Numerical Investigation of Thermally Developing Non-Darcy Forced Convection in a Porous Circular Duct with Asymmetric Entrance Temperature Under LTNE Condition

Transport in Porous Media - This paper numerically investigates the heat transfer performance of thermally developing non-Darcy forced convection in a fluid-saturated porous medium tube under...     

Optimizing Acetylene Sorption through Induced-fit Transformations in a Chemically Stable Microporous Framework

Developing practical storage technologies for acetylene (C₂H₂) is important but challenging because C₂H₂ is useful but explosive. Here, a novel metal–organic framework (MOF) ( FJI-H36 ) with adaptive channels was prepared. It can effectively capture C₂H₂ (159.9 cm³ cm⁻³) at 1 atm and 298 K, possessing a record-high storage density (561 g L⁻¹) but a very low adsorption enthalpy (28 kJ mol⁻¹) among all the reported MOFs. Structural analyses show that such excellent adsorption performance comes from the synergism of active sites, flexible framework, and matched pores; where the adsorbed-C₂H₂ can drive FJI-H36 to undergo induced-fit transformations step by step, including deformation/reconstruction of channels, contraction of pores, and transformation of active sites, finally leading to dense packing of C₂H₂. Moreover, FJI-H36 has excellent chemical stability and recyclability, and can be prepared on a large scale, enabling it as a practical adsorbent for C₂H₂. This will provide a useful strategy for developing practical and efficient adsorbents for C₂H₂ storage.     

Synthesis of Stimuli-responsive Nanoparticles by Solution Polymerization

In this study, stimuli-responsive nanoparticles were prepared by solution polymerization. Two synthesis routes are proposed to synthesize the particles, the monomer route and the polymer/monomer route. For the monomer route, pH and thermal sensitive nanoparticles were synthesized from acrylic acid and N-isopropylacrylamide. For the polymer/monomer route, the pH sensitive nanoparticles were synthesized from chitosan and acrylic acid. The effect of reaction time, initiator concentration and agitation rate on the particle size and the size distribution were investigated. The stimuli-responsive nanoparticles could be directly blended with other polymers to prepare stimuli-responsive functional membranes.     

三维网络结构NiCo2S4@NiCo2O4复合电极的制备及其性能

超级电容器具有更大的功率密度、优秀的循环稳定性、极快的充放电速度、超长的循环寿命以及环境友好等突出特点,其性能与构件关系密切,其中最根本的就是组成它的电极材料。本研究主要采用传统的水热法制备出钴酸镍(NiCo₂O₄)电极材料,进而通过离子交换(二次水热)制得镍钴硫(NiCo₂S₄),最后利用化学浴沉积(CBD)法使其与钴酸镍复合,得到最终所需的三维网络结构NiCo₂S₄@NiCo₂O₄复合电极。经过表面形貌表征、循环伏安测试、恒电流充放电测试以及比电容计算分析等可以证明:三维网络结构NiCo₂S₄@NiCo₂O₄复合电极的比电容及循环稳定性等远远优于复合前单一的纯NiCo₂O₄电极材料,具有极大应用前景。     

Catalytic palladium nanoparticles supported on nanoscale MOFs: a highly active catalyst for Suzuki–Miyaura cross-coupling reaction

Pd nanoparticles have been successfully supported on nanoscale metal-organic frameworks (NMOFs) by using a simple and effective microwave-assisted impregnation process. The resulting composite, representing as a highly active NMOFs supported metal nanoparticles catalyst for the Suzuki cross-coupling reaction between aryl/heteroaryl halides and arylboronic acids, is well characterized, and its high activity and good recyclability are discussed in details. It reveals that, compared to the corresponding bulk MOFs and conventional active carbon materials, nanoscale MOFs as novel support materials for heterogeneous catalysts can exhibit superior performance in the catalytic reactions.     

A new coding scheme of QC-LDPC codes for optical transmission systems

Based on Galois Field (GF(q)) multiplicative group, a new coding scheme for Quasi-Cyclic Low-Density Parity-Check (QC-LDPC) codes is proposed, and the new coding scheme has some advantages such as the simpler construction, the easier implementation encoding, the lower complexity of the encoding and decoding, the more flexible adjustment of the code length as well as the code rate and so forth. Under the condition of considering the characteristics of optical transmission systems, an irregular QC-LDPC (3843,3603) code to be suitable for optical transmission systems is constructed by applying the proposed new coding scheme. The simulation result shows that the net coding gain (NCG) of the irregular QC-LDPC (3843,3603) code is respectively improved 2.14 dB, 1.19 dB, 0.24 dB and 0.14 dB more than those of the classic RS (255,239) code in ITU-T G.975, the LDPC (32640,30592) code in ITU-T G.975.1, the regular SCG-LDPC (3969,3720) code constructed by the Systematically Constructed Gallager (SCG) coding scheme and the regular QC-LDPC (4221,3956) code at the bit error rate (BER) of 10-8. Furthermore, all the five codes have the same code rate of 93.7%. Therefore, the irregular QC-LDPC (3843,3603) code constructed by the proposed new coding scheme has the more excellent error-correction performance and can be better suitable for optical transmission systems.     

B2型钽钨有序合金的声子色散关系的研究

应用改进分析型嵌入原子法模型和晶格动力学理论,计算了金属钽和钨以及B2型钽钨有序合金的原子力常数和动力学矩阵,模拟了金属钽和钨沿[ζ00]、[ζζ0]和[ζζζ]三个对称方向以及B2型钽钨有序合金沿[ζζ0]、[11ζ]、[ζζζ]和[ζ00]四个对称方向的声子色散曲线,并分析了B2型钽钨有序合金的振动特性.结果表明:金属钽和钨的声子色散曲线的模拟结果和实验值符合较好,特别在低频附近二者几乎一致;B2型钽钨有序合金中既有声频振动支又有光频振动支,光频支的振动频率均大于声频支的振动频率,且光频带和声频带之间由宽度约为0.82THz的空带隔开;在[ζζ0]对称方向上有六个独立振动模,在[11ζ]、[ζζζ]和[ζ00]对称方向上六个振动模均兼并为四个独立振动模,且各振动支的振动方向不尽相同.     

Learning errors of linear programming support vector regression

In this paper, we give several results of learning errors for linear programming support vector regression. The corresponding theorems are proved in the reproducing kernel Hilbert space. With the covering number, the approximation property and the capacity of the reproducing kernel Hilbert space are measured. The obtained result (Theorem 2.1) shows that the learning error can be controlled by the sample error and regularization error. The mentioned sample error is summarized by the errors of learning regression function and regularizing function in the reproducing kernel Hilbert space. After estimating the generalization error of learning regression function (Theorem 2.2), the upper bound (Theorem 2.3) of the regularized learning algorithm associated with linear programming support vector regression is estimated.     

Learning rates for multi-kernel linear programming classifiers

In this paper, we consider the learning rates of multi-kernel linear programming classifiers. Our analysis shows that the convergence behavior of multi-kernel linear programming classifiers is almost the same as that of multi-kernel quadratic programming. This is implemented by setting a stepping stone between the linear programming and the quadratic programming. An upper bound is presented for general probability distributions and distribution satisfying some Tsybakov noise condition.