On the origin of the metallic and anisotropic magnetic properties of NaxCoO2 (x approximately equal to 0.75)

Nonstoichiometric Na(x)CoO2 (0.5 < x < 1) consists of CoO2 layers made up of edge-sharing CoO6 octahedra and exhibits strongly anisotropic magnetic susceptibilities as well as metallic properties. A modified Curie-Weiss law was proposed for systems containing anisotropic magnetic ions to analyze the magnetic susceptibilities of Na(x)CoO2 (x approximately 0.75), and implications of this analysis were explored. Our study shows that the low-spin Co4+ (S = 1/2) ions of Na(x)CoO2 generated by the Na vacancies cause the anisotropic magnetic properties of Na(x)CoO2 and suggests that the six nearest-neighbor Co3+ ions of each Co4+ ion adopt the intermediate-spin electron configuration, thereby behaving magnetically like low-spin Co4+ ions. The Weiss temperature of Na(x)CoO2 is more negative along the direction of the lower g factor (i.e., theta|| < theta(perpendicular) < 0 and g|| < g(perpendicular)). The occurrence of intermediate-spin Co3+ ions surrounding each Co4+ ion accounts for the apparently puzzling magnetic properties of Na(x)CoO2 (x approximately 0.75), i.e., the large negative Weiss temperature, the three-dimensional antiferromagnetic ordering below approximately 22 K, and the metallic properties. The picture of the magnetic structure derived from neutron scattering studies below approximately 22 K is in apparent conflict with that deduced from magnetic susceptibility measurements between approximately 50 and 300 K. These conflicting pictures are resolved by noting that the spin exchange between Co3+ ions is more strongly antiferromagnetic than that between Co4+ and Co3+ ions.     

Low-power divide-by-3 injection-locked frequency dividers implemented with injection transformers

A new divide-by-3 injection-locked frequency divider (ILFD) is proposed. The ILFD consists of a 7.6 GHz voltage controlled oscillator (VCO) and two transformers, which are in series with the crosscoupled transistors in the VCO for signal injection. The proposed CMOS ILFD has been implemented with the TSMC 0.13 μm CMOS technology. At the supply voltage of 0.8 V, the core power consumption is 1.25 mW, and the free-running frequency of the ILFD is tunable from 7.2 to 7.87 GHz. At the input power of 0 dBm, the total divide-by-3 locking range is from 21.56 to 23.63 GHz as the tuning voltage is varied from 0.0 to 0.8 V. The phase noise of the locked ILFD output is lower than that of the free-running ILFD in the divide-by-3 mode.     

Wavelet Transform Method for Coupled Map Lattices

The purpose of this paper is twofold. First, we derive a necessary and sufficient condition for local synchronization in coupled map lattices (CMLs) with symmetric coupling. In particular, we also identify the best choice of coupling strength in the sense that such a coupling strength gives the fastest convergence rate of initial values toward the synchronous manifold. Furthermore, such a coupling strength is independent of the choice of the individual chaotic map. In addition, it is demonstrated that the wavelet transform method, which is proposed by Wei et al., can greatly increase the applicable ranges of coupling strengths, the parameters of the individual chaotic map, and the number of nodes for local synchronization of CMLs.     

一種製作互補式金氧半複晶矽薄膜電晶體之製程簡化技術

一種製作互補式金氧半複晶矽薄膜電晶體之製程簡化技術被提出, 其藉由大角度離子佈植摻雜物穿透閘極間隙壁. 藉由此技術方法,可使用同一光罩而同時形成淡摻雜區域於閘極間隙壁氧化層下方且形成高摻雜區域,其可於互補式金氧半複晶矽薄膜電晶體製作時簡化製程步驟.更進一步地,藉由此技術,與傳統的單一高濃度摻雜源汲極技術相較,此方法可得到更好的N型及P型薄膜電晶體特性,且無需增加光罩製程.     

Design and implementation of a bistable microcantilever actuator for magnetostatic latching relay

This paper presents the design, fabrication, and implementation of a microcantilever actuator with magnetostatic latching for performing low power bistable relay applications. This unique bistable feature consists of a low-stiffness torsion/cantilever beam system with circular-shaped support and a permanent magnet for holding the closed state with a permalloy soft magnetic circuit. The special circular support is designed to enhance the stiffness of the overhang beams. First, mechanical modeling of the leveraged torsion/cantilever beams was performed by Castigliano’s theorem so as to deduce the spring stiffness of system. Then the device has been prepared by a laminated photoresist sacrificial layer process (LPSLP). Finally, mechanical performance was characterized by atomic force microscopy (AFM), combined with finite element simulation using ANSYS™ package and analysis model as well. Switching between two stable states of the microactuator was successfully validated with WYKO NT1100 optical profiling system.     

Conductive carbon tape as a sample platform for microwave-based MALDI MS detection of proteins and phosphoproteins

In this study, we developed a novel microwave-assisted protein preparation and digestion method for matrix-assisted laser desorption/ionization (MALDI) time-of-flight mass spectrometry analysis and identification of proteins that involves using conductive carbon tape as a sample platform for sample preparation (reduction and alkylation) and digestion under microwave heating and as a plate for MALDI analysis. This method allows for the enzymatic digestion products of proteins to be directly analyzed by MALDI mass spectrometry and results in a marked reduction in sample loss. Our protocol requires only a small volume (1 μL) of reaction solvent, which increases the frequency of enzyme-to-protein contact, thereby resulting in more efficient digestion of sample than conventional in-solution digestion methods. To test this protocol, we used magnetic iron (II, III) oxide nanoparticles as concentrating probes to enrich phosphopeptides from a mixture of peptides in enzymatically digested protein samples. We found that the one-pot on-tape-based protein preparation and digestion under microwave heating combined with the on-tape-based enrichment method not only dramatically reduced the time required for phosphopeptides analysis but also allowed for the simultaneous identification of phosphoproteins. The advantages of our protocol include ease of use, high digestion efficiency, high specificity, and rapid (15 min) identification of proteins and enrichment of phosphopeptides in a mixture of enzymatically digested protein samples.     

MEMS microwell and microcolumn arrays: novel methods for high-throughput cell-based assays

Although the cell-based assay is becoming more popular for high throughput drug screening and the functional characterization of disease-associated genes, most researchers in these areas do not use it because it is a complex and expensive process. We wanted to create a simple method of performing an on-chip cell-based assay. To do this, we used micro-electro-mechanical systems (MEMS) to fabricate a microwell array chip comprised of a glass substrate covered with a photoresist film patterned to form multiple microwells and tested it in two reverse transfection experiments, an exogenous gene expression study and an endogenous gene knockdown study. It was used effectively in both. Then, using the same MEMS technology, we fabricated a complementary microcolumn array to be used as a drug carrier device to topically apply drugs to cells cultured in the microwell array. We tested the effectiveness of microwell-microcolumn on-chip cell-based assay by using it in experiments to identify epidermal growth factor receptor (EGFR) activity inhibitors, for which it was found to provide effective high throughput and high content functional screening. In conclusion, this new method of cell-based screening proved to be a simple and efficient method of characterizing gene function and discovering drug leads.     

Comparison of CID versus ETD‐based MS/MS fragmentation for the analysis of doubly derivatized steroids

Electrospray ionization coupled with collision‐induced dissociation (CID) and tandem mass spectrometry (MS/MS) is a commonly used technique to analyze the chemical composition of steroids. However, steroids are structurally similar compounds, making it difficult to interpret their product‐ion spectra. Electron transfer dissociation (ETD), a relatively new technique for protein and peptide fragmentation, has been shown to provide more detailed structural information. In this study, we compared the ability of CID with that of ETD to differentiate between eight 3,20‐dioxosteroids that had been derivatizated with a quaternary ammonium salt, Girard reagent P (GirP), at room temperature or after exposure to microwave irradiation to generate doubly charged ions. We found that the derivatization of steroid with GirP hydrazine occurred in less than 10 min when the reaction was carried out in the presence of microwave irradiation compared to 30 min when the reaction was carried out at room temperature. According to the MS/MS spectra, CID provided rich, structurally informative ions; however, the spectra were complex, thereby complicating the peak assignment. In contrast, ETD generated simpler spectra, making it easier to recognize individual peaks. Remarkably, both CID and ETD were allowed to differentiate of steroid isomers, 17α‐hydroxyprogesterone (17OHP) and deoxycorticosterone (DOC), but the signature ions obtained from CID were less intense than those generated by ETD, which generated much clearer spectra. These results indicate that ETD in conjunction with CID can provide more structural information for precise characterization of steroids. Copyright © 2013 John Wiley & Sons, Ltd.     

Structural Bayesian Linear Regression for Hidden Markov Models

Linear regression for Hidden Markov Model (HMM) parameters is widely used for the adaptive training of time series pattern analysis especially for speech processing. The regression parameters are usually shared among sets of Gaussians in HMMs where the Gaussian clusters are represented by a tree. This paper realizes a fully Bayesian treatment of linear regression for HMMs considering this regression tree structure by using variational techniques. This paper analytically derives the variational lower bound of the marginalized log-likelihood of the linear regression. By using the variational lower bound as an objective function, we can algorithmically optimize the tree structure and hyper-parameters of the linear regression rather than heuristically tweaking them as tuning parameters. Experiments on large vocabulary continuous speech recognition confirm the generalizability of the proposed approach, especially when the amount of adaptation data is limited.