Zur Dünnschicht-Chromatographie von zinnorganischen Verbindungen

With the aid of thinlayer chromatography mixtures of tetraorganotin compounds containing no polar groups can be separated and identified if they differ in the number of phenyl groups per molecule or in the nature of their aliphatic residues (alkyl or allyl). For analytical purposes mixtures of 80% hexane and 20% benzene by volume are suitable as mobile phase. For preparative purposes (requiring multiple development) pure hexane or hexane containing 10 to 20% benzene by volume are best suited as mobile phase, where the amount of hexane depends on the problem of separation at hand. The R_f values of the mobile phases vary linearly with the number of the phenyl groups x and decrease with increasing x. The straight lines connecting the R_f values cross at x=4 (tetraphenyltin). Dithizone and silver nitrate are suitable spray reagents for organotin compounds. By spraying the plates with dithizone tetra-, tri-, and di-organotin compounds can be immediately distinguished on the basis of the ensuing colourations. With silver nitrate allyltin compounds and polystannanes can be distinguished from alkyltin compounds; tetra-phenyltin is not coloured by this reagent.     

Penetration and Crosslinking of Macromolecules

One of the topical problems of macromolecular chemistry is the question of the extent to which macromolecules in solution or in the amorphous solid state can engage in mutual interpenetration. The present article is concerned primarily with recent investigations on cross-linking copolymerization and with inter-and intramolecular reactions of polymers as a contribution to the study of molecular shape.     

The dependence of low-energy electron attachment to CF3Br on electron and vibrational energy

In a joint experimental and theoretical effort, we have studied dissociative electron attachment (DEA) to the CF3Br molecule at electron energies below 2 eV. Using two variants of the laser photoelectron attachment method with a thermal gas target (T(G) = 300 K), we measured the energy dependent yield for Br- formation over the range E = 3-1200 meV with resolutions of about 3 meV (E < 200 meV) and 35 meV. At the onsets for excitation of one and two quanta for the C-Br stretching mode nu3, downward cusps are detected. With reference to the recommended thermal (300 K) attachment rate coefficient k(A)(CF3Br) = 1.4 x 10(-8) cm3 s(-1), absolute cross sections have been determined for Br- formation. In addition, we studied Br- and (CF3Br)Br- formations with a seeded supersonic target beam (10% CF3Br in helium carrier gas, with a stagnation pressure of 1-4 bars and nozzle temperatures of 300 and 600 K) and found prominent structure in the anion yields due to cluster formation. Using the microwave pulse radiolysis swarm technique, allowing for controlled variation of the electron temperature by microwave heating, we studied the dependence of the absolute DEA rate coefficient on the mean electron energy E over the range of 0.04-2 eV at gas temperatures T(G) ranging from 173 to 600 K. For comparison with the experimental results, semiempirical resonance R-matrix calculations have been carried out. The input for the theory includes the known energetic and structural parameters of the neutral molecule and its anion; the parameters of the resonant anion curves are chosen with reference to the known thermal rate coefficient for the DEA process. For the gas temperature T(G) = 300 K, good overall agreement of the theoretical DEA cross section with the experimental results is observed; moreover, rate coefficients for Br- formation due to Rydberg electron transfer, calculated with both the experimental and the theoretical DEA cross sections, are found to agree with the previously reported absolute experimental values. At T(G) = 300 K, satisfactory agreement is also found between the calculated and experimental attachment rate coefficients for mean electron energies E = 0.04-2 eV. The strong increase of the measured rate coefficients with rising gas temperature, however, could be only partially recovered by the R-matrix results. The differences may result from the influence of thermal excitations of other vibrational modes not included in the theory.     

Biooxidation of bridged cycloketones using Baeyer-Villiger monooxygenases of various bacterial origin

Bridged cycloketones were synthesized and utilized as substrates to study biooxidations mediated by Baeyer-Villiger monooxygenases (BVMO) of various bacterial origin. The required enzymes were heterologously produced by recombinant overexpression systems based on Escherichia coli to enable facile recycling of the required nicotinamide cofactors during the whole-cell biotransformations. Ketone precursors of various structural demands were chosen to evaluate steric limitations and flexibility of the active site of BVMOs. By desymmetrization of the prochiral substrates, four to six stereogenic centers were generated within a single biooxidation step. The enzyme library investigated in this study allowed access to antipodal lactone products with excellent enantioselectivity in several cases. Together with a distinct substrate acceptance profile, the recently proposed classification into two groups of cycloketone converting BVMOs was supported by the biotransformation results obtained within this study.     

A probabilistic approach to classifying metabolic stability

Metabolic stability is an important property of drug molecules that should-optimally-be taken into account early on in the drug design process. Along with numerous medium- or high-throughput assays being implemented in early drug discovery, a prediction tool for this property could be of high value. However, metabolic stability is inherently difficult to predict, and no commercial tools are available for this purpose. In this work, we present a machine learning approach to predicting metabolic stability that is tailored to compounds from the drug development process at Bayer Schering Pharma. For four different in vitro assays, we develop Bayesian classification models to predict the probability of a compound being metabolically stable. The chosen approach implicitly takes the "domain of applicability" into account. The developed models were validated on recent project data at Bayer Schering Pharma, showing that the predictions are highly accurate and the domain of applicability is estimated correctly. Furthermore, we evaluate the modeling method on a set of publicly available data.