Marko Jovanović

Comparison of Full Factorial Design, Central Composite Design, and Box-Behnken Design in Chromatographic Method Development for the Determination of Fluconazole and Its Impurities

This paper presents the development and optimization of a liquid chromatographic method for the determination of fluconazole and its impurities by experimental design methodology. Four experimental design types were applied: two-level full factorial design, central composite design, Box-Behnken design, and three-level full factorial design. The advantages and drawbacks of each design are described and detailed statistical evaluation of mathematical models was performed. The central composite design and three-level full factorial design created significantly better models comparing to the other methods. As the central composite design requires a smaller number of experiments, its models were used for theoretical examination of experimental space. Multiobjective optimization aiming to achieve maximal separation of all investigated substances and minimal analysis duration was performed by a grid point search. The defined optimal separation was achieved on a C18 (125 mm × 4 mm, 5 µm particle size) column with a mobile phase consisting of acetonitrile and water (5 mM ammonium formate) (15:85, v/v); a column temperature of 25°C; a flow rate of 1.2 mL min−1; and a detection wavelength of 260 nm.

Thorough investigation of the retention mechanisms and retention behavior of amides and sulfonamides on amino column in hydrophilic interaction liquid chromatography.

In this paper detailed analysis of a mixture of four amides (tropicamide, nicotinamide, tiracetam, and piracetam) and six sulfonamides (sulfanilamide, sulfacetamide, sulfamethoxazole, sulfafurazole, furosemide, and bumetanide) on aminopropyl column in hydrophilic interaction chromatography (HILIC) was carried out. Since, there are no papers on the topic of the assessment of the contribution of ion-exchange retention mechanism involved in the separation of the acidic compounds on aminopropyl column in HILIC mode, the authors utilized the retention data of the acidic sulfonamides for this purpose. Next, broad range of the aqueous buffer concentrations in the mobile phase was examined providing the separation under either HILIC or RP conditions. Turning points between these two mechanisms were determined and then the fitting of the experimental data in the localized and non-localized adsorption models in both RP and HILIC regions was assessed. Since not many papers in the literature were dealing with the estimation of factor influence on the retention behavior of neutral and acidic compounds on aminopropyl column in HILIC, Box-Behnken design and Response Surface Methodology were applied. On the basis of the obtained data, ten quadratic models were proposed and their adequacy was confirmed using ANOVA test. Furthermore, retention data was graphically evaluated by the construction of 3D response surface plots. Finally, good predictive ability of the suggested models was proved with five additional verification experiments.

Quality by Design approach in the development of hydrophilic interaction liquid chromatographic method for the analysis of iohexol and its impurities

This study presents the development of hydrophilic interaction liquid chromatographic method for the analysis of iohexol, its endo-isomer and three impurities following Quality by Design (QbD) approach. The main objective of the method was to identify the conditions where adequate separation quality in minimal analysis duration could be achieved within a robust region that guarantees the stability of method performance. The relationship between critical process parameters (acetonitrile content in the mobile phase, pH of the water phase and ammonium acetate concentration in the water phase) and critical quality attributes is created applying design of experiments methodology. The defined mathematical models and Monte Carlo simulation are used to evaluate the risk of uncertainty in models prediction and incertitude in adjusting the process parameters and to identify the design space. The borders of the design space are experimentally verified and confirmed that the quality of the method is preserved in this region. Moreover, Plackett-Burman design is applied for experimental robustness testing and method is fully validated to verify the adequacy of selected optimal conditions: the analytical column ZIC HILIC (100 mm × 4.6 mm, 5 μm particle size); mobile phase consisted of acetonitrile-water phase (72 mM ammonium acetate, pH adjusted to 6.5 with glacial acetic acid) (86.7:13.3) v/v; column temperature 25 °C, mobile phase flow rate 1 mL min(-1), wavelength of detection 254 nm.

Assessment of β‐lactams retention in hydrophilic interaction chromatography applying Box–Behnken Design

In this paper, the retention prediction models for mixture of β-lactam antibiotics analyzed by hydrophilic interaction chromatography (HILIC) are presented. The aim of the study was to investigate the retention behavior of some organic acids and amphoteric compounds including cephalosporins (cefotaxime, cefalexin, cefaclor, cefuroxime, and cefuroxime axetil) and penicillins (ampicillin and amoxicillin). Retention of substances with acidic functional group in HILIC is considered to be interesting since the majority of publications in literature are related to basic compounds. In the beginning of the study, classical silica columns were chosen for the retention analysis. Then, preliminary study was done and factors with the most significant influence on the retention factors were selected. These factors with the impact on the retention factors were investigated employing Box-Behnken design as a tool. On the basis of the obtained results the mathematical models were created and tested using ANOVA test and finally verified. This approach enables the presentation of chromatographic retention in many ways (three-dimensional (3-D) graphs and simple two-dimensional graphical presentations). All of these gave the possibility to predict the chromatographic retention under different conditions. Furthermore, regarding the structure of the analyzed compounds, the potential retention mechanisms in HILIC were suggested.

THEORETICAL AND EMPIRICAL MODELS IN HYDROPHILIC INTERACTION LIQUID CHROMATOGRAPHY

□ Hydrophilic interaction liquid chromatography (HILIC) has found application in the determination of polar molecules. This alternative type of chromatography has become increasingly popular in the last five years and a considerable number of articles regarding the elucidation of its retention mechanisms and the development of methods are available. In order to achieve these goals, various types of theoretical and empirical mathematical models were utilized. However, the knowledge obtained from these models was the subject of various interpretations. The goal of this article is to list and discuss the papers in which these models were used in HILIC and give proper theoretical background of the models. In general, theoretical models in HILIC were applied in order to acquire the knowledge of dominant processes in the retention mechanism, retention prediction, and quantitative description of the influence of mobile phase-related factors on the retention and selectivity of the analytes on different types of stationary phases. In addition, empirical models were used for detailed assessment of the retention behavior in terms of both single factor and factor interactions influence on analyte retention and selectivity. Furthermore, they have also shown a great potential in the optimization of HILIC methods in the method development process.

Hydrophilic interaction liquid chromatography in analysis of granisetron HCl and its related substances. Retention mechanisms and method development

In this paper separation of granisetron and its two related substances in HILIC mode is presented. Separation was done on silica column derivatized with sulfoalkylbetaine groups (ZIC-HILIC). Firstly, retention mechanisms were assessed whereby retention factors of substances were followed in wide range of acetonitrile content (80-97%), at constant concentration of aqueous buffer (10mM) as well as at constant pH value of 3.0. Further, in order to developed optimal HILIC method, Design of Experiments (DoE) methodology was applied. For optimization full factorial design 3(2) was employed. Influence of acetonitrile content and ammonium acetate concentration were investigated while pH of the water phase was kept at 3.3. Adequacy of obtained mathematical models was confirmed by ANOVA. Optimization goals (α>1.15 and minimal run time) were accomplished with 94.7% of acetonitrile in mobile phase and 70 mM of ammonium acetate in water phase. Optimal point was in the middle of defined Design Space. In the next phase, robustness was experimetally tested by Rechtschaffen design. The investigated factors and their levels were: acetonitrile content (±1%), ammonium acetate molarity in water phase (±2 mM), pH value of water phase (±0.2) and column temperature (±4 °C). The validation scope included selectivity, linearity, accuracy and precision as well as determination of limit of detection (LOD) and limit of quantification (LOQ) for the related substances. Additionally, the validation acceptance criteria were met in all cases. Finally, the proposed method could be successfully utilized for estimation of granisetron HCl and its related substances in tablets and parenteral dosage forms, as well as for monitoring degradation under various stress conditions.

Five different columns in the analysis of basic drugs in hydrophilic interaction liquid chromatography

AbstractFive different columns (two silica, two cyanopropyl and one diol) were investigated in hydrophilic interaction liquid chromatography (HILIC). For the assessment of columns behavior in HILIC mode, six basic drugs (lamotrigine, thioridazine, clozapine, chlorpheniramine, pheniramine and sulpiride) were chosen. The assessment of the influence of the concentration of organic modifier on analytes’ retention on each column was provided by fitting the retention data into theoretical models. Utilizing the statistical analysis, the selection of the model that provides better prediction of the retention behavior was enabled. Dual RP-HILIC mechanism was suggested on cyanopropyl and diol columns, therefore the transition points between the two mechanisms on these columns were calculated. Furthermore, in order to investigate the impact of three factors simultaneously on the retention behavior of the analyzed substances on Betasil Silica column, chemometrically-aided empirical models were built. The experiments were conducted according to the matrix of Box-Behnken design and on the basis of the retention data, six quadratic models were obtained and their adequacy was confirmed using ANOVA test. The obtained coefficients of quadratic models enabled the elucidation of both single factor and factor interactions influence. This was also graphically presented in 3D response surface plots.

Retention mechanism assessment and method development for the analysis of iohexol and its related compounds in hydrophilic interaction liquid chromatography

AbstractHydrophilic interaction liquid chromatography (HILIC) has emerged in recent years as a valuable alternative to reversed-phase liquid chromatography in the analysis of polar compounds. Research in HILIC is divided into two directions: the assessment of the retention mechanism and retention behavior, and the development of HILIC methods. In this work, four polar neutral analytes (iohexol and its related compounds A, B, and C) were analyzed on two silica and two diol columns in HILIC mode with the aim to investigate thoroughly the retention mechanisms and retention behavior of polar neutral compounds on these four columns. The adsorption and partition contribution to the overall HILIC retention mechanism was investigated by fitting the retention data to linear (adsorption and partition) and nonlinear (mixed-retention and quadratic) theoretical models. On the other hand, the establishment of empirical second-order polynomial retention models on the basis of D-optimal design made possible the estimation of the simultaneous influence of several mobile-phase-related factors. Furthermore, these models were also used as the basis for the application of indirect modeling of the selectivity factor and a grid point search approach in order to achieve the optimal separation of analytes. After the optimization goals had been set, the grids were searched and the optimal conditions were identified. Finally, the optimized method was subjected to validation. Figureᅟ

Chemometrically assissted optimization and validation of RP-HPLC method for the analysis of itraconazole and its impurities

This paper presents the chemometrically assisted optimization and validation of the RP-HPLC method intended for the quantitative analysis of itraconazole and its impurities in pharmaceutical dosage forms. To reach the desired chromatographic resolution with a limited number of experiments in a minimum amount of time, Box- -Behnken design was used to simultaneously optimize some important chromatographic parameters, such as the acetonitrile content in the mobile phase, pH of the aqueous phase and the column temperature. Separation between itraconazole and impurity F was identified as critical and selected as a response during the optimization. The set optimal mobile phase composition was acetonitrile/ water pH 2.5 adjusted with o-phosphoric acid (50:50, V/V). Separations were performed on a Zorbax Eclipse XDB-C18, 4.6 × 150 mm, 5 μm particle size column with the flow rate 1 mL min-1, column temperature set at 30 °C and UV detection at 256 nm. The established method was then subjected to method validation and the required validation parameters were tested. For the robustness evaluation, fractional factorial 24-1 design was utilized and factors that might significantly affect the system performance were defined. As other validation parameters were also found to be suitable, the possibility to apply the proposed method for the determination of itraconazole, its impurities B and F in any laboratory under different circumstances has been proven.

Optimization of the Separation of Ephedrine, Pseudoephedrine, Phenylephrine, and Synephrine by Hydrophilic Interaction Liquid Chromatography Employing Experimental Design Methodology

Hydrophilic interaction liquid chromatography (HILIC) has emerged as an alternative separation technique for small polar compounds. Design of Experiments (DoE) methodology in the analysis of the retention behavior and method development have been shown to be successful. This article addresses the retention behavior and optimization of chromatographic separation of a model mixture consisting of four sympathomimetic drugs (ephedrine, pseudoephedrine, phenylephrine, synephrine), on a bare silica column by HILIC. The acetonitrile content, pH of the aqueous phase, and concentration of ammonium acetate in aqueous phase were optimized according to the matrix of Box-Behnken design, and retention data were fitted to second-order polynomial models. The obtained coefficients of determination R2 were higher than 0.990 and adjusted R2 exceeded 0.972, while the lowest value for Predicted R2 was 0.845. Beside of statistical analysis, graphical estimation of retention behavior was performed. In order to optimize the separation, indirect modeling of selectivity factors of critical peak pairs (the diastereoisomers ephedrine/pseudoephedrine and the positional isomers phenylephrine/synephrine) and grid point search approach were employed. After the development of the grids, optimal chromatographic conditions were selected (acetonitrile content 91%, pH of the aqueous phase 4.5 and ammonium acetate concetration 35 m mol L−1), providing a total analysis time of 14 min.