A novel analytical low-cost flow system based on a 0.6 MPa (84 psi) diaphragm pump applied to on-line trace pre-concentration in flame AAS and ICP-OES

A novel inexpensive 0.6 MPa (84 psi) flow system based on a low-cost diaphragm pump has been developed. The unfavourable strong pulsation of the pump has been overcome by using highly flexible silicone tubing as a pulse suppression coil. This results in a smooth pulse-free continuous flow of 100 mL min(-1) in circulation. This flow rate is much too high for a flow-injection system; however, with a restrictor capillary the flow rate required can be tapped off down to a range of 0.1-50 mL min(-1). By employing diaphragm pumps in an analytical flow system the pressure gap between HPLC pumps (2-40 MPa) and peristaltic pumps (<0.2 MPa), mainly used in FIA systems, can be filled. Due to the higher pressure delivered by diaphragm pumps relative to peristaltic pumps, the new flow system can be applied to on-line sample pre-concentration and matrix separation in flame AAS and ICP-OES by using standard HPLC pre-columns or small ion-exchange columns, respectively. In this way, very low detection limits in flame AAS have been reached (Cd 0.07 micro g L(-1), Cu 0.05 micro g L(-1), Co 0.9 micro g L(-1), Ni 0.8 micro g L(-1), Mn 0.7 micro g L(-1), Pb 0.8 micro g L(-1) and Tl 0.2 micro g L(-1)).     

Flow inconsistency: The evil twin of column switching—Hardware aspects

Solvent flow, generated by HPLC pumps is consistent and accurate. This statement, while true for single column (one dimensional) liquid chromatography applications, may not apply to column switching applications. Connection of pumps and/or columns to one flow path may cause substantial pressure changes. Immediate post valve switch pressure differences between pumps can cause backflow where the mobile phase stored at higher pressure will temporary flow into the lower pressure area. A more common side effect of column switching is flow inconsistency during pump pressurization. For the duration of pump pressurization, liquid flow through the column will be smaller than expected since the HPLC column acts like a flow restrictor.     

Means of validation of HPLC systems and components

Summary The performance of HPLC pumps is determined by applying Hagen-Poiseuille's law. As restrictor a fused silica column is used. The pulsation height and frequency is measured via a pressure transducer. Fourier transformation of pulsation data permits the characterization of pump identity. The absolute calibration of HPLC detectors and the determination of their linear and dynamic range as well as the noise level can be achieved by a single measurement applying the exponential dilution flask. Wavelength accuracy of variable wavelength detectors can be determined by comparing spectral data of aromatic hydrocarbons from the literacture with those determined with the detectors. Presented at the 21^st ISC held in Stuttgart, Germany, 15^th–20^th September, 1996     

Characterization of Tritium Exposures by Measuring Tritiated Metabolites in Urine

A high-performance liquid chromatography (HPLC)-based method was developed to look for the presence of characteristic urinary metabolites associated with different tritium-exposure situations. Non-volatile metabolites in urine were isolated by evaporating an aliquot of urine samples, at room temperature under nitrogen, from animals percutaneously exposed to tritiated thymidine, tritiated formaldehyde, tritium-gas-contaminated metal surfaces and tritiated pump oil. A total of 40 fractions were collected at 1 min intervals with a flow rate of 1 ml·min^-1, and their tritium activities were measured. The activity profile of tritium showed that the ratios of non-volatile tritiated metabolites in fraction I (0–20 min) to fraction II (20–40 min) were noticeably different among the animals exposed to tritiated thymidine (77.2±4.5), tritiated formaldehyde (40.9±3.3), tritium-gas-contaminated metal surfaces (16.5±2.5), or tritiated pump oil (8.7±0.4) 24 h post-exposure. Our results suggest that, if the nature of a tritium exposure is unknown, comparison of the ratio of fraction I to fraction II in non-volatile tritiated metabolites may be useful in characterizing the source and the nature of tritium exposure.     

Membrane emulsification using membranes of regular pore spacing: Droplet size and uniformity in the presence of surface shear

During membrane emulsification it is shown that the size of the drops formed at the membrane surface may increase with increasing dispersed phase injection rate through the membrane, or it may decrease, depending on the prevailing conditions. This is illustrated using a paddle stirrer positioned above flat disc membranes with regular arrays of pores of 20 μm diameter and spacing between the pores of 80 and 200 μm. In the former case an additional mechanism for drop detachment is the push-off force, which is determined by the geometry of the drops as they deform at the membrane surface. When dispersing sunflower oil in to aqueous solutions containing Tween 20, drop sizes between 60 and 200 μm were produced, and in the case of the membrane when the push-off force was working the Coefficient of Variation of the drops formed was below 10%. The push-off force may be added to the shear-drag force to predict drop detachment. For the 200 μm pore spaced membrane this force is much less prominent than the 80 μm spaced membrane. The capillary-shear model has been modified to include this push-off force. The experimental study required the use of very low dispersed phase injection rates as well as very high rates. Hence, two different types of pumps were used to provide these: a peristaltic and syringe pumps. A small study comparing the drop size, and size distributions, showed that the pump type did not influence the drops produced by the membrane emulsification process.     

An integrated micropump and electrospray emitter system based on porous silica monoliths

The work presents the design of an integrated system consisting of a high-pressure electroosmotic (EO) micropump and a microporous monolithic emitter, which together generate a stable and robust electrospray. Both the micropump and electrospray emitter are fabricated using a sol-gel process. Upon application of an electric potential of sufficient amplitude (>2 kV), the pump delivers fluids with an electroosmotically induced high pressure (>1 atm). The same potential is also harnessed to electrostatically generate a stable electrospray at the porous emitter. Electrokinetic coupling between pump and spray produces spray features different from sprays pressurized by independent mechanical pumps. Four typical spray modes, each with different drop sizes and charge-to-mass ratios, are observed and have been characterized. Since the monolith is silica-based, this integrated device can be used for a variety of fluids, especially organic solvents, without the swelling and shrinking problems that are commonly encountered for polymer monoliths. The maximum pressure generated by a 100 microm id monolithic pump is 3 atm at an applied voltage of 5 kV. The flow rate can be adjusted in the range of 100 nL/min to 1 microL/min by changing the voltage. For a given applied voltage across the pump and emitter system, it is seen that there exists one unique flow rate for which flow balance is achieved between the delivery of liquid to the emitter by the pump and the liquid ejection from the emitter. Under such a condition, a stable Taylor cone is obtained. The principles that lead to these results are also discussed.     

New approach to large-volume injection in reversed-phase high performance liquid chromatography: determination of atrazine and hydroxyatrazine in soil samples

A well established method of direct injection of larger than conventional sample volumes ranging from 0.1 mL to 10 mL in HPLC is the injection valve method in which a loop of tubing is totally or partially filled with sample. Recent HPLC pumps have a flow-rate setting accuracy of +/- 1-2% over a flow-rate range from 0.1 mL/min to 10 mL/min and the flow stability is 0.2% or less. Quarternary low pressure gradient pumps are widely available and used, but all their hydraulic lines are seldom utilised. The idea of using one line of a common commercial HPLC quaternary low-pressure pump for direct on-column injection (pumping) of large sample volumes ranging from 1 mL to 100 mL was tested. This approach was evaluated during practical work on the development of an RP-HPLC method of determination of residual atrazine and hydroxyatrazine. In lysimetric environmental experiments hydroxyatrazine was formed in situ in a soil column by hydrolysis of atrazine. The results proved the applicability of this approach not only in experiments with model mixtures of analytes at microg/L levels in solutions. Analysis of 20 mL of soil leachates and extracts of soil samples containing atrazine and hydroxyatrazine at the 10 microg/kg level (in dry soil) revealed that good figures-of-merit were preserved, even in the presence of a large excess of humic substances.     

Capillary column SFC instrumentation based on piston pumps with a programmable back-pressure regulator for on-line mixing of mobile phases

An alternative phase delivery system based on piston pumps and a back-pressure regulator has been developed for capillary column SFC. The chromatography is not affected by the fast piston pump refill. A homogeneous on-line mixing of binary phases with simultaneous pressure programming is easily accessible without any additional computing. Acceptable reproducibilities (< 3.5% RSD for external and < 2.0% RSD for internal standard methods) were found with mixtures of 2-propanol/CO₂ as mobile phases using UV detection and split ratios of 1:60 and 1:120. Variation and control of the split are easily done by simple flow rate volumetric changes.     

High dead volume pumps: Delay time reduction by short-term low-ratio split flow in LC/MS applications

The aim of our work was to develop a low-cost, simple and reliable solution to reduce LC/MS analysis time by compensating for limitations inherent to high dead volume standard HPLC pumps. In our approach, we utilized a temporary (sub-1 min) low ratio flow split (approximately 1:10) at 5 mL/min pump flow before the column. During this short period, 90% of the entire pump flow is delivered to waste and used for fast pump/system equilibration. Although full-time flow splitting is widely used in capillary/nano applications (usually with high split ratios in the hundreds or thousands), to our knowledge, this is the first time that short-term low-ratio flow splitting has been used in conventional LC/MS applications.     

Porous glass electroosmotic pumps: design and experiments

An analytical model for electroosmotic flow rate, total pump current, and thermodynamic efficiency reported in a previous paper has been applied as a design guideline to fabricate porous-structure EO pumps. We have fabricated sintered-glass EO pumps that provide maximum flow rates and pressure capacities of 33 ml/min and 1.3 atm, respectively, at applied potential 100 V. These pumps are designed to be integrated with two-phase microchannel heat exchangers with load capacities of order 100 W and greater. Experiments were conducted with pumps of various geometries and using a relevant, practical range of working electrolyte ionic concentration. Characterization of the pumping performance are discussed in the terms of porosity, tortuosity, pore size, and the dependence of zeta potential on bulk ion density of the working solution. The effects of pressure and flow rate on pump current and thermodynamic efficiency are analyzed and compared to the model prediction. In particular, we explore the important tradeoff between increasing flow rate capacity and obtaining adequate thermodynamic efficiency. This research aims to demonstrate the performance of EOF pump systems and to investigate optimal and practical pump designs. We also present a gas recombination device that makes possible the implementation of this pumping technology into a closed-flow loop where electrolytic gases are converted into water and reclaimed by the system.     

An electrochemical detector with a dropping mercury electrode for high-performance liquid chromatography

The electrochemical flow-through cell described has an active volume of less than 1 μl, and incorporates a dropping mercury electrode with drop times of about 0.05 s. Its performance as a detector for high-performance liquid chromatography is assessed for p-nitrophenol and nitrobenzene. The detection limits are 4–5 ng; the relative standard deviation of the peak height is better than 5% for the range 5–150 ng. The dependence of the response on flow rate, mercury pressure and drop time is described.     

Wall modified photonic crystal fibre capillaries as porous layer open tubular columns for in-capillary micro-extraction and capillary chromatography

Wall modified photonic crystal fibre capillary columns for in-capillary micro-extraction and liquid chromatographic separations is presented. Columns contained 126 internal parallel 4 μm channels, each containing a wall bonded porous monolithic type polystyrene-divinylbenzene layer in open tubular column format (PLOT). Modification longitudinal homogeneity was monitored using scanning contactless conductivity detection and scanning electron microscopy. The multichannel open tubular capillary column showed channel diameter and polymer layer consistency of 4.2 ± 0.1 μm and 0.26 ± 0.02 μm respectively, and modification of 100% of the parallel channels with the monolithic polymer. The modified multi-channel capillaries were applied to the in-capillary micro-extraction of water samples. 500 μL of water samples containing single μg L⁻¹ levels of polyaromatic hydrocarbons were extracted at a flow rate of 10 μL min⁻¹, and eluted in 50 μL of acetonitrile for analysis using HPLC with fluorescence detection. HPLC LODs were 0.08, 0.02 and 0.05 μg L⁻¹ for acenaphthene, anthracene and pyrene, respectively, with extraction recoveries of between 77 and 103%. The modified capillaries were also investigated briefly for direct application to liquid chromatographic separations, with the retention and elution of a standard protein (cytochrome c) under isocratic conditions demonstrated, proving chromatographic potential of the new column format, with run-to-run retention time reproducibility of below 1%.     

Evaluation of an automated on-line method for generating calibration curves in inductively coupled plasma atomic emission spectrometry

Summary A high-performance liquid chromatography pump (HPLC) is used for the computer-controlled on-line generation of calibration standards in inductively coupled plasma-atomic emission spectrometry. Such a pump, employed as a solution-delivery device, is well suited for feeding a nebulizer, since it is capable of delivering a range of constant flows and of mixing selected stock solutions or solvents automatically; this capability translates into the capacity to dilute stock or sample solutions and to introduce the resulting mixtures at constant flow into any of a number of nebulizers. Further, the range of useful nebulizers is broadened, since the mixture is supplied under high pressure. This feature makes the method especially useful for devices such as the jet-impact nebulizer (JIN), which requires a pressurized, constant feed. At present, the principal limitation of the new approach is the flow resolution of available HPLC pumps. Although this limitation does not ordinarily cause difficulties in schemes for automatic dilution or calibration, it can be a serious shortcoming if the apparatus is to be used for automated standard-additions analysis.Dedicated to Professor Dr. Wilhelm Fresenius on the occasion of his 80th birthday     

INFLUENCE OF TURBULENCE AND VISCOSITIES ON THE KINETICS OF DROP BREAKING

The proposed model assumes breaking to be caused by instant pulsation with sufficiently high amplitude. The critical amplitude is estimated taking Into account capillary pressure Inside the drop and viscous shear tension on Its surface. The kinetics of drop size decrease Is described using the distribution of pulsations and statistics of residence time of emulsion In the most turbullzed area of flow.

The model was verified In vessels of four sizes with 17 agitators; relation of dispersed-to-continuous phase viscosities varied by approximately 600 times, fleximal turbulent dissipation rate values varied In the range of 10-10,000 W/kg.

The Influence of viscosities was found to Increase with an Increase 1n turbulence Intensity. The minimal mean drop diameter is obtained when emulsifying liquids with a definite viscosities ratio. It Increases even If this ratio Is changed as a result of a decrease In one of the viscosities.

The model allows to predict drop size distribution and to optimize emulsifying devices.     

LC Determination of 4-Bromoaniline in Green Algae Chlamydomonas reinhardtii After Continuous-Flow Microextraction

In this study, the determination of 4-Bromoaniline (4-BA) in green algae Chlamydomonas reinhardtii (C. reinhardtii) was investigated by applying continuous-flow microextraction (CFME) combined with high-performance liquid chromatography (HPLC). Continuous-flow microextraction was conducted in a homemade glass chamber, i.e. the sample solution flowed through a constant volume drop of solvent in the chamber at a constant flow rate. The effects of different factors on extraction efficiencies were also investigated and these factors included the kind of extraction solvent, solvent drop volume, sample flow rate, extraction time and addition amount of salt. Under the optimum extraction conditions (extraction solvent, carbon tetrachloride; solvent drop volume, 3.5 μL; sample flow rate, 1.0 mL min⁻¹; extraction time, 10 min; no addition of salt), the calibration plot was set up by plotting peak area against a series of 4-Bromoaniline concentrations (0.01–10 μg mL⁻¹) in aqueous solution. The correlation coefficient (r) was 0.9990. The limit of detection (LOD) was 0.6 ng mL⁻¹. The precision of this method was obtained by successive five time analyses of 100-ng mL⁻¹ standard solution of 4-Bromoaniline, and the relative standard deviation (RSD) was 3.5%. The concentration factor was calculated by the ratio of peak area of the analyte obtained after and before extraction and found to be 10.6. 4-Bromoaniline residues in Chlamydomonas. reinhardtii cells and tap water samples were satisfactorily analyzed according to the method described above.     

Low flow high-performance liquid chromatography solvent delivery system designed for tandem capillary liquid chromatography-mass spectrometry

A solvent delivery system is described that is designed to increase the efficiency of liquid chromatography-mass spectrometry (LC/MS) analyses. Gradients formed by using two low pressure syringe pumps are stored in a length of narrow bore tubing (gradient loop) mounted on a standard high pressure switching valve. The preformed gradient is pushed through the column by using a high pressure syringe pump. The system is fully automated and can be controlled with either a personal computer or the mass spectrometer data system. Advantages include gradient operation without the use of split flows, pressure programed flow control for rapid sample loading and recycling to initial conditions, and a flow rate range of 0.1–20 μL/min, which is suitable for packed capillary columns 50–500 μm in diameter. The system has been used extensively for rapid molecular weight determinations of intact protein samples, as well as LC/MS and liquid chromatography-tandem mass spectrometry analyses of complex peptide mixtures.     

A New Monolithic‐Type HPLC Column For Fast Separations

The application of a new silica‐based, monolithic‐type HPLC‐column for fast separations is presented. The column is prepared according to a new sol‐gel process, which is based on the hydrolysis and polycondensation of alkoxysilanes in the presence of water soluble polymers. The method leads to “rods” made of a single piece of porous silica with a defined pore structure, i. e. macro‐ and mesopores. The main feature of silica rod columns is a higher total porosity, about 15% higher than of conventional particulate HPLC columns. The resulting column pressure drop is therefore much lower, allowing operation at higher flow rates including flow gradients. Consequently, HPLC analysis can be performed much faster, as it is demonstrated by various applications.     

Effects of the thermal heterogeneity of the column on chromatographic results

The influence of the thermal heterogeneity of HPLC columns on retention data was investigated. The retention factor of the retained compound phenol was measured at 24 increasing values of the flow rate, from 0.025 to 4.9 mL/min, on six different packing materials prepared with the same batch of silica particles (5 microm diameter, 90 A pore size). One column was packed with the neat silica particles, another with the silica endcapped with trimethylchlorosilane (TMS)(C(1), 3.92 micromol/m(2)), and the other four with silica first derivatized with octadecyl-dimethyl-chlorosilane (C(18), 0.42, 1.01, 2.03, and 3.15 micromol/m(2)), and second endcapped with TMS. Four different sources of heat contributing to raise the column temperature were considered: (1) the heat supplied by the hot high-pressure pump chamber to the solvent; (2) the adiabatic (dS=0) compression of the solvent in the high-pressure pump; (3) the isenthalpic (dH=0) decompression of the solvent during its migration along the porous chromatographic bed; and (4) the heat released by the friction of the solvent percolating through the column bed. The main contributions appear to be the heat supplied to the solvent by the HP pump and the friction heat. The average column temperature (ACT) was indirectly derived from the measurements of the first moment, mu(1), of phenol peak, of the column pressure drop, DeltaP, and of the retention factors of the phenol peak apices as a function of the flow rate applied. If the column is placed in a still-air bath at 298 K (and its temperature is not externally controlled), a longitudinal temperature gradient is established along the column and the average column temperature is about 6 K higher when this column is operated at 4.9 mL/min than when the flow rate is only 0.025 mL/min. If the column is placed in a heated air bath (temperature controlled at 316 or 338 K), the ACT changes by less than 3 K over the whole flow rate range applied.     

Long-term stable electroosmotic pump with ion exchange membranes

We present the design, fabrication and test of a novel inline frit-based electroosmotic (EO) pump with ion exchange membranes. The pump is more stable than previous types due to a new flow component that ensures a controlled width of the diffusion layer close to the ion exchange membranes. The pump casing is constructed in polymers while the EO active part, the frit, is made in a nanoporous silica. The pressure capability of the pump is Deltapm/DeltaV = 0.15 bar V(-1). The flow rate to current ratio is Qm/I = 6 microL min(-1) mA(-1). This translates to Deltapm = 4.5 bar and Qm = 6 microL min(-1) at DeltaV = 30 V. The pump has been tested with four different buffer concentrations. In order to investigate day-to-day reproducibility each Q-p pump characteristic has been recorded several times during hour-long operation runs under realistic operating conditions.