Pavel Májek

Use of a polar ionic liquid as second column for the comprehensive two-dimensional GC separation of PCBs.

The orthogonality of three columns coupled in two series was studied for the congener specific comprehensive two-dimensional GC separation of polychlorinated biphenyls (PCBs). A non-polar capillary column coated with poly(5%-phenyl-95%-methyl)siloxane was used as the first ((1)D) column in both series. A polar capillary column coated with 70% cyanopropyl-polysilphenylene-siloxane or a capillary column coated with the ionic liquid 1,12-di(tripropylphosphonium)dodecane bis(trifluoromethane-sulfonyl)imide were used as the second ((2)D) columns. Nine multi-congener standard PCB solutions containing subsets of all native 209 PCBs, a mixture of 209 PCBs as well as Aroclor 1242 and 1260 formulations were used to study the orthogonality of both column series. Retention times of the corresponding PCB congeners on (1)D and (2)D columns were used to construct retention time dependences (apex plots) for assessing orthogonality of both columns coupled in series. For a visual assessment of the peak density of PCBs congeners on a retention plane, 2D images were compared. The degree of orthogonality of both column series was, along the visual assessment of distribution of PCBs on the retention plane, evaluated also by Pearsons correlation coefficient, which was found by correlation of retention times t(R,i,2D) and t(R,i,1D) of corresponding PCB congeners on both column series. It was demonstrated that the apolar+ionic liquid column series is almost orthogonal both for the 2D separation of PCBs present in Aroclor 1242 and 1260 formulations as well as for the separation of all of 209 PCBs. All toxic, dioxin-like PCBs, with the exception of PCB 118 that overlaps with PCB 106, were resolved by the apolar/ionic liquid series while on the apolar/polar column series three toxic PCBs overlapped (105+127, 81+148 and 118+106).

Fluorescence spectroscopy and multivariate methods for the determination of brandy adulteration with mixed wine spirit

The addition of mixed wine spirit to brandy is easy way to adulterate brandy. To avoid the misleading of the consumers, it is necessary to develop reliable method to detect adulteration of brandy. In this work excitation emission matrix fluorescence in combination with parallel factor analysis (PARAFAC) and partial least squares (PLS) regression was used to determine the content of mixed wine spirit in adulterated brandy samples. Excitation emission matrix fluorescence spectra were measured in the emission wavelength range of 485-580 nm and in the excitation wavelength range of 363-475 nm. The model created using PARAFAC-PLS was able to predict the mixed wine spirit level in adulterated brandy with the root mean square error of prediction (RMSEP) value of 1.9% and a square of the correlation (R(2)) between the reference contents and the predicted values of 0.995.

Simultaneous determination of caffeine, caramel and riboflavin in cola-type and energy drinks by synchronous fluorescence technique coupled with partial least squares

The aim of this work was to develop a multivariate method for the rapid determination of caffeine and Class IV caramel in cola-type soft drinks and of caffeine, Class III caramel and riboflavin in energy drinks using synchronous fluorescence spectra. The synchronous fluorescence spectra were recorded at constant wavelength difference 90 nm from 200 to 500 nm. Reference values of analyte concentrations by high performance liquid chromatography (HPLC) with fluorescence detection combined with the standard addition method were used to create the partial least squares (PLS) models. High coefficients of determination (>0.99) were obtained in 0.2-4.2, 0.25-5.25, 0.4-10.0 and 0.007-0.054 mg L(-1) range for caffeine, Class III caramel, Class IV caramel and riboflavin, respectively. The PLS models were used to determine the concentration of analytes in different drink samples. The method provided comparable results with those found using the HPLC method.

Classification of plum spirit drinks by synchronous fluorescence spectroscopy.

Synchronous fluorescence spectroscopy was used in combination with principal component analysis (PCA) and linear discriminant analysis (LDA) for the differentiation of plum spirits according to their geographical origin. A total of 14 Czech, 12 Hungarian and 18 Slovak plum spirit samples were used. The samples were divided in two categories: colorless (22 samples) and colored (22 samples). Synchronous fluorescence spectra (SFS) obtained at a wavelength difference of 60 nm provided the best results. Considering the PCA-LDA applied to the SFS of all samples, Czech, Hungarian and Slovak colorless samples were properly classified in both the calibration and prediction sets. 100% of correct classification was also obtained for Czech and Hungarian colored samples. However, one group of Slovak colored samples was classified as belonging to the Hungarian group in the calibration set. Thus, the total correct classifications obtained were 94% and 100% for the calibration and prediction steps, respectively. The results were compared with those obtained using near-infrared (NIR) spectroscopy. Applying PCA-LDA to NIR spectra (5500-6000 cm(-1)), the total correct classifications were 91% and 92% for the calibration and prediction steps, respectively, which were slightly lower than those obtained using SFS.

On the determination of a detector response enhancement factor for flow modulated comprehensive two-dimensional gas chromatography.

Comparison of conventional one-dimensional gas chromatography (1D-GC) and comprehensive two-dimensional gas chromatography (GC×GC) shows that the detector signal in GC×GC is significantly enhanced. The value of this enhancement depends on the operating parameters, such as the modulation time and the non-modulated ((1)D column) and corresponding modulated ((2)D column) peak widths. Determination of the enhance factors in flow modulated GC×GC for separation of hydrocarbon samples using flame ionization detection (FM GC×GC-FID) was studied. A term named detector response enhance factor (DREF) has been designed to express the value of the signal enhancement. Two methods were used to calculate DREF: (i) employing (2)D peak height, and (ii) using a theoretical approach. Comparison of FM GC×GC-FID and 1D-GC-FID method illustrates that the DREF increases from 10 to 33 times, in 1-6s modulation time intervals.

On retentivity tuning by flow in the second column of different comprehensive two dimensional gas chromatographic configurations

Retentivity tuning in comprehensive two dimensional GC separations of aliphatics (linear and cyclic hydrocarbons) and aromatics in gasoline by changing the carrier gas flows in the column series at constant working temperature parameters of both columns is discussed. Comprehensive 2D techniques studied include GC×GC with cryogenic and differential flow modulation and non-modulated transfer (NMT). In all configurations, the first dimension was a non-polar column and the second dimension a polar column. Using three different flows (0.6, 1.0 and 1.4mL/min) of helium carrier gas in cryogenic modulated GC×GC illustrated that, as expected, retention of the solutes on the (1)D and (2)D columns increased but the separation quality was nearly constant. A change of carrier gas pressure (p(m)=175-125kPa) on the (1)D and (2)D columns joint point at constant inlet pressure (p(i)=525kPa) in NMT, induces an increase of the carrier gas flow rate on the (1)D and a decrease on the (2)D column, respectively. The higher retentivity of the (2)D column improved the group type separation of aliphatic/cyclic hydrocarbons and aromatics and a higher distribution of aromatics on the 2D retention plane was noted. Retentivity tuning was also performed in flow modulated GC×GC by operating the (1)D column at 0.8mL/min and the (2)D column at 20 and 26mL/min. The higher retentivity at 20mL/min improved the group type separation of aliphatic/cyclic hydrocarbons and aromatics in the 2D retention plane.

Considerations on the determination of the limit of detection and the limit of quantification in one-dimensional and comprehensive two-dimensional gas chromatography

Methods based on the blank signal as proposed by IUPAC procedure and on the signal to noise ratio (S/N) as listed in the ISO-11843-1 norm for determination of the limit of detection (LOD) and quantitation (LOQ) in one-dimensional capillary gas chromatography (1D-GC) and comprehensive two-dimensional capillary gas chromatography (CG×GC) are described in detail and compared for both techniques. Flame ionization detection was applied and variables were the data acquisition frequency and, for CG×GC, also the modulation time. It has been stated that LOD and LOQ estimated according to IUPAC might be successfully used for 1D-GC-FID method. Moreover, LOD and LOQ decrease with decrease of data acquisition frequency (DAF). For GC×GC-FID, estimation of LOD by IUPAC gave poor reproducibility of results while for LOQ reproducibility was acceptable (within ±10% rel.). The LOD and LOQ determined by the S/N concept both for 1D-GC-FID and GC×GC-FID methods are ca. three times higher than those values estimated by the standard deviation of the blank. Since the distribution pattern of modulated peaks for any analyte separated by GC×GC is random and cannot be predicted, LOQ and LOD may vary within 30% for 3s modulation time. Concerning sensitivity, 1D-GC-FID at 2Hz and of GC×GC-FID at 50Hz shows a ca. 5 times enhancement of sensitivity in the modulated signal output.

Determination of the adulterants in adulterant–brandy blends using fluorescence spectroscopy and multivariate methods

The addition of water or ethanol to brandy is an easy way to adulterate brandy. To avoid the misleading of consumers, it is necessary to develop a reliable method to detect the adulteration of brandy. In this work excitation–emission matrix fluorescence in combination with parallel factor analysis (PARAFAC) and partial least squares (PLS) regression was used to determine the content of water, ethanol and methanol in adulterated brandy samples. Excitation–emission matrix fluorescence spectra were measured in the emission wavelength range of 510–600 nm and in the excitation wavelength range of 393–497 nm. The model created using PARAFAC-PLS was able to predict the water, ethanol and methanol level in adulterated brandy with root mean square error of prediction values of 0.24%, 0.20% and 0.22%,respectively, and coefficients of determination of prediction between the reference content and the predicted values of 0.993, 0.997 and 0.995, respectively.

The non-modulated transfer of total effluent for serially coupled columns in gas chromatography

The non-modulated transfer (NMT) of the total effluent of the separation of polychlorinated biphenyls (PCBs) on two columns coupled in series has been studied. A non-polar poly(5%-phenyl-95%-methyl)siloxane column (40 m x 100 microm x 0.1 microm) and a polar 70%-cyanopropyl-polysilphenylene-siloxane column (4 m x 0.1 mm x 0.1 microm) were used as (1)D and (2)D columns, respectively. The effluents from both the (1)D column and the (1)D+(2)D column series were monitored independently by two FIDs. Data from the (1)D and (1)D+(2)D integration reports were used to evaluate the NMT experiment. (1)D retention times, t(R,i,D1), were directly accessible from (1)D integration report while (2)D retention times, t(R,i,B), were calculated for all corresponding peak pairs from (1)D and (1)D+(2)D integration reports as a difference t(R,i,D2)=t(R,i,D1+D2)-t(R,i,D1). Search for corresponding peaks of PCB congeners in the (1)D and (1)D+(2)D chromatograms is elucidated in detail on standard PCB samples and on PCB congeners present in the technical formulation Arochlor 1242. Both retention times (t(R,i,D1) and t(R,i,D2)) as well as peak widths at half height (w(h,i)) and peak heights (h(i)) obtained from integration reports were used to construct 2D and 3D images for PCB NMT separations on serially coupled columns. The performance of the NMT procedure is illustrated by the separation of (i) standard PCB solutions, (ii) a mixture of the 209 PCBs, and (iii) a mixture of Arochlor 1242 and hydrocarbons on the DB-5+BPX-70 column series.

Computerized optimization of flows and temperature gradient in flow modulated comprehensive two-dimensional gas chromatography

Informational entropy and syentropy percent were used to optimize the flows in the first (1D) and in the second (2D) dimension ((1)Fm and (2)Fm, respectively) as well as the temperature program rate (r) for the flow modulated GC×GC-FID separation of C6-C12 aromatic hydrocarbons in a low boiling petrochemical sample. The separations were performed on a column series consisting of a 25m×0.25mm i.d.×0.2μm df of the polar ionic liquid SLB-IL 100 (1,9-di(3-vinylimidazolium)nonane bis(trifluoromethylsulfonyl)imide) in the first dimension and 5m×0.25mm i.d.×0.25μm df apolar HP-5MS (5% phenyl-95% methylpolysiloxane) in the second dimension. A dependence of a distribution of individual aromatic hydrocarbons in the 2D retention plane on the carrier gas flows ((1)Fm, and (2)Fm,) and temperature gradient (r) was examined in this study. It was found that informational entropy and synentropy percent are advantageous criteria to characterize the distribution of peaks in the 2D retention plane. Maximum informational entropy and synentropy percents correspond to the maximum distribution of C6-C12 aromatic hydrocarbons in the corresponding 2D retention plane gained by the given separation using optimized values of individual carrier gas column volume flows and the temperature rate at the temperature programmed GC×GC separations.