On liquid motion in a circular cylinder with horizontal axis

Summary The flow pattern of the slosh motion of a homogeneous, nonviscous (inviscid) and incompressible fluid with a free surface, contained in a rigid circular canal, has been dealt with analytically and experimentally. The axis of the canal is perpendicular to the direction of gravity. Considered are transverse (lateral to axis) oscillations of the liquid. The shape of the free liquid surface is determined numerically by means of a simple procedure. Theoretically calculated streamlines (path lines) are found to be in good qualitative agreement with experimentally observed trajectories of small spheres, made of plastic material and immersed in the liquid. The plastic particles and the liquid (water with a solution of salt added) have the same density.     

De nouveaux stannures ternaires de rhodium et d'elements des terres rares: TR1+xRh2Sn4−x, 0 ⩽ x ⩽ 0,5, TR = La - SmNew ternary stannides: TR1+xRh2Sn4−x, 0≦x≦0.5, TR = La - Sm

New ternary stannides are reported: RE_1+xRh₂Sn_4?x, 0 ? x ? 0,5 (RE = La ? Sm). NdRh₂Sn₄ has been characterized by single - crystal X-ray diffraction analysis. Its structure is of a new type with space group Pnma and Z = 4: a = 18,535(3), b = 4,463(1), $\text{c}\text{= 7,229(1)}\text{A}\text{?}$, D_x = 9,16 g.cm^?1, μ(AgKα) = 13,6 mm^?1, F(000) = 1300, R = 0,040 for 496 independent reflexions (R_w = 0,047). One of the tin sites can be partly occupied by rare earths atoms. The structure types of NdRh₂Sn₄ and Y₂Rh₃Sn₅ are closely related. LaRh₂Sn₄ does not exhibit superconductivity above 1.4 K.     

Si/SiGe n-MODFETs on thin SiGe virtual substrates prepared by means of He implantation

Si/SiGe n-type modulation-doped field-effect transistors grown on a very thin strain-relieved Si/sub 0.69/Ge/sub 0.31/ buffer on top of a Si(100) substrate were fabricated and characterized. This novel type of virtual substrate has been created by means of a high dose He ion implantation localized beneath a 95-nm-thick pseudomorphic SiGe layer on Si followed by a strain relaxing annealing step at 850/spl deg/C. The layers were grown by molecular beam epitaxy. Electron mobilities of 1415 cm/sup 2//Vs and 5270 cm/sup 2//Vs were measured at room temperature and 77 K, respectively, at a sheet carrier density of about 3/spl times/10/sup 12//cm/sup 2/. The fabricated transistors with Pt-Schottky gates showed good dc characteristics with a drain current of 330 mA/mm and a transconductance of 200 mS/mm. Cutoff frequencies of f/sub t/=49 GHz and f/sub max/=95 GHz at 100 nm gate length were obtained which are quite close to the figures of merit of a control sample grown on a conventional, thick Si/sub 0.7/Ge/sub 0.3/ buffer.     

Calorimetric determination of the enthalpy change for the binding of methyl orange to poly(vinylpyrrolidone) in aqueous solution

Thermodynamic parameters have been determined for the interaction of methyl orange and poly(vinylpyrrolidone) in aqueous solutions containing tris, tris and hydrochloric acid, or water only. Enthalpy changes were determined by use of a flow microcalorimeter. Free energy changes were calculated from the results of equilbrium dialysis studies. The results are interpreted in terms of the behaviour of poly(vinylpyrrolidone) in these systems. Analysis of data shows that the hydrophobic interactions in the systems are in the order: tris > wateronly > tris-HC1.     

Differential competitive learning for centroid estimation and phoneme recognition.

A comparison is made of a differential-competitive-learning (DCL) system with two supervised competitive-learning (SCL) systems for centroid estimation and for phoneme recognition. DCL provides a form of unsupervised adaptive vector quantization. Standard stochastic competitive-learning systems learn only if neurons win a competition for activation induced by randomly sampled patterns. DCL systems learn only if the competing neurons change their competitive signal. Signal-velocity information provides unsupervised local reinforcement during learning. The sign of the neuronal signal derivative rewards winners and punishes losers. Standard competitive learning ignores instantaneous win-rate information. Synaptic fan-in vectors adaptively quantize the randomly sampled pattern space into nearest-neighbor decision classes. More generally, the synaptic-vector distribution estimates the unknown sampled probability density function p( x). Simulations showed that unsupervised DCL-trained synaptic vectors converged to class centroids at least as fast as, and wandered less about these centroids than, SCL-trained synaptic vectors did. Simulations on a small set of English phonemes favored DCL over SCL for classification accuracy.     

A new technique for on-line and off-line high speed computation

A novel technique for both online and offline computation is presented. With this technique, a reconstruction analysis in elementary particle physics, otherwise prohibitively long, has been accomplished. It will be used online in an upcoming Fermilab experiment to reconstruct more than 100000 events per second and to trigger on the basis of that information. The technique delivers 40 gigaoperations per second, has a bandwidth on the order of gigabytes per second, and has a modest cost. An overview of the program, details of the system, and performance measurements are presented