THE GROWTH OF ICE DENDRITES UNDER MIXED CONVECTION CONDITIONS

Dendritic solidification of ice from a supercooled melt is studied under conditions which include thermal convection, forced convection, and the effect of the moving solid-liquid interface, which manifests itself as apparent convection. A theory based on Oseen rectilinear flow suggests that heat transfer from the tip of a growing dendrite can be described by a linear superposition of Peclet numbers due to the individual mechanisms of convection by forced bulk motion of the fluid and by the moving boundary which creates the apparent convection effect

This simple result was tested by plotting the experimental data over the full range available of rates of dendritic growth, in terms of the Nusselt number, as a function of the Peclet number, Pe^*, which is based on the tip radius and the sum of the forced velocity of the fluid and the velocity of the tip of the dendrite. This procedure clearly reveals three regions in which different mechanisms control the process: 1) Dendritic growth is dominated by apparent and natural convection when the Peclet number, Pe^*, is below about 2 10^-3 2) When Pe^* is equal to or greater than 10^-1, dendritic growth is dominated by forced convection. 3) Between these limits, the combined effects of apparent and natural convection compete for dominance with forced convection. The equation which correlates the data over the full range of operating conditions, including the mixed convection region is obtained by non linear superposition of the Nusselt number correlations for Pe^* less than 2 10^-3 and greater than 10^minus;1. The resulting Nusselt number represents the experimental data well over the full range studied.     

Phosphorus-overpressure rapid thermal annealing of indium phosphide

A rapid thermal annealing technique for InP is described in which a controllable phosphorus overpressure, generated by heating red phosphorus, suppresses the dissociation of InP at required annealing temperatures. Two annealing configurations were used to independently study the effects of phosphorus overpressures, anneal temperatures and gas flow rates on the post-anneal electrical and morphological properties of low dose Si-implanted InP:Fe. The advantage of phosphorus overpressure annealing over close-contact annealing is shown, and comparison is made with Si₃N₄ encapsulated annealing. Gas flow velocities close to the sample are found to significantly affect the surface morphology, and a static layer immediately above the sample is found to be beneficial.     

Speaker distinguishing distances: a comparative study

Speaker discrimination is a vital aspect of speaker recognition applications such as speaker identification, verification, clustering, indexing and change-point detection. These tasks are usually performed using distance-based approaches to compare speaker models or features from homogeneous speaker segments in order to determine whether or not they belong to the same speaker. Several distance measures and features have been examined for all the different applications, however, no single distance or feature has been reported to perform optimally for all applications in all conditions. In this paper, a thorough analysis is made to determine the behavior of some frequently used distance measures, as well as features, in distinguishing speakers for different data lengths. Measures studied include the Mahalanobis distance, Kullback-Leibler (KL) distance, T ² statistic, Hellinger distance, Bhattacharyya distance, Generalized Likelihood Ratio (GLR), Levenne distance, L ₂ and L _∞ distances. The Mel-Scale Frequency Cepstral Coefficient (MFCC), Linear Predictive Cepstral Coefficients (LPCC), Line Spectral Pairs (LSP) and the Log Area Ratios (LAR) comprise the features investigated. The usefulness of these measures is studied for different data lengths. Generally, a larger data size for each speaker results in better speaker differentiating capability, as more information can be taken into account. However, in some applications such as segmentation of telephone data, speakers change frequently, making it impossible to obtain large speaker-consistent utterances (especially when speaker change-points are unknown). A metric is defined for determining the probability of speaker discrimination error obtainable for each distance measure using each feature set, and the effect of data size on this probability is observed. Furthermore, simple distance-based speaker identification and clustering systems are developed, and the performances of each distance and feature for various data sizes are evaluated on these systems in order to illustrate the importance of choosing the appropriate distance and feature for each application. Results show that for tasks which do not involve any limitation of data length, such as speaker identification, the Kullback Leibler distance with the MFCCs yield the highest speaker differentiation performance, which is comparable to results obtained using more complex state-of-the-art speaker identification systems. Results also indicate that the Hellinger and Bhattacharyya distances with the LSPs yield the best performance for small data sizes.     

CONQUEST: A Coarse-Grained Algorithm for Constructing Summaries of Distributed Discrete Datasets

In this paper we present a coarse-grained parallel algorithm, CONQUEST, for constructing bounded-error summaries of high-dimensional binary attributed data in a distributed environment. Such summaries enable more expensive analysis techniques to be applied efficiently under constraints on computation, communication, and privacy with little loss in accuracy. While the discrete and high-dimensional nature of the dataset makes the problem difficult in its serial formulation, the loose-coupling of distributed servers hosting the data and the heterogeneity in network bandwidth present additional challenges. CONQUEST is based on a novel linear algebraic tool, PROXIMUS, which is shown to be highly effective on a serial platform. In contrast to traditional fine-grained parallel techniques that distribute the kernel operations, CONQUEST adopts a coarse-grained parallel formulation that relies on the principle of sampling to reduce communication overhead while maintaining high accuracy. Specifically, each individual site computes its local patterns independently. Various sites cooperate in dynamically orchestrated work groups to construct consensus patterns from these local patterns. Individual sites may then decide to continue their participation in the consensus or leave the group. Such parallel formulation implicitly resolves load-balancing and privacy issues while reducing communication volume significantly. Experimental results on an Intel Xeon cluster demonstrate that this strategy is capable of excellent performance in terms of compression time, ratio, and accuracy with respect to post-processing tasks.     

Effects of in situ and ex situ formations of silica nanoparticles on polyethersulfone membranes

This study deals with the observed changes in the structure and performance of polyethersulfone (PES) membranes due to in situ formation and ex situ addition of silica particles (SiO₂). Hydrolysis and condensation of tetraethyl orthosilicate (TEOS) inside the PES polymer matrix and the reaction of TEOS with ammonium hydroxide were chosen to form in situ and ex situ SiO₂ formations, respectively. The resultant structure confirmed by X-ray diffraction for the composite PES membranes showed the retention of the amorphous nature even after the addition of SiO₂. The FTIR study revealed the functional groups corresponding to silica networks with enhanced OH signatures on the surface of the composite membranes. Field emission scanning electron microscopic images showed the variation in the surface and cross-sectional structures for the pure and composite membranes. Considerable reduction in the thickness of the skin, difference in the pore structure and ‘finger-like’ cross-sectional morphology with the presence of SiO₂ was observed in PES membranes. Both SiO₂/PES composite membranes were showed a minor change in their glass transition temperature (T _g). The ex situ methodically formed composite membrane displayed an increase in the pure water flux and decrease in bovine serum albumin rejection as compared to in situ and pure PES membranes. These kinds of composite membranes can be utilized for water treatment applications demanding higher water flux.     

1-Indanone and 1,3-indandione Derivatives as Ligands for Misfolded α-Synuclein Aggregates

The development of imaging agents for in vivo detection of alpha-synuclein (α-syn) pathologies faces several challenges. A major gap in the field is the lack of diverse molecular scaffolds with high affinity and selectivity to α-syn fibrils for in vitro screening assays. Better in vitro scaffolds can instruct the discovery of better in vivo agents. We report the rational design, synthesis, and in vitro evaluation of a series of novel 1-indanone and 1,3-indandione derivatives from a Structure-Activity Relationship (SAR) study centered on some existing α-syn fibril binding ligands. Our results from fibril saturation binding experiments show that two of the lead candidates compounds 8 and 32 bind α-syn fibrils with binding constants (K_d) of 9.0 and 18.8 nM, respectively, and selectivity of greater than 10× for α-syn fibrils compared with amyloid-β (Aβ) and tau fibrils. Our results demonstrate that the lead ligands avidly label all forms of α-syn on PD brain tissue sections, but only the dense core of senile plaques in AD brain tissue, respectively. These results are corroborated by ligand-antibody colocalization data from Syn211, which shows immunoreactivity toward all forms of α-syn aggregates, and Syn303, which displays preferential reactivity toward mature Lewy pathology. Our results reveal that 1-indanone derivatives have desirable properties for the biological evaluation of α-synucleinopathies.     

IoT enabled lung cancer detection and routing algorithm using CBSOA-based ShCNN

The Internet of Things (IoT) has tremendously spread worldwide, and it influenced the world through easy connectivity, interoperability, and interconnectivity using IoT devices. Numerous techniques have been developed using IoT-enabled health care systems for cancer detection, but some limitations exist in transmitting the health data to the cloud. The limitations can be accomplished using the proposed chronological-based social optimization algorithm (CBSOA) that effectively transmits the patient's health data using IoT network, thereby detecting lung cancer in an effective way. Initially, nodes in the IoT network are simulated such that patient's health data are collected, and for transmission of such data, routing is performed in order to transmit the health data from source to destination through a gateway based on cloud service using CBSOA. The fitness is newly modeled by assuming the factors like energy, distance, trust, delay, and link quality. Finally, lung cancer detection is carried out at the destination point. At the destination point, the acquired input data is fed to preprocessing phase to make the data acceptable for further mechanism using data normalization. Once the feature selection is done using Canberra distance, then the lung cancer detection is performed using shepard convolutional neural network (ShCNN). The process of routing as well as training of ShCNN is performed using the CBSOA algorithm, which is devised by the inclusion of the chronological concept into the social optimization algorithm. The proposed approach has achieved a maximum accuracy of 0.940, maximum sensitivity of 0.941, maximum specificity of 0.928, and minimum energy of 0.452.