Effect of the ionic strength of salts on retention and overloading behavior of ionizable compounds in reversed-phase liquid chromatography I. XTerra-C18

The influence of the salt concentration (potassium chloride) on the retention and overloading behavior of the propranolol cation (R'-NH2+ -R) on an XTerra-C18 column, in a methanol:water solution, was investigated. The adsorption isotherm data were first determined by frontal analysis (FA) for a mobile phase without salt (25% methanol, v/v). It was shown that the adsorption energy distribution calculated from these raw adsorption data is bimodal and that the isotherm model that best accounts for these data is the bi-Moreau model. Assuming that the addition of a salt into the mobile phase changes the numerical values of the parameters of the isotherm model, not its mathematical form, we used the inverse method (IM) of chromatography to determine the isotherm with seven salt concentrations in the mobile phase (40% methanol, v/v; 0, 0.002, 0.005, 0.01, 0.05, 0.1 and 0.2 M). The saturation capacities of the model increase, q(s,1) by a factor two and q(s,2) by a factor four, with increasing salt concentration in the range studied while the adsorption constant b1 increases four times and b2 decreases four times. Adsorbate-adsorbate interactions vanish in the presence of salt, consistent with results obtained previously on a C18-Kromasil column. Finally, besides the ionic strength of the solution, the size, valence, and nature of the salt ions affect the thermodynamic as well as the mass transfer kinetics of the adsorption mechanism of propranolol on the XTerra column.     

Application of mass spectrometry for study of the adsorption of multicomponent surfactant mixtures at the solid/solution interface

The selective adsorption of the components of a polydisperse gemini surfactant blend (alkylbenzenesulfonate-Jeffamine salt, ABSJ) in aqueous solution onto Berea sandstone, a reference material in enhanced oil recovery (EOR), was investigated. The individual adsorption isotherms of the four, benzene-ring containing ABSJ components with different alkyl chain lengths (ranging from decyl to tridecyl of the alkyl chain length) were simultaneously determined by using a four-channel electrospay ionization mass spectrometer (ESI-MS) for concentration analysis. This analytical device provided selective information (based on the differences in the mass to charge ratio) on the adsorption of each component in the mixed surfactant system. The overall isotherm obtained from the superposition of the individual isotherms determined by ESI-MS agreed well with the isotherm determined by UV spectrometry; the UV equipment is benzene-ring sensitive, irrespective of the alkyl chain length. The S-shaped isotherms reached a plateau at the critical micelle concentration. Longer-chain surfactants adsorbed preferentially over the short chain homologs, independently of solution concentration. This analytical device provided the net adsorption isotherm. Most analytical methods are not component selective, and thus they are not able to measure the individual isotherms in multicomponent solutions. Here, we report on a novel method which describes the selective determination of the individual adsorption isotherms of surfactants in a multicomponent mixture. The theoretical background of the method is described in detail.     

High-throughput isotherm determination and thermodynamic modeling of protein adsorption on mixed mode adsorbents

The thermodynamic modeling of protein adsorption on mixed-mode adsorbents functionalized with ligands carrying both hydrophobic and electrostatic groups was undertaken. The developed mixed mode isotherm was fitted with protein adsorption data obtained for five different proteins on four different mixed mode adsorbents by 96-well microtitre plate high throughput batch experiments on a robotic workstation. The developed mixed mode isotherm was capable of describing the adsorption isotherms of all five proteins (having widely different molecular masses and iso-electric points) on the four mixed mode adsorbents and over a wide range of salt concentrations and solution pH, and provided a unique set of physically meaningful parameters for each resin-protein-pH combination. The model could capture the typically observed minimum in mixed mode protein adsorption and predict the precise salt concentration at which this minimum occurs. The possibility of predicting the salt concentration at which minimum protein binding occurs presents new opportunities for designing better elution strategies in mixed mode protein chromatography. Salt-protein interactions were shown to have important consequences on mixed mode protein adsorption when they occur. Finally, the mixed mode isotherm also gave very good fit with literature data of BSA adsorption on a different mixed mode adsorbent not examined in this study. Hence, the mixed mode isotherm formalism presented in this study can be used with any mixed mode adsorbent having the hydrophobic and electrostatic functional groups. It also provides the basis for detailed modeling and optimization of mixed mode chromatographic separation of proteins.     

The significance of a second adsorption phase with weakly adsorbed species for the calculation of the surface concentrations of a mixture: methanol–DME and methanol–ethene adsorption in SAPO-34

With the calorimetric (adsorption heat versus coverage) curve also measured together with the adsorption isotherm, the simultaneous use of both curves showed that there were two phases of adsorption in the adsorption of methanol, dimethyl ether, ethene and propane in SAPO-34. The dual-site Langmuir equation gave good fits to the adsorption data to support the interpretation that a second (type 2) adsorption phase occurred in the high-pressure region in addition to a first (type 1) adsorption phase on the acid sites at lower pressures. Adsorption experiments and calculations using binary gas mixtures showed that due to the existence of two types of adsorption, the multicomponent Langmuir isotherm equation (Langmuir competitive adsorption model) calculated incorrect surface concentrations when the concentrations were high. In contrast, the ideal adsorbed solution theory (IAST) calculated correct surface concentrations in the adsorption of mixtures.     

Effect of charge regulation on steric mass-action equilibrium for the ion-exchange adsorption of proteins

A thermodynamic formalism is developed for incorporating the effects of charge regulation on the ion-exchange adsorption of proteins under mass-overloaded conditions as described by the steric mass-action (SMA) isotherm. To accomplish this, the pH titration behavior of a protein and the associated adsorption equilibrium of the various charged forms of a protein are incorporated into a model which also accounts for the steric hindrance of salt counterions caused by protein adsorption. For the case where the protein is dilute, the new model reduces to the protein adsorption model described recently by the authors which accounts for charge regulation. Similarly, the new model reduces to the steric mass-action isotherm developed by Brooks and Cramer which applies to mass-overloaded conditions for the case where charge regulation is ignored so that the protein has a fixed charge. Calculations using the new model were found to agree with experimental data for the adsorption of bovine serum albumin (BSA) on an anion-exchange column packing when using reasonable physical properties. The new model was also used to develop an improved theoretical criterion for determining the conditions required for an adsorbed species to displace a protein in displacement chromatography when the pH is near the protein pI.     

Multicomponent adsorption modeling: isotherms for ABE model solutions using activated carbon F-400

Biobutanol has attracted significant interest in recent decades and is seriously considered as a potential biofuel to partly replace gasoline. However, some production challenges must be addressed to make butanol economically viable such as the low product concentration and product toxicity inhibiting the microorganism. To alleviate these limitations, several in situ or ex situ separation techniques have been investigated in view of their integration to the biobutanol production process to enhance its economic viability. One of these techniques is adsorption which is one of the most energy-efficient techniques used for biobutanol separation. Considering the number of chemical species present in the ABE fermentation broth, it is essential to develop multicomponent adsorption isotherms for all components as a first step to design a high performance adsorption process. Few multicomponent isotherm models have been proposed such as multicomponent Langmuir and Freundlich. In this study, these two models as well as artificial neural networks were used to model the isotherms of each component in an ABE fermentation broth as a function of the equilibrium concentrations of all components for activated carbon F-400. Results showed that the multicomponent Langmuir model was not accurate due to the many simplifying assumptions. The multicomponent Freundlich and feedforward neural network (FFNN) isotherm models were able to predict the behavior of multicomponent systems very well. Indeed, the predictive model of the experimental data had a coefficient of determination (R²) of 0.97 and 0.99, for multicomponent Freundlich and FFNN isotherm models, respectively.     

New algorithms for the computation of column dynamics of multicomponent liquid phase adsorption

New and efficient numerical algorithms were developed for simulating column dynamics of multicomponent liquid phase adsorption. Simple and realistic models are used for the simulation. Langmuir form of isotherm and linear driving force rate expressions are employed in the model equations. Algorithms were formulated for three different rate control mechanisms, namely, film diffusion control, particle diffusion control and combined film and particle diffusion control. The algorithms derived are explicit with the exception of the requirement of solving a nonlinear equation in one single variable which is the concentration of a reference species. Thus the tedious iterative calculation procedure for solving simultaneous nonlinear equations in a multicomponent fixed bed system is avoided. Example calculations indicated very good numerical accuracy as verified from an independent check by means of an overall mass balance.     

Modeling of adsorption in hydrophobic interaction chromatography systems using a preferential interaction quadratic isotherm

A preferential interaction quadratic isotherm model for hydrophobic interaction chromatographic systems is presented in this paper. In this isotherm, the nonlinear effect of salt on the capacity factor is described using the preferential interaction model developed by Perkins et al. [J. Chromatogr. A, 766 (1997) 1]. This is then coupled with a quadratic nonlinear isotherm to describe nonlinear adsorption behavior at high solute concentrations. The resulting preferential interaction quadratic isotherm is examined for its ability to describe solute adsorption behavior under both linear and nonlinear conditions over a wide range of salt concentrations in HIC systems. The results indicate that this isotherm is well suited for predicting nonlinear adsorption behavior in HIC systems for both proteins and low-molecular mass HIC displacers.     

Effect of the ionic strength of salts on retention and overloading behavior of ionizable compounds in reversed-phase liquid chromatography II. Symmetry-C18

In a companion paper, we describe the influence of the concentration and the nature of salts dissolved in the mobile phase (methanol:water, 40:60, v/v) on the adsorption behavior of propranolol (R'-NH2+ -R, Cl-) on XTerra-C18. The same experiments were repeated on a Symmetry-C18 column to compare the adsorption mechanisms of this ionic compound on these two very different RPLC systems. Frontal analysis (FA) measurements were first carried out to determine the best isotherm model accounting for the adsorption behavior of propranolol hydrochloride on Symmetry with a mobile phase without salt (and only 25% methanol to compensate for the low retention in the absence of salt). The adsorption data were best modeled by the bi-Moreau model. Large concentration band profiles of propranolol were recorded with mobile phases having increasing KCl concentrations (0, 0.002, 0.005, 0.01, 0.05, 0.1 and 0.2 M) and the best values of the isotherm coefficients were determined by the inverse method (IM) of chromatography. The general effect of a dissociated salt in the mobile phase was the same as the one observed earlier with XTerra-C18. Increasing the salt concentration increases the two saturation capacities of the adsorbent and the adsorption constant on the low-energy sites. The adsorption constant on the high-energy sites decreases and the adsorbate-adsorbate interactions tend to vanish with increasing salt concentration of the mobile phase. The saturation capacities decrease with increasing radius of the monovalent cation (Na+, K+, Cs+, etc.). Using sulfate as a bivalent anion (Na2SO4) affects markedly the adsorption equilibrium: the saturation capacities are drastically reduced, the high-energy sites nearly disappear while the adsorption constant and the adsorbate-adsorbate interactions on the low-energy sites increase strongly. The complexity of the thermodynamics in solution might explain the different influences of these salts on the adsorption behavior.     

Contribution of concentration-dependent surface diffusion in ternary adsorption kinetics of ethane, propane and n-butane in activated carbon

The role of concentration-dependent surface diffusion in the adsorption kinetics of a multicomponent system is investigated in this paper. Ethane, propane and n-butane are selected as the model adsorbates and Ajax activated carbon as the model adsorbent. Adsorption equilibrium isotherm and dynamic parameters extracted from single-component systems are used to predict the ternary adsorption equilibria and kinetics. The effect of concentration-dependent surface diffusion on the adsorption kinetics predictions is studied by comparing the results of two mathematical models with the experimental data. Three diffusion mechanisms, macropore, surface and micropore diffusions are incorporated in both models. The distinction between these two models is the use of the chemical potential gradient as the driving force for the diffusion of the adsorbed species in one model and the concentration gradient in the other. It was found that the model using the chemical potential gradient provides a better prediction of the ternary adsorption kinetics data, suggesting the importance of the concentration dependency of the surface diffusion, which is implicitly reflected in the chemical potential gradient. The kinetic model predictions are also affected by the way how single-component adsorption equilibrium isotherm data are fitted.     

An Analysis of the Interactions of BSA with an Anion-Exchange Surface Under Linear and Non-Linear Conditions

The interactions of BSA with an anion-exchange adsorbent have been studied to aid in the understanding of protein adsorption in ion-exchange chromatography. Linear chromatography, flow microcalorimetry and isotherm measurements were used to analyze adsorption energetics in the linear and overloaded regions of the equilibrium isotherm. The effects of salt type, salt and protein concentration, and temperature are reported. It was observed that under all conditions studied the adsorption process was entropically driven. This was contrary to expectations, since at the pH selected ion exchange is expected to dominate. A major driving force for the adsorption of BSA on the anion exchanger was concluded to be the increase in entropy from the release of water due to interactions between hydrophobic regions on the protein and adsorbent. The data further suggest that the conformational entropy change accompanying protein adsorption on the ion exchanger may also be significant.     

Adsorption of the enantiomers of 3-chloro-1-phenyl-propanol on silica-bonded chiral quinidine carbamate

The interactions of 3-chloro-1-phenyl-propanol with a quinidine carbamate-bonded chiral stationary phase under NPLC conditions were studied by measuring the adsorption isotherm data of its enantiomers by frontal analysis, modeling these data with a suitable isotherm model, and comparing the experimental overloaded elution band profiles with those calculated with this isotherm and the equilibrium dispersive model of liquid chromatography. The affinity energy distribution was calculated from the adsorption isotherm data. The results show that the surface of the adsorbent is heterogeneous and exhibits a bimodal adsorption energy distribution. This fact is interpreted in terms of the presence of two different types of adsorption sites on the stationary phase, nonselective and enantioselective sites. Albeit the bi-Langmuir isotherm model successfully accounts for the single-component data corresponding to both enantiomers, the competitive bi-Langmuir isotherm model does not allow an accurate prediction of the overloaded band profiles of the racemic mixture. Thermodynamic data are drawn for explanation. Some aspects of the retention mechanism are discussed in the light of the data obtained.     

Ion-exchange displacement chromatography of proteins Dextran-based polyelectrolytes as high affinity displacerss

Dextran-based polyelectrolyte displacers were successfully employed for the displacement purification of proteins in ion-exchange displacement systems. The effect of molecular mass was investigated by examining the efficacy of DEAE-dextran and dextran sulfate displacers of various molecular masses in cation- and anion-exchange systems, respectively. Induced salt gradients produced during these displacement experiments were measured in order to study their effect on the protein separations. The unique characteristics of these displacements were well predicted by simulations obtained from a steric mass action (SMA) ion-exchange model. These displacements differ from the traditional vision of displacement chromatography in several important ways: the isotherm of the displacer does not necessarily lie above the feed component isotherms; the concentration of the displaced proteins can sometimes exceed that of the displacer; higher-molecular-mass displacers are not necesarily more efficacious than lower-molecular-mass compounds; and the salt gradients induced by the adsorption of the displacer produce different salt micro-environments for each displaced protein.     

Adsorption-desorption isotherm hysteresis of phenol on a C18-bonded surface

Single component adsorption and desorption isotherms of phenol were measured on a high-efficiency Kromasil-C18 column (N = 15000 theoretical plates) with pure water as the mobile phase. Adsorption isotherm data were acquired by frontal analysis (FA) for seven plateau concentrations distributed over the whole accessible range of phenol concentration in pure water (5, 10, 15, 20, 25, 40, and 60 g/l). Desorption isotherm data were derived from the corresponding rear boundaries, using frontal analysis by characteristic points (FACP). A strong adsorption hysteresis was observed. The adsorption of phenol is apparently modeled by a S-shaped isotherm of the first kind while the desorption isotherm is described by a convex upward isotherm. The adsorption breakthrough curves could not be modeled correctly using the adsorption isotherm because of a strong dependence of the accessible free column volume on the phenol concentration in the mobile phase. It seems that retention in water depends on the extent to which the surface is wetted by the mobile phase, extent which is a function of the phenol concentration, and of the local pressure rate, which varies along the column, and on the initial state of the column. By contrast, the desorption profiles agree well with those calculated with the desorption isotherms using the ideal model, due to the high column efficiency. The isotherm model accounting best for the desorption isotherm data and the desorption profiles is the bi-Langmuir model. Its coefficients were calculated using appropriate weights in the fitting procedure. The evolution of the bi-Langmuir isotherm parameters with the initial equilibrium plateau concentration of phenol is discussed. The FACP results reported here are fully consistent with the adsorption data of phenol previously reported and measured by FA with various aqueous solutions of methanol as the mobile phase. They provide a general, empirical adsorption model of phenol that is valid between 0 and 65% of methanol in water.     

Recombinant protein purification using gradient-assisted simulated moving bed hydrophobic interaction chromatography. Part I: selection of chromatographic system and estimation of adsorption isotherms

The design of gradient simulated moving bed (SMB) chromatographic processes requires an appropriate selection of the chromatographic system followed by the determination of adsorption isotherm parameters in the relevant range of mobile phase conditions. The determination of these parameters can be quite difficult for recombinant target proteins present in complex protein mixtures. The first part of this work includes the estimation of adsorption isotherm parameters for streptokinase and a lumped impurity fraction present in an Escherichia coli cell lysate for a hydrophobic interaction chromatography (HIC) matrix. Perturbation experiments were carried out using a Butyl Sepharose matrix with purified recombinant protein on buffer equilibrated columns as well as with crude cell lysate saturated columns. The Henry constants estimated for streptokinase were found to exhibit in a wide range a linear dependence on the salt concentration in the mobile phase. These parameters were applied in subsequent investigations to design a simulated moving bed (SMB) process capable to purify in a continuous manner recombinant streptokinase from the E. coli cell lysate.     

Comparison between the adsorption behaviors of an organic cation and an organic anion on several reversed-phase liquid chromatography adsorbents

Adsorption data of an organic cation (propranololium chloride) and an organic anion (sodium 1-naphthalene sulfonate) were measured by frontal analysis on two RPLC adsorbents, Symmetry-C18 and XTerra-C18, with aqueous solutions of methanol as the mobile phases. The influence of supporting neutral salts on the adsorption behavior of these two ions are compared. The Henry constants are close (H approximately 5). The four sets of isotherm data are all well accounted for using the bi-Moreau model. However, the isotherms of the two ions behave differently at high concentrations. The initial behaviors of all the isotherms are antilangmuirian but remain so in a much wider concentration range for the cation than for the anion, due to its stronger adsorbate-adsorbate interactions on the low-energy adsorption sites. The retention times of both ions increase with increasing concentration of neutral salt in the mobile phase, suggesting the formation of ion-pair complexes, with Cl- for the cation and with Na+ for the anion. The adsorbate-adsorbate interactions vanish in the presence of salt and the bi-Moreau isotherm model tends toward a bi-Langmuir model. Differences in adsorption behavior are also observed between the cation and the anion when bivalent inorganic anions and cations, respectively, are dissolved in the mobile phase. High concentration band profiles of 1-naphthalene sulfonic acid are langmuirian, except in the presence of a trivalent cation, while those of propranolol are antilangmuirian under certain conditions even with uni- or divalent cations.     

Adsorption-desorption isotherm hysteresis of beta-lactoglobulin A with a weakly hydrophobic surface.

Adsorption-desorption isotherms of bovine beta-lactoglobulin A (beta-lact A) on a weakly hydrophobic stationary phase (C1-ether) were measured by frontal analysis. The adsorption isotherms obtained at different pH were found to be dramatically different in shape, column capacity and desorption reversibility. At pH 4.5, an S-shaped adsorption isotherm was observed whereas at pH 6.0 a Langmuir isotherm was found. In addition, the desorption isotherm at pH 6.0 was found to overlap with the adsorption isotherm, and the adsorption-desorption process of beta-lact A under this condition could be characterized by a fully reversible Langmuir model. The desorption isotherm at pH 4.5, however, did not retrace the adsorption isotherm, resulting in hysteresis loops. A higher aggregate (tetramer) of beta-lact A is shown to be in an equilibrium with the beta-lact A protomer (dimer) at pH 4.5 whereas the dimer alone is predominant at pH 6.0. It is further shown that changes in the absorption coefficient between the adsorption and the desorption cycles for the tetramer at pH 4.5 can account for the hysteresis. The results demonstrate that pH can be a sensitive parameter in protein adsorption isotherm behavior and ultimately the behavior of species in preparative-scale chromatography.     

Competitive adsorption isotherm modelling of heterocyclic nitrogenous compounds,pyridine and quinoline,onto granular activated carbon and bagasse fly ash

In the present study, simultaneous adsorption of quinoline and pyridine onto adsorbents such as granular activated carbon (GAC) and bagasse fly ash (BFA) from pyridine–quinoline binary aqueous solution was studied at various temperatures (288–318 K). Gathered equilibrium adsorption data were further analysed using various multicomponent competitive isotherm models such as non-modified and modified competitive Langmuir isotherms, extended-Langmuir isotherm, extended-Freundlich model, Sheindorf–Rebuhn–Sheintuch (SRS) model, and non-modified and modified competitive Redlich–Peterson isotherm model. It was observed that increase in pyridine concentration decreased the total adsorption yield and the individual adsorption yield for both the quinoline and pyridine for both the adsorbents GAC and BFA at all the temperatures studied. Identical trend was observed during the equilibrium uptake of pyridine on to GAC and BFA with an increase in quinoline concentration. The extended-Freundlich model satisfactorily represented the binary adsorption equilibrium data of quinoline and pyridine onto GAC and BFA.