Meßtechnik zur Charakterisierung von Gas/Flüssigkeits-Reaktoren

Measuring techniques for characterizing gas/liquid reactors . The article considers measuring techniques for the characterization of the phases in gas/liquid reactors. These include the velocity profile and turbulence in the continuous phase as well as holdup, size, and velocity of the dispersed phase. Also, methods for the determination of the specific interfacial area and the volumetric mass transfer coefficients are discussed. Measuring methods have made enormous advances in recent years. In spite of this, there is a lack of reliable methods, especially for use during reactions without disturbing the state of the reactor. It is suspected that probe technique will gradually fall into disuse and more and more attention will be focussed on contractless methods. This calls for further development of ultrasonic methods and methods relying on emission of light.     

Verbesserung des Vakuums in Isoliergefäßen aus Glas

Improving the vacuum would increase the insulation capacity of commercially available glass containers by a factor of about two. To do this, pressures of 10^?4 mbar must be attained. Different approaches were taken to meet this objective. As an alternative to today's commonly practiced sealing method, three possibilities are introduced which eliminate the necessity of pump stub narrowing which causes reduced conductance. Furthermore, depending on the sealing process used, the gases occuring during sealing are either avoided completly or they can at least be reduced by having the chance of exact temperature control resp. by a lower separation temperature. A completly different approach was taken by using Getters, of which a metal hydride on a TiVMn base proved especially suitable. It is activated by heat and can be integrated into the normal production process without any problem. Aside from a shorter pump period, the added advantage here is the containment of the gasses occuring during sealing as well as that of the longterm gasses.     

Selbstbestimmung

Ohne Zusammenfassung     

Randomized function evaluation on a ring

LetR be a unidirectional asynchronous ring ofn identical processors each with a single input bit. Letf be any cyclic nonconstant function ofn boolean variables. Moran and Warmuth (1986) prove that anydeterministic algorithm that evaluatesf onR has communication complexity (n logn) bits. They also construct a family of cyclic nonconstant boolean functions that can be evaluated inO(n logn) bits by a deterministic algorithm.This contrasts with the following new results:

Darstellung,Struktur und Reaktivität quartärer Ammoniumcellulosate

Tetraalkylammonium cellulosates were prepared by means of an exchange reaction between cellulose and tetraalkylammonium methoxides in anhydrous methanol and dimethylsulfoxide up to a degree of substitution of about 0,7 without chain degradation. The course of this reaction and the structure of the cellulosates in comparison to lithium and sodium cellulosate has been studied. The reaction rate and the equilibrium substitution decreased with increasing size of the cation in the methoxide. The distribution of the alcoholate groups in equilibrium at C-2, C-3 and C-6 of the glucopyranosyl unit in the cellulosates, which was investigated by methylation, hydrolysis of the methylcelluloses, and analysis of the methylglucoses by gas chromatography, was independent of the nature of the cation. Under the given reaction conditions the hydroxyl group at C-2 was more acidic than those at C-3 and C-6, while for the latter practically no difference was found. The tetraalkylammonium cellulosates are highly reactive intermediates for the preparation of cellulose derivatives. The reactivity of the cellulosates increased with increasing size of the cation, Li^⊕ < Na^⊕ < N(CH₃)₄^⊕ <[(CH₃)₃N? CH₂? C₆H₅]^⊕.     

Anwendung von statistischen Prüfverfahren im Bereich der Mischtechnik

Application of statistical testing to mixing. Assessment of the homogeneity of mixtures of solids and determination of the necessary mixing times are performed with the aid of statistical testing methods. The results from the basis for necessary decisions. Since reliable statements are fundamentally impossible in mathematical statistics, all such decisions are, with a certain probability, false, i.e. involve some risk. In studies of mixing, the first kind of error concerns the risk of rejecting a mixture although its homogeneity satisfies requirements, while the second kind of error concerns the risk of accepting a mixture although its homogeneity is insufficient. It is shown that barriers can be introduced for both kinds of error, and how to do this. Both the sample size and the limits of the acceptance range for the sample variance are thus laid down. If no decision is to be made concerning acceptance or rejection of a given mixture, but information is merely required about the mixing quality after a given time then this is accomplished by statement of confidence intervals.