Qualitative and quantitative analysis of complex temperature-programmed desorption data by multivariate curve resolution

The substantial amount of information carried in temperature-programmed desorption (TPD) experiments is often difficult to mine due to the occurrence of competing reaction pathways that produce compounds with similar mass spectrometric features. Multivariate curve resolution (MCR) is introduced as a tool capable of overcoming this problem by mathematically detecting spectral variations and correlations between several m/z traces, which is later translated into the extraction of the cracking pattern and the desorption profile for each desorbate. Different from the elegant (though complex) methods currently available to analyze TPD data, MCR analysis is applicable even when no information regarding the specific surface reaction/desorption process or the nature of the desorbing species is available. However, when available, any information can be used as constraints that guide the outcome, increasing the accuracy of the resolution. This approach is especially valuable when the compounds desorbing are different from what would be expected based on a chemical intuition, when the cracking pattern of the model test compound is difficult or impossible to obtain (because it could be unstable or very rare), and when knowing major components desorbing from the surface could in more traditional methods actually bias the quantification of minor components. The enhanced level of understanding of thermal processes achieved through MCR analysis is demonstrated by analyzing three phenomena: i) the cryogenic desorption of vinyltrimethylsilane from silicon, an introductory system where the known multilayer and monolayer components are resolved; ii) acrolein hydrogenation on a bimetallic Pt–Ni–Pt catalyst, where a rapid identification of hydrogenated products as well as other desorbing species is achieved, and iii) the thermal reaction of Ti[N(CH₃)₂]₄ on Si(100), where the products of surface decomposition are identified and an estimation of the surface composition after the thermal reaction is afforded. Since this work constitutes, to the best of our knowledge, the first effort to introduce multivariate analysis to TPD data, the procedures, algorithms and strategies employed are described in full detail.     

Narrow band limit of the equivalence between Anderson and Kondo Hamiltonians

It is shown that for a narrow band version of the Anderson Hamiltonian, the transformation of Schrieffer and Wolff can be carried out to all orders in the mixing parameter. The resulting Kondo-like Hamiltonian is in diagonal form. The results can be used to derive a systematic approximation scheme for the Anderson Hamiltonian, through use of the level shift operator of Primas. The resulting interaction parameters remain regular when the energy of the localized level approaches the Fermi level.     

Detrended cross-correlation analysis of electroencephalogram

In the paper we use detrended cross-correlation analysis (DCCA) to study the electroencephalograms of healthy young subjects and healthy old subjects. It is found that the cross-correlation between different leads of a healthy young subject is larger than that of a healthy old subject. It was shown that the cross-correlation relationship decreases with the aging process and the phenomenon can help to diagnose whether the subject's brain function is healthy or not.     

Narrow band limit of the periodic Anderson model with an interatomic hybridization

We analyze the single-particle part of the periodic Anderson model starting from the atomic representation for band (5d) and localized (4f) orbitals. This system with an interatomic hybridization, is transformed into a two-band structure with an intraatomic mixing between the component bands. The resultant hybridized-band structure has narrow lower band with a singular density of states (DOS) at its upper edge.The importance of this singularity for orbitally degenerate 4f or 5f systems is briefly discussed.     

Narrow band filters in both transmission and reflection with metal/dielectric thin films

We present a novel narrow band filter operating in both transmission and reflection for the first time to our knowledge. This proposed structure consists of one unsymmetrical dielectric Fabry-Perot cavity and an ultrathin metal film with n ≈ k. Theoretical analysis shows that both the reflectance and transmittance at the central wavelength are maximums. Due to the high absorption induced by the metal, a good rejection level can be obtained for a wide spectral range. In addition, the changes of peak value ratio R_max/T_max is also investigated by adjusting the amount of dielectric stacks. We finally demonstrate the experimental results to verify these designs.     

脑电信号的标度分析及其在睡眠状态区分中的应用

脑电信号具有长程幂律相关性及多重分形的标度特性,并随生理病理状态改变.本文首次针对睡眠脑电信号应用单重分形去趋势波动分析(detrended fluctuation analysis,简记为DFA)方法与多重分形奇异谱对睡眠脑电信号的标度特征进行系统的对比研究.发现DFA标度指数α对于不同导联和样本组间的差异较为敏感,随睡眠状态的变化不规律;而多重分形奇异强度区间Δα随睡眠状态的变化更为规律,睡眠Ⅰ期至Ⅳ期不断增大,并且导联间差异和样本组间差异均较小.多重分形Δα参数更适合作为判定睡眠状态的定量参数.     

A quantitative analysis of an athermal design for a long period grating based tunable filter

In this paper, a quantitative analysis of an athermal design for a unique all-fiber tunable filter based on the combination of a single resonant band long period grating (LPG) with an electro-optic polymer second cladding layer is presented. The single resonant band LPG is used to select the resonant wavelength, which can be tuned by changing the refractive index of an electro-optic polymer second cladding layer via external electric field. Although the basic operational principle and implementation of this unique tunable filter have been previously reported, this paper is focused on the quantitative analysis of the athermal design, which may significantly enhance the practicability of the proposed tunable filter.     

Experimental study of hydraulic engine mounts using multiple inertia tracks and orifices: Narrow and broad band tuning concepts

Hydraulic engine mount tuning concepts with one inertia track and one decoupler are well understood. However, the dynamic response with multiple tracks or orifices is not. To overcome this void in the literature, dynamic tuning concepts of hydraulic engine mounts, with emphasis on multiple (n-)inertia tracks/orifices, are experimentally examined. A new prototype mount concept is designed, built, and experimentally evaluated in a controlled manner. Refined linear time-invariant models of fixed decoupler-type designs are developed to critically assess the dynamic stiffness measurements and to explore a family of alternate designs. Three narrowband devices are investigated for accurately predicting the frequencies corresponding to peak loss angles for the first time, in addition to examining and validating an n = 3 track mount. Two broadband devices are also successfully evaluated by tuning damping introduced by orifice-type tracks. A special broad-tuned design utilizing a controlled ‘leakage’ path flow area is then suggested, and the role of fluid resistance in achieving the desired performance is clarified. Finally, a production mount with unknown configuration is diagnosed using the proposed models with n tracks.     

基于改进的符号转移熵的心脑电信号耦合研究

试图探究动力系统中的耦合关系一直以来都是国内外众多学者关注的热点,传统的时间序列符号化分析方法会使研究结果受序列非平稳性的严重影响,本文在原有转移熵的研究基础上,应用粗粒化提取,经过理论与实验的分析,发现心脑电信号耦合研究中的转移熵值在不同提取情况下对应不同的分布趋势,并选择效果最好的信号数据提取方法用在其后的应用分析中. 此外,对时间序列符号化方法提出改进,采用动态的自适应分割方法. 实验结果表明,无论清醒期还是睡眠期,改进的符号转移熵算法观测分析到的心脑电信号耦合作用更显著,能更好的捕捉到信号中的动态信息、系统动力学复杂性的改变,更利于医学临床实践应用中的检测,在分析非平稳的时间序列上具有更好的效果. 关键词： 心脑电信号 粗粒化 符号转移熵 基本尺度     

试用聚类分析对船只噪声和脑电分类

聚类分析属模式识别的一种方法,不同样本由于其信息特征有差异,经信号处理与计算机学习,可以在减维的特征空间被分类识别。本文介绍用非线性映射减维变换的方法,对船只噪声和脑电进行分类识别,想通过对两种不同领域里的样本试分类,来说明聚类分析方法的广泛适用性。     

基于自适应模板法的脑电信号转移熵分析

脑电信号是一种产生机理相当复杂且非常微弱的随机信号, 综合反映了大脑组织的脑电活动及大脑的功能状态. 由于脑电信号的微弱性, 传统的基本模板方法在脑电信号分析上得到了良好的应用. 为进一步提升分析脑电信号的性能, 提出了一种新的基于自适应模板的转移熵方法并分析了青少年脑电与成年人脑电信号. 结果表明: 对于青少年脑电还是成年人脑电, 与基本模板法相比, 基于自适应模板法的转移熵可以更显著地表示脑电信号的耦合作用, 并且具有更好的区分度, 这将能更好地捕捉到信号中的动态信息、系统动力学复杂性的改变. 同时, 该方法将更有利于医学临床诊断的辅助检测, 对脑电信号是否处于病理状态的诊断提供了新的更好的判断依据.     

RbVO_3的光谱分析及能带结构研究

采用杂化前驱体的湿化学合成法制备了偏钒酸铷(RbVO3)发光材料。利用粉末X射线衍射(XRD)、红外光谱仪(IR)等对RbVO3的结构进行了表征,利用紫外-可见(UV-Vis)吸收光谱、光致发光光谱等研究了该发光材料的光致发光性能,并采用基于密度泛函理论的CASTEP程序计算了RbVO3的能带。结果表明,在357 nm紫外光的激发下,RbVO3产生发光峰在525 nm附近、色坐标为(0.318 0,0.430 9)的绿白光发射,发光亮度高;RbVO3计算的光学带隙为2.67 eV,与实验值较为吻合。     

Wavelet analysis of learning and forgetting of photostimulation rhythms for a nonstationary electroencephalogram

Quantitative parameters characterizing transient processes of mastering and forgetting of photostimulation (PST) rhythms for a nonstationary electroencephalogram (EEG) are developed on the basis of a continuous wavelet transformation. Nonstationarity factor K _nst(μ), as well as rhythm mastering K _M (μ) and confinement K _C (μ) factors are calculated for various spectral ranges μ. Photoflash mastering time τ _M = τ _S + τ _I , which is the sum of latent silence period τ _S after PST actuation and the rhythm increasing period τ _I is calculated. In the case of PST, the EEG rhythm retardation time τ _R relative to the beginning of PST is calculated. Rhythm forgetting time τ _F = τ _P + τ _D after PST actuation is the sum of the preservation time τ _P of the corresponding rhythm over a certain time interval and its decay period τ _D . The lag time τ _L of the EEG signal relative to the PST signal after its removal is determined. The proposed method is used in quantitative analysis and classification of transient processes characterizing the properties of the central nervous system. Possible applications of the method in analysis of various nonstationary signals in physics are discussed.     

有限开敞介质光子晶体的模式及其带结构分析

针对光子晶体在行波管中的应用,对有限开敞介质光子晶体的模式和带结构进行了分析和计算. 分析表明,有限开敞介质光子晶体中所能存在的模式包括EH和HE混合模式,如果是二维光子晶体,还存在E模式,E₁模式即为无限光子晶体中的E极化. 计算表明,被用作行波管慢波电路的光子晶体可以不必拥有完全带隙,而只需要具有带间隙即可. 关键词： 光子晶体 本征值方程 行波管 带结构     

TOF-SIMS analysis of polystyrene/polybutadiene blend using chemical derivatization and multivariate analysis

Chemical imaging with high spatial resolution is one of the features of TOF-SIMS. However, degradation of the sample due to primary ion bombardment becomes problematic when the analysis area is small. Although polystyrene (PS) and polybutadiene (PB) separately show relatively distinct spectra, observation of their phase separation in PS/PB blends is difficult when the analysis area is small because degradation of both polymers and especially PS leads to disappearance of their characteristic peaks, resulting in low chemical image contrast. We therefore investigated the application of various forms of multivariate analysis (MVA) to the TOF-SIMS image data to improve the chemical image contrast. PCA, MCR, and the other forms of MVA provided improvement in contrast, but the images were still obscure and observation of phase separation remained difficult. Chemical derivatization using osmium tetroxide was also investigated, and found to give clear images of phase separation in the PS/PB blend. In quantitative determinations with MVA and chemical derivatization, PLS demonstrated the best predictive capability and chemical derivatization resulted in large deviations from both the bulk chemical composition and the determinations with MVA, particularly in regions of low PB content.     

EDXRF and multivariate statistical analysis of fragments from Marajoara ceramics

In this work, 102 fragments of Marajoara ceramics, belonging to the National Museum collection (Rio de Janeiro, Brazil), were analyzed using energy dispersive X‐ray fluorescence (EDXRF) and principal component analysis (PCA) in order to identify possible groups of samples that present similar behaviors or different characteristics. This information will give an important aid to a more accurate classification of these artifacts. The EDXRF measurements were carried out with a portable system developed in the Nuclear Instrumentation Laboratory consisting of an Oxford TF3005 X‐ray tube, with W anode, and an Si‐PIN XR‐100CR detector from Amptek, working at 25 kV and 100 µA, acquisition time of 600 s and a beam collimation of 2 mm. PCA applied to the X‐ray fluorescence results revealed a clear cluster separation to the samples. Copyright © 2009 John Wiley & Sons, Ltd.     

Quality control of fragrances using Raman spectroscopy and multivariate analysis

An analytical methodology using Raman spectroscopy and chemometrics was developed for direct, fast and non‐destructive discrimination and prediction of the properties of fragrances according to their composition. The soft independent modeling of class analogies was used as a supervised classification method for fragrances classification, and partial least squares regression as a multivariate calibration method for the prediction of physicochemical properties of fragrances, such as density and refractive index. From 155 fragrance samples, the model exhibited a high success rate for all of the studied fragrance classes, with 100% correct classification. In the multivariate calibration model, adequate correlation was observed between the measured and partial least squares regression‐predicted data for refractive index and density, with a relative standard error of prediction between 0.02% and 0.07%, respectively. This study demonstrates the wide applicability of the methodology for the discrimination, classification, and prediction of complex olfactory mixtures in quality control of fragrances. Copyright © 2015 John Wiley & Sons, Ltd.