Search

Polymeric dispersants are an important ingredient in many consumer products. Their separations and identifications in final product formulation can be very challenging due to the presence of multiple polymeric dispersants at different levels and the presence of other polymeric and small-molecule components. In this study, using nearly comprehensive two-dimensional liquid chromatography (2D-LC), various water-soluble polymer and co-polymer dispersants were separated with aqueous size exclusion chromatography (SEC) in the first dimension (1D) and gradient elution reversed-phase liquid chromatography (RPLC) in the second dimension (2D). Detection of the polymeric dispersants was accomplished by evaporative light scattering detector (ELSD). A large ID (8.0 mm) SEC column used in common one-dimensional SEC practices was directly adopted in the 2D setup for rapid method development. A close representation of fully comprehensive 2D separation was achieved even with 60% of 1D eluent diverted to waste, demonstrating the flexibility and versatility of having SEC in 1D for two dimensional separation of polymers. Important method parameters, such as 2D column dimensions and flow rate, gradient conditions, and buffer pH were studied. Practical aspects of routine industrial applications such as solvent consumption and analysis time were also considered. This method was exploited for quick identification of polymeric dispersants in commercial detergent samples. Nine detergent samples were screened and polymeric dispersants and additional polymer features were detected in the samples.

Cation exchange chromatography (CEX) is a widely used technique for the removal of monoclonal antibody (mAb) aggregates. At present, resins are mainly used for this purpose, as convective types of adsorbents such as membrane adsorbers (MAs) have often not demonstrated overall comparable performance for this particular application. Fiber-based adsorbents can overcome the current limitations of MAs with respect to permeability, binding capacity, and adsorbent cost, and could therefore be a viable alternative to resins for the removal of mAb aggregates. It has not been evaluated, however, whether and under which conditions the use of such adsorbents is feasible for this purpose. In the present study, the use of fiber-based CEX adsorbents for mAb aggregate removal was examined. Two types of fiber-based adsorbents, an uncontrolled grafted and a controlled grafted fiber-based adsorbent, were evaluated with respect to permeability, dynamic mAb binding capacity (DBC), resolution capabilities, and the performance in bind and elute (B/E) and frontal chromatography mode with respect to typical performance indicators. The permeabilities of the fiber-based adsorbents ranged from 200 to 1700 mD, making it possible to use the fiber-based adsorbents at larger bed heights than membrane adsorbers with fast mobile phase velocities. Antibody DBCs ranged from 20 to 41 g/L at 150 cm/h, and at higher mobile phase velocities exceeded the DBC of an existing resin material, Poros 50 HS, which has frequently been used for aggregate removal. Both fiber types showed good resolution capabilities of monomer and aggregates, and provided better resolution per column length than Poros 50 HS. Typical purity and yield constraints were fulfilled for both fiber types in both B/E and frontal chromatography mode for mobile phase velocities ranging up to 480 cm/h and 1060 cm/h. The overall performance of the controlled grafted fibers was found to be superior to the performance of uncontrolled grafted fiber-based adsorbents due to higher productivity and lower buffer consumption. The overall performance of the fiber-based adsorbents was found to be comparable to the performance of Poros 50 HS at typical operating conditions. The results in this study indicate that the use of fiber-based adsorbents for mAb aggregate removal is feasible with a performance that is comparable to the performance of an existing resin material. Depending on the cost of the adsorbents and the use scenario, the usage of such adsorbents could be beneficial.

Ultra-high voltage capillary electrophoresis with high electric field strength has been applied to the separation of the charge variants, drug conjugates, and disulfide isomers of monoclonal antibodies. Samples composed of many closely related species are difficult to resolve and quantify using traditional analytical instrumentation. High performance instrumentation can often save considerable time and effort otherwise spent on extensive method development. Ideally, the resolution obtained for a given CE buffer system scales with the square root of the applied voltage. Currently available commercial CE instrumentation is limited to an applied voltage of approximately 30kV and a maximum electric field strength of 1kV/cm due to design limitations. The instrumentation described here is capable of safely applying potentials of at least 120kV with electric field strengths over 2000V/cm, potentially doubling the resolution of the best conventional CE buffer/capillary systems while decreasing analysis time in some applications. Separations of these complex mixtures using this new instrumentation demonstrate the potential of ultra-high voltage CE to identify the presence of previously unresolved components and to reduce analysis time for complex mixtures of antibody variants and drug conjugates.

A facile and efficient dynamic coating method using an ionic complementary peptide was established for high-performance separation of 8-aminopyrene-1,3,6-trisulfonic acid (APTS)-labeled carbohydrates in a hybrid poly(dimethylsiloxane) (PDMS)/glass microfluidic channel. EAK16-II with a sequence of [(Ala-Glu-Ala-Glu-Ala-Lys-Ala-Lys)2] can readily self-organize into a complete coating layer tightly adsorbed on both hydrophobic PDMS and hydrophilic glass surfaces, which efficiently suppressed nonspecific analyte adsorption and minimized electroosmotic flow (EOF). Separation conditions were systematically investigated with respect to EAK16-II concentration, running buffer, buffer pH, and field strength (Esep). Under the optimal conditions, rapid and reproducible separations of maltodextrin ladder, glycans from glucosamine capsules, tablets, and pomegranate peel extracts were achieved with over 450000 theoretical plates per meter in the hybrid PDMS/glass microchannels dynamically coated with 1.0mg/mL EAK16-II-0.05% n-dodecyl β-d-maltoside (DDM), and the relative standard deviation (RSD) values were less than 3.2% (n=4) for the migration times. The present work provides a facile and efficient means to minimize EOF and nonspecific analyte adsorption in microfluidic chips fabricated in various substrates, thereby broadening the applications of microfluidic chips in complicated biological assays.

Continuous processing of recombinant proteins was accomplished by combining continuous matrix-assisted refolding and purification by tandem simulated moving bed (SMB) size-exclusion chromatography (SEC). Recombinant proteins, N(pro) fusion proteins from inclusion bodies were dissolved with NaOH and refolded in the SMB system with a closed-loop set-up with refolding buffer as the desorbent buffer and buffer recycling of the refolding buffer of the raffinate by tangential flow filtration. For further purification of the refolded proteins, a second SMB operation also based on SEC was added. The whole system could be operated isocratically with refolding buffer as the desorbent buffer, and buffer recycling could also be applied in the purification step. Thus, a significant reduction in buffer consumption was achieved. The system was evaluated with two proteins, the N(pro) fusion pep6His and N(pro) fusion MCP-1. Refolding solution, which contained residual N(pro) fusion peptide, the cleaved autoprotease N(pro), and the cleaved target peptide was used as feed solution. Full separation of the cleaved target peptide from residual proteins was achieved at a purity and recovery in the raffinate and extract, respectively, of approximately 100%. In addition, more than 99% of the refolding buffer of the raffinate was recycled. A comparison of throughput, productivity, and buffer consumption of the integrated continuous process with two batch processes demonstrated that up to 60-fold higher throughput, up to 180-fold higher productivity, and at least 28-fold lower buffer consumption can be obtained by the integrated continuous process, which compensates for the higher complexity.

The use of either a polyampholyte buffer or a simple buffer system for the high-performance cation-exchange chromatofocusing of monoclonal antibodies is demonstrated for the case where the pH gradient is produced entirely inside the column and with no external mixing of buffers. The simple buffer system used was composed of two buffering species, one which becomes adsorbed onto the column packing and one which does not adsorb, together with an adsorbed ion that does not participate in acid-base equilibrium. The method which employs the simple buffer system is capable of producing a gradual pH gradient in the neutral to acidic pH range that can be adjusted by proper selection of the starting and ending pH values for the gradient as well as the buffering species concentration, pKa, and molecular size. By using this approach, variants of representative monoclonal antibodies with isoelectric points of 7.0 or less were separated with high resolution so that the approach can serve as a complementary alternative to isoelectric focusing for characterizing a monoclonal antibody based on differences in the isoelectric points of the variants present. Because the simple buffer system used eliminates the use of polyampholytes, the method is suitable for antibody heterogeneity analysis coupled with mass spectrometry. The method can also be used at the preparative scale to collect highly purified isoelectric variants of an antibody for further study. To illustrate this, a single isoelectric point variant of a monoclonal antibody was collected and used for a stability study under forced deamidation conditions.

For reversed phase separation amino acids are usually derivatized. Several derivatization reactions are carried out at basic pH. In the present work, influence of three basic buffer solutions on liquid chromatography electrospray ionization mass-spectrometric (LC–ESI-MS) analysis of amino acid derivatives was studied. Borate buffer – the most common derivatization buffer – was found to influence ESI ionization up to 23 min retention time. For 9-fluorenylmethylmethoxycarbonyl chloride (Fmoc-Cl derivatization) carbonate buffer should be preferred as it provides higher responses. Hexafluoroisopropanol (HFIP) buffer improves chromatographic peak shapes and responses for diethyl ethoxymethylenemalonate (DEEMM) derivatives.

An ethylene glycol solution was used as the electrophoretic running buffer in unmodified cyclic olefin copolymer (COC) microchips to minimize the interactions between the analytes and the hydrophobic walls of the plastic microchannels, enhance the resolution of the analytes and eliminate the uncontrollable dispersion caused by uneven liquid levels and non-uniform surfaces of the separation channels. Five amino acids that were labeled with fluorescein isothiocyanate (FITC) were used as model analytes to examine the separation efficiency. The effects of ethylene glycol concentration. pH and sodium tetraborate concentration were systematically investigated. The five FITC-labeled amino acids were effectively resolved using a COC microchip with an effective length of 2.5 cm under optimum conditions, which included using a running buffer of 20 mmol/L sodium tetraborate in ethylene glycol:water (80:20, v/v), pH 6.7. A theoretical plate number of 4.8 x 10(5)/m was obtained for aspartic acid. The system exhibited good repeatability, and the relative standard deviations (n = 5) of the peak areas and migration times were no more than 3.4% and 0.7%, respectively. Furthermore, the system was successfully applied to elucidate these five amino acids in human saliva. (C) 2011 Elsevier B.V. All rights reserved.

Two fluoroalcohols – 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) and 1,1,1,3,3,3-hexafluoro-2-methyl-2-propanol (HFTB) – were evaluated as volatile buffer acids in basic mobile phases for LC–ESI-MS determination of acidic and basic compounds. HFIP and HFTB as acidic buffer components offer interesting possibilities to adjust retention behavior of different analytes and expand the currently rather limited range of ESI-compatible buffer systems for basic mobile phases. Comparing with commonly used basic buffer components the fluoroalcohols did not suppress the ionization of the analytes, for several analytes ionization enhancement was observed. RP chromatographic retention mechanisms were evaluated and compared to traditional buffer system. All trends in retention of the acidic and basic analytes can be interpreted by the following model: the neutral fluoroalcohols are quite strongly retained by the stationary phase whereas their anions are less retained, thus their amount on the stationary phase is dependent on mobile phase pH; the anions of the fluoroalcohols form ion pairs in the mobile phase with the basic analytes; the fluoroalcohols on the stationary phase surface compete with acidic analytes thereby hindering their retention; the fluoroalcohols on the stationary phase bind basic analytes thereby favoring their retention.

A novel anion exchange membrane adsorber is presented which shows excellent impurity removal under different buffer conductivities ranging from 2 to 27 mS/cm. The membrane utilizes a primary amine ligand (polyallylamine) and was designed specifically to bind impurities at high salt concentrations. Studies with DNA, endotoxin, and virus spiked into buffer at varying salt conditions were done, resulting in clearance of >3, 4, and 4 LRV, respectively, with negligible change on increasing salt up to 27 mS/cm conductivities. Verification of virus removal in mAb feedstocks is also shown. The data are compared with other membrane adsorbers and a conventional resin which utilize traditional chemistries to demonstrate improved purification performance with the primary amine ligand. Additional data on scale-up of the membrane adsorber device is discussed. A stacked flat-sheet design was implemented to ensure linear scale-up of performance using bovine serum albumin (BSA) as a model. The linearly scalable device, coupled with the highly effective membrane for virus, DNA, and endotoxin removal, represents a step forward in polishing technology for the purification of monoclonal antibodies and recombinant proteins.

Flow field flow fractionation (Fl-FFF) was coupled to flow cytometry to improve the performance of suspension arrays. Size-based separation of the protein-conjugated microspheres by Fl-FFF was performed and the results demonstrated that, the separation could tolerate a wide range of carrier fluid conditions (pH values, salt concentrations, and buffer compositions) favorable for immunoassays. The immuno-complex remained intact during Fl-FFF, as revealed by fluorescence measurements before and after the Fl-FFF separation, and SDS-PAGE of the eluted proteins. The sample throughput of the suspension array can be increased several folds by using particles of different sizes and separating them with Fl-FFF before flow cytometric measurement. Moreover, the gel result hinted that the continuous wash inside the Fl-FFF system may lower the assay background, another possible advantage of the two-dimensional suspension array system.

The application of quantum dots in capillary electrophoresis immunoassay was studied for the first time. Quantum dots were conjugated with antibody and subsequently tested by electrophoretic separation of free antibody and antibody–antigen complex. Antibody was fluorescently labeled by quantum dots via conjugation procedures and its electrophoretic characteristics were effectively modified due to the attachment of quantum dots. The determination of human IgM by direct CE based immunoassay could be easily achieved by simply changing the pH value of separation buffer. Polymer additive influenced the separation too but the effect was not as significant as buffer pH adjustment. Satisfactory separation of complex from free antibody could be achieved with 20 mM sodium tetraborate as separation buffer, at pH 9.8. The immunoassay application of quantum dots in CE offers considerable advantages and can be readily applied to other large bio-molecules.

In agreement with our previous studies and those of other authors, it is shown that much better fits of retention time as a function of pH are obtained for acid–base analytes when pH is measured in the mobile phase, than when pH is measured in the aqueous buffer when buffers of different nature are used. However, in some instances it may be more practical to measure the pH in the aqueous buffer before addition of the organic modifier. Thus, an open methodology is presented that allows prediction of chromatographic retention of acid–base analytes from the pH measured in the aqueous buffer. The model presented estimates the pH of the buffer and the pKa of the analyte in a particular acetonitrile/water mobile phase from the pH and pKa values in water. The retention of the analyte can be easily estimated, at a buffer pH close to the solute pKa, from these values and from the retentions of the pure acidic and basic forms of the analyte. Since in many instances, the analyte pKa values in water are not known, the methodology has been also tested by using Internet software, at reach of many chemists, which calculates analyte pKa values from chemical structure. The approach is successfully tested for some pharmaceutical drugs.

A simple on-line preconcentration method of protein for capillary electrophoresis (CE) using a cellulose acetate (CA)-coated porous membrane was proposed. CA membrane is fabricated at one of the ends of the column that allows the passage of buffer ions but excludes larger protein molecules. Protein sample is continuously electrokinetically loaded and trapped by the membrane. When injection is completed, the direction of the electric field is switched and the trapped proteins are then separated by conventional CE procedure. The results achieved showed that the preconcentration mechanism of this method was based on size-exclusion effect. Bovine serum albumin (BSA) was used for model protein sample, and signal enhancement of 550-fold with 15 min injection time was achieved.

An organic buffer system will be discussed that is suitable for the separation of neutral as well as charged molecules be means of micellar electrokinetic chromatography (MEKC). The buffers are based on the combination of a long chain alkyl acid, such as lauric acid with ammonium hydroxide or an organic base such as tris-hydroxymethylaminomethane (Tris). The resulting buffer system is able to separate neutral compounds based on its micellar properties. These buffers exhibit much reduced conductivity compared to traditional MEKC buffers, such as sodium dodecylsulfate (SDS), which contain inorganic salts. They also have inherent buffer capacity at high pH resulting from the basic buffer component, which in our studies had pK values from about 8-11. The separations that were observed showed high efficiency with plate counts in many cases above 500,000 plates per meter. The reduced conductivity allowed for the application of much higher electric fields, resulting in very fast analysis times. Alternatively, an increase in detection sensitivity could be achieved, as the reduced conductivity allowed for the use of capillaries with lager internal diameters. Combinations of different alkyl acids and organic bases provided for significant flexibility in selectivity tuning. Finally, the fact that the organic micellar buffer systems discussed here do not contain inorganic ions, allows for coupling with mass spectrometric (MS) detection. The possibility of MS detection combined with the high speed in analysis that can be obtained using these organic buffer systems, could make this approach an interesting option for high throughput analysis of combinatorial libraries.

Membrane adsorbers provide an attractive alternative to traditional bead-based chromatography columns used to remove trace impurities in downstream applications. A linearly scalable novel membrane adsorber family designed for the efficient removal of trace impurities from biotherapeutics, are capable of reproducibly achieving greater than 4 log removal of mammalian viruses, 3 log removal of endotoxin and DNA, and greater than 1 log removal of host cell protein. Single use, disposable membrane adsorbers eliminate the need for costly and time consuming column packing and cleaning validation associated with bead-based chromatography systems, and minimize the required number and volume of buffers. A membrane adsorber step reduces process time, floor space, buffer usage, labor cost, and improves manufacturing flexibility. This "process compression" effect is commonly associated with reducing the number of processing steps. The rigid microporous structure of the membrane layers allows for high process flux operation and uniform bed consistency at all processing scales.

A robust and simple interface for microchip electrophoresis-mass spectrometry (MCE-MS) was developed using a spray nozzle connected to the exit of the separation channel of the microchip. The spray nozzle was attached to the microchip using a polyether ether ketone screw without adhesive, thus allowing easy replaced. Sample injection and electrophoretic separation was performed by control of the voltage only. The analysis of a few basic drugs was performed using the optimized MCE-MS system. The separation was improved by using a high-viscosity separation buffer and a spray nozzle with a small bore size. This system was also applied to the separation of peptides and protein-trypsin digests. Sample adsorption was minimized by adding acetonitrile to the separation buffer when using a quartz microchip.

The influence of the relative buffer zone lengths on the efficiency was investigated in partial filling micellar electrokinetic chromatography using sodium dodecyl sulfate as separation additive. Varying relative zone lengths were obtained by applying identical initial separation zone lengths but different total lengths of the capillaries. Plate numbers of a homologous series of ω-phenylalcohols were measured to indicate the effect of both a changing relative zone length during the run and a counter pressure applied on the cathodic buffer reservoir. The magnitude and the course of these plate numbers are discussed on the basis of models for flow profile superposition and flow profile deformation which are caused by an intersegmental pressure arising at the boundary between the two buffer zones with different electroosmotic flow velocities. Calculation of the intersegmental pressure and of the resulting laminar flow components in the buffer zones on the basis of some equations for electroosmotic and hydrodynamic flow supported the interpretation that a long background buffer zone should be avoided in case of mismatched electroosmotic flows.

The potential of counter pressure-moderated partial filling micellar electrokinetic chromatography (PF-MEKC) was investigated in this work. Plate numbers of homologous omega-phenylalcohols were measured in a two-plug PF-MEKC system varying the concentrations and hence the ionic strengths of the background buffer compared to the sodium dodecyl sulfate-containing separation buffer and the counter pressure on the cathodic buffer reservoir. It was observed that plate numbers are strongly influenced by both the buffer concentrations and the counter pressure. Highest plate numbers were obtained with a buffer system where the concentrations are adjusted such that the electroosmotic flow velocities in both zones are equal. Differences in the local electroosmotic flow velocities of the zones caused by different buffer concentrations are responsible for tremendously reduced plate numbers. The efficiency drop is explained in several models by the formation of an intersegmental pressure which produces a parabolically shaped laminar flow component in both zones. Thus, the electroosmotic plug-like flow profile is distorted and the efficiency is reduced. The effect of counter pressure on efficiency turned out to be very complex in dependence on the buffer system applied.

Multicomponent adsorption data of a fermentation broth containing adipoyl-7-amino-3-deacetoxycephalosporanic acid (adipoyl-7-ADCA), a cephalosporin precursor for 7-ADCA, and two key impurities, alpha-hydroxyadipoyl-7-ADCA and alpha-aminoadipoyl-7-ADCA were obtained from batch equilibrium and frontal chromatography tests. Amberlite XAD-1600 was chosen as the resin. A rate model was applied to simulate the chromatograms. An alkaline buffer, which by itself has no affinity for the resin, was used as the eluent. The widely used reversed-phase modulator model is inaccurate in explaining the stepwise elution data. A new model, the induced competition model, has been developed to account for apparent retention of the buffer in the presence of adsorbed species. Close agreement between the simulations and the data was achieved with the new model.

The separation of proteins using high-performance chromatofocusing with linear or concave pH gradients formed using simple mixtures of buffering species in the elution buffer is investigated experimentally. The separation achieved is comparable to that using polyampholyte elution buffers with these types of systems. More specifically, protein band widths at one half of the band height in the range between 0.1 and 0.025 pH units were observed, and good resolution was achieved of protein variants differing by a single amino acid residue in separation times of 30 min or less. An especially useful elution buffer is investigated that contains only four buffering species and that produces a linear pH gradient in the range between pH 9.5 and 6.0 when used together with a particular high-performance column packing made specifically for chromatofocusing. This elution buffer and column packing combination is evaluated by using it for the chromatofocusing of equine myoglobin and human hemoglobin variants. Additional applications are described in which a polyethyleneimine derivatized silica column packing and a pH gradient that is concave in shape are used for the separation of proteins in an E. coli cell lysate.

A fused-silica capillary that is wall-modified via chemically bonding a sulfonated polymer to the capillary wall has a uniform negative charge density on its surface and produces an electroosmotic flow (EOF) greater than 4 x 10(-4) cm2 V(-1) s(-1) The EOF is nearly independent of buffer pH over the pH range of 2 to 10 and is lower than the EOF obtained for the bare fused-silica capillary at the more basic pH but is higher at the more acidic buffer pH. Optimization of buffer pH can be based on analyte pKa values to improve the overall quality of the capillary zone electrophoresis (CZE) separation of complex mixtures of weak acid and base analytes. Because of the high EOF in an acidic buffer, the capillary is useful for the separation of weak organic bases which are in their cation forms in the acidic buffer. EOF for the sulfonic acid bonded phase capillary can be adjusted via buffer additives such as organic solvent, tetraalkylammonium salts, multivalent cations and alkylsulfonic acids. The advantages of utilizing buffer pH and the EOF buffer modifiers to enhance migration time, selectivity, and resolution in CZE separations with this capillary are illustrated using a series of test analyte mixtures of inorganic anions, carboxylic acids, alkylsulfonic acids, benzenesulfonic acids, sulfas, pyridines, anilines or small-chain peptides.

The optimization of background electrolyte compositions for capillary zone electrophoresis in combination with conductivity detection focusing on maximal detector response is discussed. A theoretical approach pointing out the influence of the electrolyte co- and counter-ion mobilities on the detector signal has been developed. Using this model, running buffer compositions providing optimum S/N ratios for the selected analytes could be calculated. The results derived from these examinations have been verified by experimental investigations, namely the determination of inorganic and organic anionic solutes.

The buffer properties of biopolymers (proteins and nucleic acids), with special reference to electrokinetic methodologies, are considered. The action of biopolymers as potential titrants is also analyzed. Since the ‘buffering capacity’ value ( β ), that is conventionally used, is a local value, we propose to use the concept of ‘buffer reserve’. The latter is obtained by integrating the β value over the necessary pH interval, and thus coincides with the concentration of strong titrant needed to induce the appropriate pH shift. With this parameter any buffer system may be characterized and its value will be a function of the ΔpH shift. The theoretical calculations performed for two real amphoteric substances show that the resistivity limit of such two buffers with the same initial β level may be quite different even in case of rather small pH shifts (several tenths of pH unit).

High quality separations of three-component mixtures of suspended micrometer-size metal oxide particles have been achieved by capillary zone electrophoresis in two different high-pH aqueous buffer systems. Mixtures of Al2O3, Fe3O4 and TiO2 and of Al2O3, Fe3O4 and Fe2O3 were separated in 5–10 min in either carbonate buffer, pH 10.6 or pyrophosphate buffer, pH 10.2. The effects of electric field strength, buffer concentration and buffer type on separation quality were determined. Under optimized conditions, the separations were nearly baseline resolved. Complete separation of a mixture of Al2O3 and UO2 particles was also accomplished. Different sized particles/aggregates of a given oxide were found to co-migrate so that, for the systems studied, particle size effects were negligible and the separations were electrophoretic in nature. Reproducibilities of migration times and peak widths, heights and shapes were found to be excellent, with relative standard deviations of migration times consistently being

The separation of eight purine bases was achieved by capillary electrophoresis in 16 min using a voltage of 15 kV and a 50 mM sodium carbonate–hydrogencarbonate buffer at a pH of 10. Carbonate buffers have many advantages. They are non-toxic, inexpensive, easy to prepare and provide a stable, reproducible electroosmotic flow. Equilibration time after a sodium hydroxide rinse is minimal; thus the total analysis time is shorter than when an acidic buffer is used. No additives to the buffer are required to obtain the majority of the separations of interest and uncoated capillaries can be used. A group of eight purines and the naturally occurring pyrimidines were baseline separated. The linearity correlation for adenine in an optimized separation of the bases was 0.995 over two orders of magnitude. The daily reproducibility was 0.1% and day-to-day reproducibility was 2.5%.

Pentosane polysulfate sodium salt (PPS) is a mixture of multiply charged anionic polysaccharides, used for urological treatment. Several constituents of the polysaccharide can be characterized by a highly reproducible fingerprint. In comparison with earlier approaches the separation efficiency has been further improved using an anionic benzene-1,2,4-tricarboxylic acid buffer (8.7 mmol l-1, pH = 4.9) with indirect UV detection (lambda = 217 nm) and a special capillary pretreatment (1 M NaOH for 10 h at 25 degrees C applying -20 kV). The method has been optimized with regard to buffer concentration and pH. The robustness was tested on several capillaries. PPS was separated from all major synthetic impurities such a sulfate, chloride and acetate. Twelve PPS batches from two manufactures were measured and compared.

Chromatofocusing utilizes internally generated, retained pH gradients to focus proteins into narrow chromatographic bands. One of the characteristics of current chromatofocusing methods is that they use expensive polyampholyte buffers containing large numbers of ill-defined components in order to generate linear or quasi-linear pH gradients. In addition to being costly to use, polyampholyte buffers also tend to associate with proteins and often yield irreproducible gradient shapes. In order to avoid the various difficulties associated with the use of polyampholyte buffers, this study investigates the use of mixtures of simple buffering species to generate quasi-linear pH gradients on a weak-base ion-exchange column packing. The ability of these gradients to separate protein mixtures was also investigated. To optimize the conditions used, a computer simulation method using a local equilibrium model was developed that predicts the shape of the pH gradient. Several experiments were performed that demonstrate the usefulness of the method and the accuracy of the model.

The two synthetic self complementary oligonucleotides pd(AT)12 and pd(GC)12 were separated by free solution capillary electrophoresis (CZE) using simple borate buffers. The effects of pH (7.5–9) and the concentration of the buffer (0.03–0.35 M) were investigated. Higher pH values and buffer concentrations led to better resolution and longer migration times, the pH having a more pronounced effect on the separation than the concentration of the buffer. It is proposed that the conformation and effective length of the oligonucleotides may have a role in their separation in free solution capillary electrophoresis.

The use of capillary zone electrophoresis is described for the separation and identification of selected mixtures of aromatic plant acids. The influence of voltage, temperature, buffer systems and buffer pH on the migration order and migration times was investigated. The best separation was obtained with a fused-silica capillary column (64.5 cm×50 μm I.D.), a running buffer of 20 mM sodium phosphate (pH 7.0), an applied voltage of 25 kV and a temperature of 40°C, combined with diode array detection. Using a methanolic-aqueous extract of Epilobium angustifolium L. as an example, the possibility of using the method developed for the identification of plant acids in plant material is demonstrated. Ferulic, gallic, protocatechuic, cinnamic, caffeic, gentisic and chlorogenic acids could be identified.

We have developed a compact polytetrafluoroethylene (PTFE) assembly-type capillary electrophoresis with chemiluminescence (CL) detection system. Luminol–microperoxidase–hydrogen peroxide chemiluminescence reaction was adopted. The device is rectangular in shape (60 mm × 40 mm × 30 mm) and includes three reservoirs (sample, migration buffer, and detection reservoirs) with electrodes. The detection reservoir includes an optical fiber to transport light at the capillary tip to a photomultiplier tube. Isoluminol isothiocyanate (ILITC) was analyzed as a model using this device with fused-silica or polytetrafluoroethylene capillary tubes 10 cm in length. We also used the sample reservoir as a reactor for an immune reaction between anti-human serum albumin immobilized on glass beads and isoluminol isothiocyanate-labeled human serum albumin. The present polytetrafluoroethylene assembly with the capillary tube was useful as a palm-sized analysis device for separation and detection, as well as a reactor.

A versatile method for the rapid capillary electrophoresis (CE) separation of ribonucleotides using a carbonate buffer was optimized. The separation of twelve 5′-ribonucleotides was obtained in 25 min using a voltage of 18 kV and a 30 m M sodium carbonate/bicarbonate buffer at a pH of 10. For the separation of the monophosphates or the diphosphates, a lower buffer concentration and pH could be used. However, for the triphosphates a 50 m M buffer at pH 11 gave the best peak sharpness and resolution. Therefore, optimal conditions for the separation of the twelve ribonucleotides were a compromise between those best for the mono- and diphosphates and those needed for the triphosphate nucleotides. The sodium carbonate/bicarbonate buffer was inexpensive and easy to prepare. In addition, the high alkalinity of the buffer resulted in a shorter analysis time and the capillary equilibration time with NaOH was minimal compared to an acidic buffer. Since bare capillaries were used, the time and expense of preparing and maintaining coated capillaries was eliminated.

Injection of a “pulse” of a secondary pH buffer just prior to the injection of analyte provides a mild pH gradient within the primary pH buffer. Separations of weakly acid or basic analytes are easily optimized using this gradient. A model explaining the impact of the pulse of secondary buffer is proposed. Experiments verify that the strength of the pH gradient depends on the effective width of the pulse, the difference in mobility of the analyte in the two buffers and on the difference in mobility between the analyte and the pulse of secondary buffer. Sixteen chlorinated phenols are baseline separated in 25 min using a 45-mM phosphate-15 mM tetraborate buffer (adjusted to pH 7.0) when 2.6% of the capillary is filled with 22.5 mM phosphate-7.5 mM tetraborate (adjusted to pH 10.0).

The effect of Mg2+, Cd2+ and Zn2+ are evaluated as buffer additives to reduce electroosmotic flow (EOF) to facilitate the capillary zone electrophoretic (CZE) separation of 2,4-dinitrophenyl (DNP) derivatives of amino acids. As the divalent cation concentration increases EOF decreases and reduction of EOF follows the order Zn2+>Cd2+>Mg2+. The electrophoretic mobility of the DNP-amino acid derivatives remains constant for a given buffer and pH over a change in divalent cation concentration. The effect of the divalent cation buffer additive is consistent with cation-exchange at the silanol sites on the fused-silica wall. Higher buffer pH favors greater cation-exchange and reduced EOF. Separation of multicomponent mixtures of DNP- l -amino acid derivatives in the presence of Mg2+, Cd2+ and Zn2+ are compared at pH 7.00. Derivative migration time, resolution and peak shape are enhanced and are best for Mg2+ as the buffer additive although Mg2+ must be used at a higher concentration than either Cd2+ or Zn2+. Raising the buffer pH to 9.25 in the presence of Mg2+ increases migration time and resolution but analysis time is also increased.

A method has been explored for the production of low background polyacrylamide gradient gel-filled capillaries. The main idea is to prepare the capillaries by filling them with short steps of gelling solutions, including a long section of buffer for detection. This method is demonstrated to be practicable by scanning the filled capillaries. Smooth gradients can be formed at a step-length of 1.5 cm if the relevant capillaries are filled with solutions in a step-down order. By using this method, the desired capillaries were prepared with a success rate of >90% and gradient position error of ±0.62 mm. As shown by capillary gradient gel electrophoresis, this new type of capillary can be used for fast and efficient separation of polyaspartate and murein oligosaccharides (from E. coli ). They also allow the use of any available UV wavelength for high detection sensitivity. The unlabelled oligosaccharides at a total concentration of 1 mg ml −1 were directly detected at 200 nm.

To separate sulfonamides using capillary zone electrophoresis, citrate buffer is superior to phosphate buffer as a background electrolyte. Separation parameters affecting the selectivity and resolution of sulfonamides were optimized. Complete separation of thirteen sulfonamides was rapidly and efficiently achieved within 2.1 min with citrate buffer at pH 6.9. The resolution of peaks between sulfamethoxypyridazine and sulfathiazole, is enhanced either on heating the capillary to 35°C or on adding methanol to the buffer electrolyte in an appropriate proportion. In combination with mobility data obtained at low pH, two p K a values of each individual sulfonamide are satisfactorily determined. Electrophoretic mobilities of sulfonamides measured at optimum pH of the buffer correlate well with those calculated from Offord's equation. Thus the order of migration of these sulfonamides, reflected in the magnitudes of their electrophoretic mobilities, depends on their ratios of charge to mass, and is primarily determined by their p K a values.

Electrolysis of the electrolyte in the buffer reservoirs can occur when performing multiple assays in capillary electrophoresis (CE). This effect, which has been termed buffer depletion, results in a pH gradient being formed across the capillary. This gradient can effect both migration time, peak efficiency and selectivity and therefore can cause problems of reproducibility which leads to poor quantitative assays in pharmaceutical analysis. In this study the factors affecting the extent of buffer depletion were investigated. This involved performing a series of multiple assay runs where a number of factors were varied such as the use of dilute or concentrated buffer solutions and single or separate vials for pre-rinse and separation. In addition, the effect of the volume of buffer in the separation vials on the degree of depletion was also examined using standard (4 ml) and small volume (0.3 ml) CE vials. From these studies, it was observed that long multiple assay runs should not be carried out with small volumes of dilute buffer when pre-rinse and separation steps are from the same vial. The rate of electrolysis also increased when using wide-bore capillaries and at elevated temperatures. The use of short-end injections, zwitterionic buffers and reversing the polarity were found to appreciably reduce the effects of buffer depletion. From this work a number of recommendations are provided to maximise the number of acceptable separations that can be obtained in routine CE injection sequences.

Improved chromatography column fittings were developed for the efficient and reliable application of an electric field to a preparative chromatography column (a process termed electrochromatography). The improved fittings contained electrodes in close proximity to the column packing media and allowed uniform electric fields to be applied. Membranes in the fittings prevented mixing of the electrode and the column eluent buffers. The membranes prevented gases and electrolytic products generated in the electrode chamber from entering the column eluent buffer. An electrode buffer solution was pumped through the electrode chamber to a large external container. The circulation of buffer through the electrode chamber removed the gases and electrolytic products and ensured a uniform electric field by helping to maintain a constant buffer composition. The membranes prevented macromolecules being separated on the column from coming in contact with the electrodes.

Ten β-blockers were simultaneously separated by capillary zone electrophoresis using citrate buffer as a background electrolyte at low pH. The effects of buffer concentration and buffer pH on the migration behavior and separation of β-blockers were systematically investigated. The results indicate that the resolution of co-migrating analytes improves with increasing buffer concentration at a low pH and that the optimum pH of the buffer decreases with increasing buffer concentration. Complete separation of the β-blockers tested was achieved within a relatively short time with an appropriate concentration of citrate buffer in the range 160–400 mM and a proper pH in the range 3.0-1.8 using a 43 cm×50 μm I.D. fused-silica capillary at 15 kV.

The enantiomeric separation of pinacidil was studied using α-cyclodextrin (α-CD), β-CD, heptakis (2,6-di-O-methyl)-β-CD (DM-β-CD), hydroxypropyl-β-CD (HP-β-CD) and γ-CD as buffer additives at acidic pH in Capillary zone electrophoresis. β-CD, HP-β-CD and γ-CD all showed chiral recognition on the compound, with HP-β-CD giving the highest chiral selectivity. Buffer concentration and HP-β-CD concentration were optimized for the separation. The separation was improved by the addition of hydroxypropyl cellulose (HPC) to the buffer. A complete separation of pinacidil enantiomers was achieved by suing 100 mM Tris-H3PO4 buffer (pH 2.3) containing 9 mM HP-β-CD and 0.05% HPC.

A wide-bore capillary electrophoresis apparatus for the analysis of analyte ions is provided. The apparatus has a wide-bore capillary that has a restriction zone at one or more ends. The capillary has an inlet end and an outlet end and an opening at each of said ends. The restriction zone is capable of providing fluid communication between the wide bore and the opening at said end. The restriction zone includes a narrow bore extending to the opening and a transition zone providing gradual change of bore diameter from the wide bore to the narrow bore. The apparatus further has a buffer source to supply buffer to the inlet end of the capillary and a power supply for supplying power to drive buffer and analyte ions through the capillary. During CE, electrodes provide electrical communication between the power supply and the inlet end and the outlet end and apply a voltage differential between said ends.

The composition of background electrolyte (BGE) in capillary electrophoresis (CE) is important for both selectivity and efficiency. Although the importance of pH and ionic strength predominates, attention should also be paid to specific conductivity, heat dissipation and buffering capacity. At a given pH/ionic strength combination, several buffers can in principle be used. Preference should be given to the so called Good type buffers because of the combination of higher buffering capacity and lower specific conductivity. Heat dissipation, leading to disturbances in migration and efficiency can thus be minimised. Buffer performance was calculated using a recently developed buffer selection and evaluation computer program.

In this study the feasibility of using pure organic solvents as separation media in capillary electrophoresis was tested. The chiral separation of individual dansyl-amino acids was compared in aqueous phosphate-borate buffer and in N-methylformamide. In both cases s-cyclodextrin was used as a chiral selector. The low solubility of s-cyclodextrin in water may have hindered the separation of some dansyl-amino acid enantiomers when the aqueous buffer was used. By contrast, N-methylformamide proved to be an excellent solvent for s-cyclodextrin. The chiral separation of ten dansyl-amino acids is demonstrated for s-cyclodextrin in N-methylformamide. The results indicate the potential of organic solvents for chiral separations by capillary electrophoresis.

The migration behavior of dichlorophenols in replicate separations, without replenishment of buffer electrolyte between runs, in micellar electrokinetic capillary chromatography (MECC) was investigated. The results indicated that the pH of the buffer electrolyte drifts as a result of electrolysis of the buffer solution during electrophoresis and that the pH drift, being independent of the micelle concentration, is responsible for the variation in electrophoretic mobilities of analytes in MECC when the buffer is not replenished. The influences of the buffer pH and micelle concentration on the migration behavior of analytes were demonstrated to be correlated at small micelle concentrations. The variation of electrophoretic mobility of an analyte as a function of buffer pH at a given micelle concentration is quantitatively described. 2,6-Dichlorophenol is selected for illustration. The variation of electrophoretic mobility agrees satisfactorily between predicted and observed data.

Metallothionein (MT) isoforms, sulphydryl-rich proteins with a high affinity for both essential and toxic trace metals, are readily separated by capillary zone electrophoresis (CZE) with an uncoated polyimide-clad fused-silica capillary using Tris-borate buffer. The metallothionein samples investigated were rabbit liver MT, rabbit liver MT-1, rabbit liver MT-2 and horse kidney MT. The effects of temperature, buffer pH, buffer concentration, buffer composition and running voltage on the separation efficiency were investigated. An improved separation efficiency compared with published methods using uncoated fused-silica capillaries was obtained, allowing the separation of peaks which previously co-migrated. The running conditions used were voltage 11 kV, temperature 20°C and buffer 110 mM Tris-110 mM borate (pH 6.90). Under these conditions, horse kidney MT exhibited five putative isoform peaks and rabbit liver MT nine or ten peaks. Interestingly, the experiments indicated that rabbit liver MT contains one major component which is not found in either isoform MT-1 or MT-2, possibly indicating that the improved separation efficiency made it possible to detect a new group or structure. Based on this initial evaluation, CZE with an uncoated fused-silica capillary using Tris-borate buffer appears to be a very useful method for the separation of metallothionein isoforms.

Adapting well-characterized existing liquid chromatographic separations for compatibility with mass spectrometric detection often requires changes in chemical and/or chromatographic conditions. Non-volatile solvent modifiers, such as phosphates or alkylsulfonates, have been widely used in chromatography because of their buffer or ion-pairing capacities, coupled with advantageous low-UV-wavelength absorption characteristics. Non-volatile components in the eluent stream are generally avoided in mass spectrometry as they quickly decrease the detector response. In addition to issues such as buffer capacity and effective ion-pairing, consideration should be given to the impact of stationary phase and column geometry. The work presented here illustrates advantages of a new high-purity, high-pore-volume stationary phase used in Waters Symmetry columns when attempting to accommodate a wide range of conditions imposed by different detection schemes. The effect of column geometry under a variety of typical LC-MS conversion situations is also included. The studies include resolution of critical hydrophilic pairs and stationary-phase loading performance relative to converting existing separations for LC-MS analysis while achieving maximum sensitivity and maintaining chromatographic fidelity.

Bovine whey comprises four major protein groups [σ-lactalbumin, β-lactoglobulin variants A and B, bovine serum albumin and immunoglobulin (specifically IgG) which have a diverse range of molecular masses, pI values, number of phenotypic variants and subunit compositions. The development of a capillary zone electrophoresis method to separate these whey proteins is described. Initially separation of the individual whey proteins was evaluated using a number of different buffer systems with pKa values above pH 7. At buffer pH values greater than 7, protein-capillary wall interactions were minimized as the majority of the whey proteins had a net negative charge because their pI values are in the range pH 4–6. A wide range of buffer additives (organic modifiers and surfactants) were also added to alter the chemistry of the separation, to further block protein-capillary wall interactions and to thus optimize the resolution of the different protein peaks. A sample buffer/separation buffer system was developed which eliminated an initial solvent trough that coincided with the IgG peak. This made it possible to quantify the IgG protein. Optimum resolution and analysis time (10 min) for the four whey proteins was achieved with a sample buffer consisting of 10 mM phosphate, pH 7.4 and a separation buffer consisting of 150 mM sodium borate, pH 8.5 containing 0.05% Tween 20. This method was succesfully used to separate a mixture of commercially purified whey proteins and to separate and quantitate the individual whey proteins in an acid whey sample.

Nine synthetic organic dyes, including seven azo compounds, used as colouring matter for textiles were separated as anions by capillary zone electrophoresis. As most of the solutes are sulfonic acids, the separation could not be effected by varying the pH of the buffer solution. Therefore, two methods were applied to adjust the electrophoretic mobility in a specific way: complexation by bis-tris-propane and interaction with linear polymers added to the buffer and acting as pseudo-phases. A buffer system containing polyethylene glycol and polyvinylpyrrolidone permits the separation of all analytes. Retention of the dyes caused by the polymeric additives was related to the solute's structure. It was demonstrated by cluster analysis that the relative decrease in the electrophoretic mobility of the dyes correlates with the number of benzoaromatic rings in the solute molecules.

This paper describes the application of combined orthogonal array design and overlapping resolution mapping to the optimization of capillary zone electrophoresis for the separation of heterocyclic amines. The factors affecting resolution, such as buffer pH, organic modifier, concentration of buffer solution, capillary tubing temperature, and electric field strength, were studied in two steps. In the first step, orthogonal array design was used to determine the most important factors and interactions. The experiments were carried out according to an OA16 (215) matrix through sixteen pre-designed trials. Based on the results of the first step, the second set of experiments was performed according to a three-dimensional overlapping resolution mapping scheme, in which eleven pre-planned trials were executed and global optimum conditions for the separation within a reasonable analysis time were obtained. The proposed conditions were successfully applied in the separation of thirteen heterocyclic amines.