Retention Index Data Research Articles

Artificial neural network prediction of quantitative structure: Retention relationships of polycyclic aromatic hydocarbons in gas chromatography

A feed-forward artificial neural network (ANN) model was used to link molecular structures (boiling points, connectivity indices and molecular weights) and retention indices of polycyclic aromatic hydrocarbons (PAHs) in linear temperature- programmed gas chromatography. A randomly taken subset of PAH retention data reported by Lee et al. [Anal. Chem. 51 (1979) 768], containing retention index data for 30 PAHs, was used to make the ANN model. The prediction ability of the trained ANN was tested on unseen data for 18 PAHs from the same article, as well as on the retention data for 7 PAHs experimentally obtained in this work. In addition, two different data sets with known retention indices taken from the literature were analyzed by the same ANN model. It has been shown that the relative accuracy as the degree of agreement between the measured and the predicted retention indices in all testing sets, for most of the studied PAHs, were within the experimental error margins (+-3 %).

VOLATILE AROMA COMPOUNDS IN MANGO FRUIT CV. ‘TOMMY ATKINS’ - A PRELIMINARY STUDY

Extraction of volatile compounds from the pulp of mango cultivar Tommy Atkins was undertaken through a simultaneous solvent distillation and extraction process using Likens and Nickerson's apparatus. Identification of volatile organic compounds was achieved in a system of high resolution gas chromatograph coupled with mass spectrometer. A total of 104 compounds were identified in the pulp of ripe fruit either positively when the mass spectrum and retention index data of the identified compound matched with that of the authentic standard run under identical conditions or tentatively when mass spectrum data and retention index values matched with the NIST (National Institute of Standards and Technology, USA) or other literature data-base. The results present the data on compounds along with their individual area representation in the chromatograms. The major classes of organic compounds identified in the flavor profile of Tommy Atkins fruit were hydrocarbons, esters, terpenes, lactones and aromatic compounds. The principal compounds were 3-carene, ethyl dodecanoate, hexyl hexanoate, methyl hexanoate, geranyl acetate, X-octalactone, X-nonalactone, citronellol, carvone, α-ionone, limonene, myrcene, β-phellandrene, terpineol (a and β), toluene, benzaldehyde and (Z)-3-hexen-1-ol.

Chemical Composition of the leaves extract of Warionia saharae of Morocco

Air-dried aerial parts of wild Warionia saharae were collected in Guelmim region (south of Morocco) during April 2003. The leaves only, were extracted with hexane in a Soxhlet apparatus for 8h. The obtained hexane extract was fractionated on the silica gel column and analyzed by GC and GC-MS using MS and retention index data. A total of 29 constituents were identified. The major components of the extract were hexadecanoic acid (17.8%), ethenyloxy-1-ocatdecane (9.5%), tridecene (7.3%), eicosene-9 (6.7%), octadecanoic acid (6.7%), (E)-2-decenol, (6.7%), eicosene-3 (5.1%) and eicosane (4.5%).

Methods for generating second dimension retention index data in comprehensive two-dimensional gas chromatography

Two methods of generating transportable second dimension retention data are outlined for comprehensive two-dimensional gas chromatography (GC×GC). They are both refinements of a previously outlined procedure, which adapted ‘isovolatile’ curves to retention prediction maps developed in our laboratory, extended to a more polar homologous series, the linear primary alcohols. The earlier work investigated maps based on alkanes and methyl ketones and methyl esters; here the method of data collection to generate the retention map has also changed, extending the retention base range. The resulting retention map permits a retention basis to be used for GC×GC data, however, a more polar retention set than alkanes is required. The calculation of ‘retention indices’ is aided by the generation of ‘fractional reference compound’ curves, by either direct manipulation of data local to the solute, or generation of a discrete curve coincident with the retention co-ordinates of the target compound.

Confirmation of the xanthone structure in thermally treated polycarbonates by reactive pyrolysis-gas chromatography/mass spectrometry

Reactive pyrolysis-gas chromatography/mass spectrometry in the presence of tetramethylammonium hydroxide was used to confirm the existence of xanthone structures in polycarbonate (PC) samples thermally treated at 300 and 400 °C. The predicted characteristic product ( M W=492) of the xanthone structures was successfully eluted on the resulting pyrogram by use of a deactivated stainless steel capillary column whose temperature was programmed up to 370 °C leading its positive assignment by its mass spectrum. Further, unequivocal identification of the specific product was carried out from the retention index data for the whole main degradation products observed on the pyrograms of the PC samples. Finally, the amounts of the xanthone structures formed in the PC samples are quantitatively discussed.

Use of artificial neural networks to predict the gas chromatographic retention index data of alkylbenzenes on carbowax-20M

Quantitative structure–activity relationships (QSARs) quantify the connection between the structure and properties of molecules and allow the prediction of properties from structural parameters. Models of relationships between structure and retention index of alkylbenzenes were constructed by means of a multilayer neural network using extended delta-bar-delta (EDBD) algorithms. The 165 group data belong to 129 alkylbenzenes at different temperatures on carbowax-20M. We proposed a new method to describe the structure of the alkylbenzene with a simple set of six numeric code depending on its molecular formula. A set of six numbers and the temperature were used as input parameters to predict the retention indices. The performance of different order of structural coding was investigated. The networks’ architecture and the learning times were optimized. The optimum ANNs could give excellent prediction results. In addition, the multiple linear regression (MLR) and nonlinear multivariate regression were applied. We have shown in our studies that, based on the structural numeric codes, ANNs give more accurate predictions of retention index data of alkykbenzenes than regression analysis.

Gibberellin Biosynthesis in Maize. Metabolic Studies with GA(15), GA(24), GA(25), GA(7), and 2,3-Dehydro-GA(9).

[17-(14)C]-Labeled GA(15), GA(24), GA(25), GA(7), and 2,3-dehydro-GA(9) were separately injected into normal, dwarf-1 (d1), and dwarf-5 (d5) seedlings of maize (Zea mays L.). Purified radioactive metabolites from the plant tissues were identified by full-scan gas chromatography-mass spectrometry and Kovats retention index data. The metabolites from GA(15) were GA(44), GA(19), GA(20), GA(113), and GA(15)-15,16-ene (artifact?). GA(24) was metabolized to GA(19), GA(20), and GA(17). The metabolites from GA(25) were GA(17), GA(25) 16alpha,17-H(2)-17-OH, and HO-GA(25) (hydroxyl position not determined). GA(7) was metabolized to GA(30), GA(3), isoGA(3) (artifact?), and trace amounts of GA(7)-diene-diacid (artifact?). 2,3-Dehydro-GA(9) was metabolized to GA(5), GA(7) (trace amounts), 2,3-dehydro-GA(10) (artifact?), GA(31), and GA(62). Our results provide additional in vivo evidence of a metabolic grid in maize (i.e. pathway convergence). The grid connects members of a putative, non-early 3,13-hydroxylation branch pathway to the corresponding members of the previously documented early 13-hydroxylation branch pathway. The inability to detect the sequence GA(12) --> GA(15) --> GA(24) --> GA(9) indicates that the non-early 3,13-hydroxylation pathway probably plays a minor role in the origin of bioactive gibberellins in maize.

Composition of Tanacetum annuum L. Oil from Morocco

Fresh aerial parts of Tanacetum annuum L. from Morocco were steam distilled to produce an oil in 0.5% yield. The oil was fractionated on the silica gel column and analyzed by GC and GC/MS using MS and retention index data and more than 130 constituents were identified. The major components of the oil were chamazulene (38–17%), myrcene (14–1%), sabinene (8.6–4%), β-eudesmol (7–3%) and camphor (18–4%). Seasonal variation of the oil composition suggested that the chamazulene content decreases and the myrcene content increases from June to October.

Gas chromatographic and mass spectrometric analysis of 36 lignin-related methoxyphenols from uncontrolled combustion of wood

Eighteen 4-hydroxy-3-methoxyphenyl (guaiacyl) compounds in wood smoke from spruce, and 18 4-hydroxy-3,5-dimethoxyphenyl (syringyl) analogues in birchwood smoke were analysed. Smoke samples were taken by a gas syringe from wood burning under varying but more or less choked conditions in laboratory pot experiments. Retention index data for a methylsilicone column and basic mass spectral data are given for all compounds. Retention index in combination with molecular mass and structure-specific fragment ions greatly facilitates qualitative analysis. The syringyl and guaiacyl analogues differ by 240 units in retention index and the matching structure-specific ions in their mass spectra by 30 mass units. The identified methoxyphenols differ in the structure of the sidechain, which may be lacking, an alkyl group, an alkenyl group, a carbonyl-containing group or an alcoholic moiety. The proportions between these groups of compounds varied strongly with combustion conditions whereas the proportions of phenols within them were quite specific for each group. The phenols with reactive olefinic linkages in the sidechain were present in larger proportions than reported in other studies of methoxyphenols.

The Metabolism of Gibberellin A20 to Gibberellin A1 by Tall and Dwarf Mutants of Oryza sativa and Arabidopsis thaliana.

The purpose of this study was to demonstrate the metabolism of gibberellin A20 (GA20) to gibberellin A1 (GA1) by tall and mutant shoots of rice (Oryza sativa L.) and Arabidopsis thaliana (L.) Heynh. The data show that the tall and dx mutant of rice and the tall and ga5 mutant of Arabidopsis metabolize GA20 to GA1. The data also show that the dy mutant of rice and the ga4 mutant of Arabidopsis block the metabolism of GA20 to GA1. [17-13C,3H]GA20 was fed to tall and the dwarf mutants, dx and dy, of rice and tall and the dwarf mutants, ga5 and ga4, of Arabidopsis. The metabolites were analyzed by high-performance liquid chromatography and full-scan gas chromatography-mass spectrometry together with Kovats retention index data. For rice, the metabolite [13C]GA, was identified from tall and dx seedlings; [13C]GA1 was not identified from the dy seedlings. [13C]GA29 was identified from tall, dx, and dy seedlings. For Arabidopsis, the metabolite [13C]GA1 was identified from tall, ga5, and ga4 plants. The amount of [13C]GA1 from ga4 plants was less than 15% of that obtained from tall and ga5 plants. [13C]GA29 was identified from tall, ga5, and ga4 plants. [13C]GA5 and [13C]GA3 were not identified from any of the six types of plant material.

Study of calcium soap pyrolysates derived from Macauba fruit ( Acrocomia sclerocarpa M.). Derivatization and analysis by GC/MS and CI-MS

An investigation has been undertaken of pyrolysates obtained from calcium soaps prepared from vegetable oils in Macauba fruit ( Acrocomia sclerocarpa M.). GC/MS showed the samples to be very complex, exhibiting more than 40 compounds. To attempt to separate and identify the constituents different techniques were used: methoximation, EI and ammonia CI-MS. n-Alkane and 1-alkane series were located and identified by the characteristic ions m/ z 85 and 83 respectively, and the molecular ions in the case of the lower members of the series. Identification of homologous series was further confirmed by a mass spectral search (NIST), visual comparison with published mass spectra and use of linear retention index data. To target the ketones and aldehydes, methoxime derivatization was employed. With EI-MS the methoximes gave a specific ion at m/ z 87, but no molecular ion. Ammonia CI-MS was used to confirm the molecular weight of both underivatized ketones and methoximes in correlation with the retention data.

List Operations on Chemical Graphs. 5. Implementation of Breadth-First Molecular Path Generation and Application in the Estimation of Retention Index Data and Boiling Points

List Operations on Chemical Graphs. 5. Implementation of Breadth-First Molecular Path Generation and Application in the Estimation of Retention Index Data and Boiling Points

Analysis of Volatile Components in Fresh Grapefruit Juice by Purge and Trap/Gas Chromatography

Qualitative and quantitative analyses of volatile flavor components in fresh grapefruit juice were performed by purge and tap/gas chromatography/mass spectrometry (P&T/GC/MS) and P&T/GC with flame-ionization detection (FID). Seventeen compounds were positively identified. Four additional compounds were tentatively identified on the basis of either MS or retention index (RI) data. Quantification of positively identified components was done by P&T/GC/FID using butyl acetate as internal standard and calibration curves. Compounds not previously identified in grapefruit juice were methyl acetate, propyl acetate, and 2-methylethyl propionate (tentatively identified)

Prediction of gas chromatographic retention index data by neural networks

Neural networks using the backpropagation algorithm can be applied to quantitative structure-physical property relationship studies. Neural networks can be trained with electrotopological indexes on monofunctional compounds to predict the corresponding retention index data. These networks can also be applied to the prediction of retention index data of acyclic and cyclic monoterpenes and a mixed set of monosubstituted and terpene compounds. Predictions by neural networks are generally in good agreement with predictions done by multilinear regression techniques. In the case of predicting retention index data of compounds from a class not represented in the training data, neural networks show strong deficiencies in comparison with multilinear regression methods.

Synthesis and spectroscopic properties of diarylpolyselanes R 2Se x (x = 2–7) and of diaryldiselenapolysulphanes (RSe) 2Sy (y = 1–15) with R = 4-Cl-C 6H 4

Reaction of RSeCl (R = 4-Cl-C 6H 4) with titanocene pentasulphide Cp 2TiS 5 affords the novel RSeS 5SeR as a yellow oil which is quite stable at 20°C. Using Cp 4′Ti 2S 4 (Cp′ = CH 3C 5H 4) as a sulphur transfer reagent the reaction with RSeCl results in a mixture of (RSe) 2S n molecules with n = 0–10, which on heating produces even longer chains as demonstrated by HPLC analysis. Titanocene pentaselenide Cp 2TiSe 5 reacts with RSeCl at 4°C to give a mixture of R 2Se x molecules (x = 1–7) and Se 8, the relative concentrations sharply decrease with increasing x. 77Se NMR and UV spectra as well as retention index data of the new molecules are reported.

List operations on chemical graphs. 4. Using edge models for prediction of retention index data

List operations on chemical graphs. 4. Using edge models for prediction of retention index data

List operations on chemical graphs. 3. Development of vertex and edge models for fitting retention index data

List operations on chemical graphs. 3. Development of vertex and edge models for fitting retention index data

Calculation of retention indices by molecular topology: III. Chlorinated dibenzodioxins

A model is developed, based entirely on chemical structure, for the accurate prediction of the retention of chlorinated dibenzodioxins (PCDDs) on a non-polar stationary phase (DB-5) and for reproducing their experimental elution sequence. The first-order molecular connectivity inex ( 1χ) and six other topological properties, the specific mono-, di- and trichloro substitution patterns, were used as structural descriptors. Thus, one can predict the retention index for any given PCDD having only a knowledge of its chlorine substitution pattern. The agreement between observed and calculated retention indices (the average residual of 1.8 units) indicates that this model works fairly well to describe the retention indices of all analysed PCDDs (37 tetra- to octachlorinated congeners). The developed model was successfully validated by extrapolation on a small set of retention indices determined for di- and trichlorinated PCDD congeners and by interpolation on a set of sixteen retention indices determined for tetra- to octachlorinated PCDDs on a low-polarity HP-5 capillary column. The regression analysis shows that the PCDD retention indices are primarily influenced by their bulk properties, i.e., the size of these molecules. This property, described best by the 1χ index, accounts for more than 97% of the variation in the retention index data. The other factors that control the magnitude of the retention indices are various chlorine substitution patterns, having either a positive or a negative effect on the magnitude of the retention indices. Finally, a mechanism for the interaction of PCDDs with the non-polar stationary phase is also discussed.

Computer assisted determination of the biological activity of polychlorinated biphenyls using gas chromatographic retention indices as molecular descriptors

Gas chromatographic retention indices of polychlorinated biphenyls (PCBs) on several different chromatographic columns are used as molecular descriptors to predict the toxicity of PCBs. Principal component and discriminant analysis are performed on the retention index data to classify a learning set of 27 PCB molecules of known toxicity. It was found that 25 of 27 PCBs were correctly classified using this method. The errors in classification are presumed to be a result of the varying degrees of toxicity of the PCB molecules. Further calculations using a set of 114 PCB congeners gave similar results. The predictions followed qualitative structure-activity relationships known for PCBs.

Experiences with symbolic programming in the automated estimation of retention index data

Gautzsch, R., Zinn, P. and Haffer, Chr.M., 1991. Experiences with symbolic programming in the automated estimation of retention index data. Chemometrics and Intelligent laboratory Systems, 12: 81–90. A method for the automatic pre-calculation of gas chromatographic retention index data is presented. The method is based on a summation of increments related to atoms and bonds which was introduced by Takács. Symbolic programming allows flexible pre-calculation for a wide range of compounds. As examples retention index data for alkanes, cycloalkylanes and alcohols are presented. In order to handle the chemical information of complex compounds comfortably, a LISP coding system for molecular information was implemented. On the basis of this LISP notation several procedures of general interest could be developed, e.g. the derivation of bond and atom lists. A nested system of recursive procedures results which is easy to extend to a wide range of applications.