Microfluidic liquid-liquid extraction system based on stopped-flow technique and liquid core waveguide capillary

In this work, a miniaturized liquid-liquid extraction system under stopped-flow manipulation mode with spectrometric detection was developed. A Teflon AF liquid-core waveguide (LCW) capillary was used to serve as both extraction channel for organic solvent flow and adsorption detection flow cell. Gravity induced hydrostatic pressure was used to drive the organic and aqueous phases through the extraction channels. During extraction process, a stable organic and aqueous phase interface was formed at the outlet of the capillary, through which the analyte in the flowing aqueous stream was extracted into the stationary organic solvent in capillary. The absorbance of the analyte extracted into the organic solvent was measured in situ by a spectrometric detection system with light emitting diode (LED) as light source and photodiode as absorbance detector. The performance of the system was demonstrated in the determination of sodium dodecyl sulfate (SDS) extracted as an ion pair with methylene blue into chloroform. The precision of the measured absorbance for a 5 mg L⁻¹ SDS standard was 6.1% R.S.D. (n = 5). A linear response range of 1-10 mg L⁻¹ SDS was obtained with 5 min extraction period. The limit of detection (LOD) for SDS based on three times standard deviation of the blank response was 0.25 mg L⁻¹.     

Chromatographic separation for domoic acid using a fragment imprinted polymer

A method for real-time visualisation of reactions performed in-capillary by the technique of electrophoretically mediated microanalysis (EMMA) is described, using a two dimensional imaging detection system. The UV absorbance detector is based on a complementary metal oxide semiconductor (CMOS) active pixel sensor. Imaging of analyte peaks absorbing at 200 nm and migrating over length of 14 mm in the capillary dimension allowed measurement of velocities and lengths of reactant and product zones. By contrast with use of single point detection, velocities of species generated by reaction anywhere within the capillary are readily measured with CMOS imaging: this is of particular benefit for EMMA experiments where reaction occurs during zone overlap. For the oxidation of glutathione by hydrogen peroxide, reaction times were varied over the range 0.5-20 s by changing voltages for electrokinetic injection and zone migration, and reactant and product peak areas were obtained for kinetic analysis of the reaction. The use of EMMA conditions with CMOS imaging allows the whole process of reaction, separation and quantification to be carried out in nanolitre volumes on-capillary in a single run on a time scale of less than 5 min.     

Portable flow-injection analyzer with liquid-core waveguide based fluorescence, luminescence, and long path length absorbance detector

We describe a multifunctional flow analysis instrument that is portable ( cm, 2.3 kg) for facile field deployment. Using a 50 cm long Teflon^® AF tubing as final reaction and optical measurement conduit, we combine a liquid-core waveguide (LCW) based fluorescence detector that is transversely illuminated by an addressable light emitting diode array, a chemiluminescence (CL) detector and an absorbance detector with a solid-state broadband (400-700 nm) source. Several illustrative experiments have been carried out to test the performance of the instrument in different detection modes. A S/N=3 limit of detection (LOD) of 0.25 μg l⁻¹ for chromium(VI) was established using the diphenylcarbazide chemistry, and an LOD of 5 μg l⁻¹ was similarly established for Al(III), using Pyrocatechol Violet (PCV) as the chelating chromogenic dye, in both cases using long path absorption detection. The LOD for aqueous hydrogen peroxide was 16 nM using a fluorescence method based on the formation of thiochrome from thiamine and 4 nM by a luminol chemiluminescence method. With a Nafion membrane diffusion scrubber (DS), the LOD was 8.0 pptv for gaseous hydrogen peroxide by the fluorescence method.     

Ultraviolet absorbance detection of colchicine and related alkaloids on a capillary electrophoresis microchip

The separation and UV absorbance detection of four toxic alkaloids, colchicine, thiocolchicine, colchicoside, and thiocolchicoside, on a microchip-based capillary electrophoresis device are reported. To increase the sensitivity of UV absorbance detection, optical cells with extended path lengths were integrated into the separation channel during the microfabrication process. The absorbance values realized on the microchip using these optical cells were proportional to the increase in average depths according to the Beer-Lambert Law, resulting in sensitivity enhancements by as much as five times. Linearity of response was observed from 5.0 to 500 mg L⁻¹ of colchicine, with detection limits ranging from 2 to 6 mg L⁻¹ depending upon the specific alkaloid and the dimension of the optical cell. The extraction of colchicine from spiked milk samples was performed and an average recovery rate of 83% with a relative standard deviation of 3.8% was determined using the optimized conditions on the microchip.     

Practical method for evaluation of linearity and effective pathlength of on-capillary photometric detectors in capillary electrophoresis

The optical characteristics of on-capillary photometric detectors for capillary electrophoresis were evaluated and five commercial detectors were compared. Plots of sensitivity (absorbance/concentration) versus absorbance obtained with a suitable testing solution yield both the linear range and the effective path length of the detector. The detector linearity is a crucial parameter when using absorbing electrolytes, such as for indirect photometric detection, and especially for highly absorbing electrolyte probe ions. The upper limits of the linear ranges (determined as 5% decline in sensitivity) for five commercial detectors ranged from 0.175 to 1.2 AU. The effective pathlength reflects the quality of the optical design of the detector and is equal to the capillary internal diameter only for a light beam passing exactly through the capillary centre, but becomes progressively shorter for imperfect optical designs. The determined effective pathlength for the five investigated detectors ranged from 49.7 to 64.6 microm for a 75 microm I.D. capillary.     

Coupling of hydrodynamically closed large bore capillary isotachophoresis with electrospray mass spectrometry

Capillary isotachophoresis with coupled columns provides efficient means for rapid electrophoretic analysis of sample volumes of up to 10 μl or more. Commercially available instruments are commonly equipped with conductivity and UV absorbance detectors; however, their on-line coupling with electrospray mass spectrometry is highly desirable. In this work we have modified the commercial coupled column isotachophoresis system for direct connection to an ion trap mass spectrometer. The design included attachment of an elution block with a short capillary transfer line directing the separated zones towards the mass spectrometer. The modification further included separation of the injection and electrode blocks from the separation columns by semipermeable membranes eliminating unwanted fluid movements in the wide bore fluoropolymer separation capillaries. Fused silica capillaries with varying internal diameter were connected as a transfer line between the elution block and mass spectrometer. The transfer line served also as the ESI tip of the sheathless electrospray interface. During the analysis the first, wide bore preseparation capillary with 0.8 mm internal diameter served for removal of the bulk sample components and preseparation of the potentially interfering analytes. After the electronic column switching the separation was finished in a 0.3 mm internal diameter capillary and the separated ITP zones were transferred in-line by a spray liquid towards the mass spectrometer. The instrumentation was tested for determination of vitamins in whole blood analysis and separation of tryptic peptides.     

On-line wall-free cell for laser-induced fluorescence detection in capillary electrophoresis

A wall-free detection method based on liquid junction in a capillary gap was proposed for laser-induced fluorescence (LIF) of capillary electrophoresis (CE). The capillary gap of the wall-free cell was fabricated by etching a 10-mm × 50-μm I.D. fused-silica capillary to obtain a polyimide coating sleeve, decoating about 6 mm at one end of both 50 μm I.D. separation and liquid junction capillary, inserting the treated capillary ends into the coating sleeve oppositely, fixing the capillaries with a gap distance of 140 μm by epoxy glue and removing the coating sleeve by burning. The theoretical model, experimental results and wall-free cell images indicated that the gap distance and applied voltage were main influence factors on the wall-free detection. Since the wall-free cell increased the absorption light path and avoided the stray light from the capillary wall, it improved the ratio of signal to noise and limit of detection (LOD) of CE-LIF. Three flavin compounds of riboflavin (RF), flavin mononucleotide sodium (FMN) and flavin adenine dinucleotide disodium (FAD) were used to evaluate the wall-free detection method. Compared with on-column cell, the LODs of the wall-free cell were improved 15-, 6- and 9-fold for RF, FMN and FAD, respectively. The linear calibration concentrations of the flavins ranged from 0.005 to 5.0 μmol/L. The column efficiency was in the range from 1.0 × 10⁵ to 2.5 × 10⁵ plates. The wall-free detection of CE-LIF was applied to the analysis of the flavins in spinach and lettuce leaves.     

Miniaturization of separation techniques using planar chip technology

Miniaturization of separation columns implies equally reduced vol- umes of injectors, detectors, and the connecting channels. Planar chip technology provides a powerful means for the fabrication of micron-sized structures such as channels. This is demonstrated by two examples. An optical absorbance detector chip exhibits the expected behavior of a 1 mm optical path length cell despite its volume of 1 nL. A capillary electrophoresis device allows integrated injections of 100 pL samples, efficiencies of 70,000 to 160,000 theoretical plates in 10 to 20 seconds, and external laser-induced fluorescence detection at any capillary length of choice between 5 and 50 mm.     

Characterization of etched electrochemical detection for electrophoresis in micron inner diameter capillaries

An enhanced etched electrochemical (EC) detection technique has been developed for CE in micron inner diameter capillaries. The design improvements allow for better alignment between the capillary bore and the electrode. This new method involves utilizing a carbon fiber microelectrode and etching both the carbon fiber and the detection end of a micrometer-sized inner diameter capillary to limit dead volume and analyte diffusion at the amperometric EC detector. To understand the factors affecting enhanced detector efficiency, a detailed examination of the relationship between detector design and performance has been completed by exploring the effects of varying electrode diameter, tip shape, and size, in addition to the etch length of the capillary outlet. The enhanced detection provides peak efficiencies as high as 75000 theoretical plates and estimated detection limits as low as 40 nM for dopamine. This etched detection method should further facilitate volume-limited sample analysis by CE.     

A multireflection cell for enhanced absorbance detection in microchip-based capillary electrophoresis devices

The design, fabrication and testing of a photolithographically fabricated, glass-based multireflection absorbance cell for microfluidic devices, in particular microchip-based capillary electrophoresis (CE) systems is described. A multireflection cell was fabricated lithographically using a three-mask process to pattern aluminum mirrors above and below a flow channel in a chip, with 30 microm diameter optical entrance or exit apertures (one in each mirror) positioned 200 microm apart. Source and detector were positioned on opposite sides, and the metal mirrors were made 1 cm square, to reduce stray light effects. Calibration curves using bromothymol blue (BTB) with a 633 nm source (He:Ne laser) were linear to at least 0.5 absorbance units, with typical r2 values of 0.9997, relative standard deviations in the slopes of +/- 1.3%, and intercepts of zero within experimental error. Effective optical pathlengths of 50-272 microm were achieved, compared to single-pass pathlengths of 10-30 microm, corresponding to sensitivity enhancements (i.e., optical path length increase) of 5 to 10-fold over single-pass devices. Baseline absorbance noise varied within a factor of two in almost all devices, depending only weakly on path length. This device can give much higher absorbance sensitivity, and should be much easier to manufacture than planar, glass-based devices previously reported.     

Lamp-based native fluorescence detection of proteins in capillary electrophoresis

The potential of a recently developed lamp-based fluorescence detector for the analysis of underivatised proteins by capillary electrophoresis (CE) was investigated. Fluorescence detection (Flu) was achieved using optical light guides to deliver excitation light from a Xenon–Mercury lamp to the capillary detection window and to collect fluorescence emission and lead it to a photomultiplier. The performance of the detector was evaluated by monitoring the native fluorescence of the amino acid tryptophan and the proteins α-chymotrypsinogen A, carbonic anhydrase II, lysozyme and trypsinogen upon excitation at 280 nm. The test compounds were analysed using background electrolytes (BGEs) of sodium phosphate at pH 3.0 and 11.3. The results were compared to experiments of CE with UV absorbance detection. For tryptophan, a linear fluorescence response was obtained with a dynamic range of over 4 orders of magnitude, and a limit of detection (LOD) of 6.7 nM. This LOD was a factor of 200 more favourable than UV detection at 280 nm, and a factor of 20 better than detection at low-UV wavelengths. All tested proteins showed linear fluorescence responses up to 250 μg/mL. LODs were typically in the 10–20 nM range. These LODs were a factor of 25 lower than for UV detection at 280 nm, and comparable to UV detection at low-UV wavelengths. Overall, Flu yields much more stable baselines, especially with a BGE of high pH. The applicability of CE–Flu is demonstrated by the analysis of a degraded protein mixture, and of an expired formulation of the protein drug human growth hormone, indicating that protein degradation products can be selectively detected.     

Understanding and improving direct UV detection of monosaccharides and disaccharides in free solution capillary electrophoresis

Direct UV detection of carbohydrates in free solution capillary electrophoresis at 270 nm is made possible by a photo-oxidation reaction. Glucose, rhamnose and xylose were shown to have unique UV absorption spectra hypothesizing different UV absorbing intermediates for their respective photo-oxidation. NMR spectroscopy of the photo-oxidation end products proved they consisted of carboxylates and not malondialdehyde as previously theorized and that oxygen thus plays a key role in the photo-oxidation pathway. Adding the photo-initiator Irgacure^® 2959 in the background electrolyte increased sensitivity by 40% at an optimum concentration of 1 × 10⁻⁴ mM and 1 × 10⁻⁸ mM for conventional 50 μm i.d. capillaries and for the corresponding extended light path capillaries, respectively.     

Preparation and characterization of p-tert-butylcalix[8]arene bonded capillaries for open-tubular capillary electrochromatography

p-tert-Butylcalix[8]arene bonded capillaries for open-tubular capillary electrochromatography were prepared with γ-glycidoxypropyltrimethoxysilane as a bridge. The bonded capillary displayed low and steady electroosmotic flow (EOF) values over the pH range from 4 to 9. Detection limits for direct spectrophotometric detection at 277 nm for benzenediols (at a signal to noise ratio of 2) were 0.96 mg l⁻¹ for the unbonded capillary and 1.48 mg l⁻¹ for the bonded capillary, showing that the bonded layer did not show significant absorbance and hence decreased sensitivity. The bonded capillaries showed good separation selectivity for o-, m- and p-benzenediols, α- and β-naphthols, and α- and β-naphthylamines. This selectivity was attributed to significant interactions between the analytes and the bonded p-tert-butylcalix[8]arene, which contributed to the electrochromatographic separation mechanism. The bonded capillaries gave high stability and reproducibility.     

A versatile total internal reflection photometric detection cell for flow analysis

A total internal reflection (TIR) flow-through cell that is highly tolerant of schlieren effects, has limited hydrodynamic dispersion and does not trap gas bubbles, and which is suitable for sensitive photometric measurements in flow analysis, is described. Light from an optical fibre is introduced into a short length of quartz capillary through the sidewall at an incident angle of ca. 53°. Under this condition, incident light undergoes total internal reflection from the external air-quartz interface and is propagated by successive reflections from the external walls through the aqueous liquid core of the cell. Detection of the transmitted beam is enabled by intentionally introducing an optical coupling medium at a predetermined distance along the capillary wall, which allows the internally reflected light to be captured by a second optical fibre connected to a charge-couple device detector.This configuration embodies a number of the desirable features of a liquid core waveguide cell (i.e. total internal reflection), a multi-reflection (MR) flow cell (i.e. minimum susceptibility to schlieren effects, low hydrodynamic dispersion and little tendency to trap bubbles), and a conventional Z-cell (wide dynamic range). When employed with a flow injection system, a limit of detection of 2.0 μg PL⁻¹ was achieved for the determination of reactive phosphate using the TIR cell, compared with LOD values of 3.8 μg PL⁻¹ and 4.9 μg PL⁻¹ obtained using the MR and Z-cells with same manifold.The combined advantages of schlieren-tolerance and lack of bubble entrapment of the MR cell with the higher S/N ratio and wider dynamic range of a conventional Z-cell, make the TIR cell eminently useful for photometric measurements of samples with widely differing refractive indices.     

Frequency-selective absorbance detection: Refractive index and turbidity compensation with dual-wavelength measurement

A frequency-selective absorbance detection approach and its applications are described. First, a digital signal processor-lock-in amplifier (DSP-LIA)-based absorbance detector was evaluated. Compared to a simple operational amplifier (TL082CP)-based detector, the DSP-LIA-based detector showed lower noise levels, but the relative advantage was reduced under very low photocurrent levels (down to few nA). A 7 cm pathlength flow cell with this commercial LIA-based detector exhibited excellent Beer's law linearity (r² = 0.9999) and a noise level of 7 micro absorbance units (μAU). The limit of detection (LOD, S/N = 3) for methyl orange (MO) was 7 nM with this detector. Finally, as a more affordable alternative to an LIA, a balanced demodulator integrated circuit chip was used to fabricate a dual wavelength-frequency-selective LED-based absorbance detector. This device successfully compensated refractive index (RI) effect and turbidity effect in test flow systems. The LOD for MO with this system was 8 nM.     

A free-flowing soap film combined with cavity ring-down spectroscopy as a detection system for liquid chromatography

We have shown that a free-flowing soap film has sufficiently high-quality optical properties to allow it to be used in the cavity of a ring-down spectrometer (CRDS). The flow rates required to maintain a stable soap film were similar to those used in liquid chromatography and thus allowed interfacing with an HPLC system for use as an optical detector. We have investigated the properties of the system in a relevant analytical application. The soap film/CRDS combination was used at 355 nm as a detector for the separation of a mixture of nitroarenes. These compounds play a role in the residue analysis of areas contaminated with explosives and their decomposition products. In spite of the short absorption path length (9 μm) obtained by the soap film, the high-sensitivity of CRDS allowed a limit of detection of 4 × 10⁻⁶ in absorption units (AU) or less than 17 fmol in the detection volume to be achieved.     

A remote flow cell for UV absorbance detection with capillary HPLC based on a single strand fiber optic

A remote flow cell based on a single strand of fused-silica fiber optic was built for UV absorbance detection with a packed capillary HPLC system, using commercially available pumps, detection electronics (Shimadzu) and fittings. This 'off-column' flow cell design is applicable to both pressure and electro-osmotically driven systems. The goals were to minimize the linearity and light leakage problems that often limit the performance of UV absorbance detection with capillary chromatography. A linear dynamic range of 10(3) (reserpine, lambda = 220 nm), and a concentration detection limit of 5.1 x 10(-8) mol l-1 were observed. Baseline noise was measured at 3.5 x 10(-5) absorbance units (AU), with a standard deviation of 1.7 x 10(-5) AU. The illuminated volume of approximately 3 nl was optimized for capillaries with inner diameters in the range 50-100 microns, and flow rates from 100 nl min-1 to 1 microliter min-1. These modifications of readily available instrumentation have allowed the construction of a practical system for fractionating complex mixtures of peptides in small amounts, prior to mass spectrometry or additional wet chemistry steps.     

Miniaturized capillary electrophoresis system with ultraviolet photometric detection combined with flow injection sample introduction using a modified falling-drop interface

A miniaturized capillary electrophoresis (CE) system with UV-Vis detection was coupled to a flow injection (FI) system for achieving high throughput continuous sample introduction. The cassette of a commercial CE instrument was modified to hold a 6.5 cm long silica capillary and a flow-through waste reservoir. The cassette was inserted into the flow-cell chamber of a commercial UV detector, with the light beam focused on the capillary and collected by two ball lenses on the cassette. The capillary inlet, left outside the cassette and detector, was positioned on the top of a vertical 3.5 mm diameter glass rod, in close contact with an electrode. Samples injected through the FI system dropped freely on top of the pillar, covering the capillary inlet and electrode. Continuous sample introduction was achieved for CE separations under non-interrupted separation voltage, which was isolated from the FI system through the discontinuity of droplets. The newly developed interface and UV detection system was used for fast separation of sulphamethoxazole (SMZ) and trimethoprim (TMP) in sulphatrim tablets, achieving a high throughput of over 48 h⁻¹, and a low carryover of 2%. Separation efficiencies of 8 μm plate height and detection limits of 1.0 mg l⁻¹ for SMZ and 0.5 mg l⁻¹ (3σ) for TMP were obtained.     

Significant improvement of signal-to-noise ratio in capillary electrophoresis through optimization of aperture width for UV absorption detection.

In capillary electrophoresis (CE), light flux passes through a capillary cell and is in most cases detected photometrically. Due to the thinness of the cell, a part of the light passes through the wall and misses hitting the sample. In most CE apparatuses, incident light is focused by converging lenses in order to condense light beams passing through the capillary. Considering the aberration of lenses and lens effects of capillary, we assumed that light beams inside were approximately parallel. Although the path lengths of light beams vary depending on their tracks, we could estimate the virtual light path length, L, by measuring absorbance when concentration and molar absorptivity of the sample solution were known. A light-restricting device consisting of narrow slits makes effectively L longer and signal intensity higher. On the other hand, noise increases as light width narrows. The signal-to-noise ratio showed a maximum at 68 microm of light width for a capillary with diameter of 75 microm. The optimized L was evaluated by the simulation. The experimental data verified it even in indirect UV detection. Our approach could help to design the optics of CE apparatuses.     

Improving detection in capillary electrophoresis with laser induced fluorescence via a bubble cell capillary and laser power adjustment

Bubble cells have been frequently employed in capillary electrophoresis (CE) to increase the light path length with UV detection to provide an increase in the observed sensitivity of CE; however this approach has not been commonly used for laser-induced fluorescence detection (LIF) with CE. In this paper we study the influence of laser power on the sensitivity of detection in using conventional and enlarged fused silica capillaries for CE with LIF. When using the bubble cell capillary, the laser power must be decreased relative to use of the conventional capillary to reduce the effects of photodegradation of the species being illuminated by the laser. Even though the light intensity was decreased, an increase in sensitivity of detection was observed for most compounds when a bubble cell was used. This increase ranged from a factor of 8 for riboflavin (410 nm excitation) to 3.2 for most aromatic compounds (266 nm excitation), when using a 3x bubble cell compared with a conventional capillary. The bubble cell capillary was used for native detection of IgG by LIF at 266 nm. A limit of detection of 60 ng mL(-1) was obtained from a 20 pg injection, which was 40 times more sensitive than silver staining in conventional SDS/PAGE.