Crystallographic defects in thermally oxidized wafer bonded silicon on insulator (SOI) substrates

The crystalline quality of wafer bonded (WB) silicon on insulator (SOI) structures thermal treated in dry oxygen ambients has been investigated by means of transmission electron microscopy and defect etching. The main crystallographic defects present in the SOI layers are dislocations, amorphous precipitates, and oxidation induced stacking faults (OISF). The evolution of the OISFs with time and temperature has also been investigated. The main feature observed is that the OISF in WB SOI structures undergo a retrogrowth process at temperatures around T = 1195°C for times of t = 2h. This result is very similar to that recently reported for oxygen implanted SOI (SIMOX) but considerably different from that observed in bulk silicon. The experimental data fits nicely a model recently proposed for the retrogrowth of OISF in thin SOI layers. This model considers that the self-interstitial supersaturation is considerably reduced compared to bulk silicon due to the relative fast point defect recombination inside the top silicon layer.     

Stability analysis of numerical interface conditions in fluid–structure thermal analysis

This paper analyses the numerical stability of coupling procedures in modelling the thermal diffusion in a solid and a fluid with continuity of temperature and heat flux at the interface. A simple one-dimensional model is employed with uniform material properties and grid density in each domain. A number of different explicit and implicit algorithms are considered for both the interior equations and the boundary conditions. The analysis shows that in general these are stable provided that Dirichlet boundary conditions are imposed on the fluid and Neumann boundary conditions are imposed on the solid; in each case the imposed values are obtained from the other domains. © 1997 John Wiley & Sons, Ltd.     

An Entropy Metric for Regular Grammar Classification and Learning with Recurrent Neural Networks

Recently, there has been a resurgence of formal language theory in deep learning research. However, most research focused on the more practical problems of attempting to represent symbolic knowledge by machine learning. In contrast, there has been limited research on exploring the fundamental connection between them. To obtain a better understanding of the internal structures of regular grammars and their corresponding complexity, we focus on categorizing regular grammars by using both theoretical analysis and empirical evidence. Specifically, motivated by the concentric ring representation, we relaxed the original order information and introduced an entropy metric for describing the complexity of different regular grammars. Based on the entropy metric, we categorized regular grammars into three disjoint subclasses: the polynomial, exponential and proportional classes. In addition, several classification theorems are provided for different representations of regular grammars. Our analysis was validated by examining the process of learning grammars with multiple recurrent neural networks. Our results show that as expected more complex grammars are generally more difficult to learn.     

Random Bit Quadrature and Approximation of Distributions on Hilbert Spaces

Foundations of Computational Mathematics - We study the approximation of expectations $${text {E}}(f(X))$$ for Gaussian random elements X with values in a separable Hilbert space H and Lipschitz...     

Counterexamples in calculus of variations in L∞ through the vectorial Eikonal equation

We show that, for any regular bounded domain $\mathrm{\Omega }?{\mathbb{R}}^n$, $n=2,3$, there exist infinitely many global diffeomorphisms equal to the identity on ?Ω that solve the Eikonal equation. We also provide explicit examples of such maps on annular domains. This implies that the ∞-Laplace system arising in vectorial calculus of variations in $L^{\infty }$ does not suffice to characterise either limits of p-Harmonic maps as $p\to \infty$ or absolute minimisers in the sense of Aronsson.     

On the utility of graphics cards to perform massively parallel simulation of advanced Monte Carlo methods

We present a case-study on the utility of graphics cards to perform massively parallel simulation of advanced Monte Carlo methods. Graphics cards, containing multiple Graphics Processing Units (GPUs), are self-contained parallel computational devices that can be housed in conventional desktop and laptop computers and can be thought of as prototypes of the next generation of many-core processors. For certain classes of population-based Monte Carlo algorithms they offer massively parallel simulation, with the added advantage over conventional distributed multi-core processors that they are cheap, easily accessible, easy to maintain, easy to code, dedicated local devices with low power consumption. On a canonical set of stochastic simulation examples including population-based Markov chain Monte Carlo methods and Sequential Monte Carlo methods, we nd speedups from 35 to 500 fold over conventional single-threaded computer code. Our findings suggest that GPUs have the potential to facilitate the growth of statistical modelling into complex data rich domains through the availability of cheap and accessible many-core computation. We believe the speedup we observe should motivate wider use of parallelizable simulation methods and greater methodological attention to their design.     

An ion cooler-buncher for high-sensitivity collinear laser spectroscopy at ISOLDE

A gas-filled segmented linear Paul trap has been installed at the focal plane of the high-resolution separator (HRS) at CERN-ISOLDE. As well as providing beams with a reduced transverse emittance, this device is also able to accumulate the ions and release the sample in bunches with a well-defined time structure. This has recently permitted collinear laser spectroscopy with stable and radioactive bunched beams to be demonstrated at ISOLDE. Surface-ionized ^{39, 44, 46}K and ⁸⁵Rb beams were accelerated to 30keV, mass separated and injected into the trap for subsequent extraction and delivery to the laser setup. The ions were neutralized in a charge exchange cell and excited with a co-propagating laser. The small ion beam emittance allowed focussing in the ion-laser overlap region, which is essential to achieve the best experimental sensitivity. Fluorescent photons were detected by a photomultiplier tube as a frequency scan was taken. A gate (typically 7-12μs wide) was set on the photomultiplier signal to accept the fluorescent photons within the time window defined by the bunch. Thus, using accumulation times of 100ms, the dominant contribution to background due to continuous laser scattering could be reduced by a factor of up to 4×10⁴ .     

Extreme \alpha -clustering in the 18O nucleus

The structure of the ¹⁸O nucleus at excitation energies above the $ \alpha$ decay threshold was studied using ¹⁴C + $ \alpha$ resonance elastic scattering. A number of states with large $ \alpha$ reduced widths have been observed, indicating that the $ \alpha$ -cluster degree of freedom plays an important role in this N $ \ne$ Z nucleus. A 0⁺ state with an $ \alpha$ reduced width exceeding the single-particle limit was identified at an excitation energy of 9.9±0.3 MeV. We discuss evidence that states of this kind are common in light nuclei and give possible explanations of this feature.     

Synthesis of E-3-Hydroxymethyl-1-methoxy-4-(1′-pentenyl)benzene

The title compound with exclusively the E-geometry has been synthesized by two independent routes. When this compound was reacted with cerium (IV) ammonium nitrate, it was found that oxidative cyclisation to the corresponding 4-hydroxybenzo-pyrans did not occur. This further defines the structural parameters required for such oxidative cyclisations.     

A fiber connectorized MEMS variable optical attenuator

A voltage-controlled moving-mirror microelectro-mechanical systems variable optical attenuator is described that has less than 1-dB fiber-to-fiber insertion loss at 1550-nm wavelength and greater than 50-dB dynamic range. The device was configured with a simple feedback circuit to operate as an optical power regulator capable of stabilizing the output power to within 0.26 dB for a 12-dB input power excursion