废水中偏二甲肼在Ni/Fe催化剂上的催化分解研究

研究了Ni/Fe催化剂对废水中偏二甲肼臭氧化分解的催化作用,考察了组分含量、体系的pH值和偏二甲肼初始浓度对催化反应的影响.结果表明,Ni/Fe催化剂对水中偏二甲肼的臭氧化具有良好的催化活性.催化剂组分含量、体系的pH值和初始浓度对反应的影响程度不大.对催化剂的XRD表征结果表明,催化剂主要由尖晶石结构的铁酸盐和FeNi3合金相组成,催化剂的良好催化性能与催化剂中尖晶石结构的铁酸盐和FeNi3合金相的形成有关.     

偏二甲肼纯度分析方法的改进

采用毛细管气相色谱内标法代替标准工作曲线法测定偏二甲肼纯度(UDMH),内标法的使用避免了微量进样不准确的缺陷.结果显示,该方法线性范围宽,偏二甲肼的标准曲线相关系数为0.9959,样品加标回收率98.39%,样品的测定标准偏差为0.0015,方法的准确度高、重现性好.     

三阶泛函微分方程的周期解的存在性

本文利用重合度理论研究了一类三阶泛函微分方程方程的2π-周期解的问题,获得了该方程存在2π-周期解的若干新结论,改进推广了有关文献中的已有结果．     

First-principle studies on the electronic structure of Fe3O4(110) surface

The first-principle was employed to study the six possible models for the Fe₃O₄(110) surface, namely the AB-terminated surface (AB model), the AB-terminated with Fe_A vacancy (AB-Fe_A vac model), the AB-terminated with Fe_B vacancy (AB-Fe_B vac model), the B-terminated surface (B model), the B-terminated surface with Fe_B vacancy (B-Fe_B vac model) and the B-terminated surface with O vacancy (B-O vac model). The stability, the electronic structure and the magnetic properties of the six surface models were also calculated. The results predict that the B-O vac model is more stable than other types of surface models. The half-metallic property remain in the AB and B models, while the other four surface models exhibit metallic properties. At the same time, the AB, AB-Fe_A vac, AB-Fe_B vac, B and the B-Fe_B vac models have ferrimagnetic properties, while the B-O vac model has antiferromagnetic property.      

O掺杂MgF2电子结构和光学性质的第一性原理研究

基于密度泛函理论(DFT)的第一性原理平面波超软赝势方法,计算了O原子不同比例(12.5％,8.33％和6.25％)掺杂MgF2晶体的几何结构、电子结构和光学性质.通过对比发现,由于O原子的掺入,体系的禁带宽度减小,材料呈现半金属性.计算也表明,O掺杂对静态介电常数和光吸收系数有重要调制作用,同时也给出了体系性质变化的...     

氢链系统中的非线性问题

考虑氢原子之间的非谐相互作用,分别用代数方法和变分法求得氢链系统中扭结与反扭结孤波解,解的特点是在非谐项的影响下,正孤波与反孤波不再对称,从而解释了实验中观察到的冰分子中L缺陷和D缺陷的对称性破缺现象,进一步的计算表明,由于非谐项的存在,正孤波与反孤波携带的能量也失去对称性,且代数方法和变分法给出一致的结果.     

混合优化RBF-BP网络的板形缺陷识别

针对传统板形模式识别方法存在精度低、鲁棒性弱的问题,提出了一种混合优化RBF-BP组合神经网络板形模式识别方法。首先利用自组织映射网络(SOM)对样本聚类,利用聚类后的网络拓扑结构确定RBF的中心,并计算RBF的宽度,克服了传统聚类算法随机选取中心导致聚类结果不稳定的问题。然后利用遗传算法(GA)良好的全局搜索能力优化整个网络的权值。RBF-BP组合神经网络是由一个RBF子网和一BP子网串联构成的,该网络同时具备BP神经网络能较好地预测未知样本的能力以及RBF神经网络的逼近速度快的优点。并以某900HC可逆冷轧机板形识别为应用背景,在MATLAB2010a环境下进行仿真实验,结果表明混合优化RBF-BP组合神经网络的板形模式识别方法能够识别出常见的板形缺陷,提高了板形缺陷识别精度并具有较好的鲁棒性,可以满足板带轧机高精度的板形控制要求。     

超级微波消解-电感耦合等离子体发射光谱（ICP-OES）法测定土壤中18种元素

为提高土壤多元素同时检测的效率,采用超级微波消解-电感耦合等离子体发射光谱法测定土壤中钾、钠、钙、镁、铜、铁、锰、锌、磷、硫、硼、砷、镉、铬、铅、钴、镓、锂等18种元素含量。比较了超级微波消解、常规微波消解和电热板消解的处理效果,采用超级微波消解法对样品进行前处理,并优化了消解条件。在最优条件下,各元素的检出限在0.05~20 mg/kg,加标回收率在86.2%~107.5%,RSD在0.1%~3.0%,方法准确度及精密度可以满足多元素同时测定的需求,且该方法具有简单、快速、成本低、用酸量少、重现性好等特点。     

A Thermal Non-Neutral Plasma Column and Its Dispersion Relation

In this paper we study the dispersion relation of thermal non-neutral plasma column by MHD theory. We use the dispersion relation to discuss the stability and dispersive properties in a non-neutral plasma-filled waveguide. The results are in agreement with those of cold non-neutral plasma.     

Fractal Aggregation Near the Threshold

In this papel, we present two fractal aggregation models, line pattern seed model (model 1) and point pattern seed model (model 2), which are particle-cluster models. Using the current models, we investigate the critical transition in fractal aggregation processes in two dimensions, and suggest a method for finding the critical transition point. The computer simulation results show that the critical concentration is P_ca=0.69±0.02 for model 1 and P_ca=0.72±0.01 for model 2, critical fractal dimension. D_c= 1.71±0.06 for model 1 and D_c=1.66±0.07 for model 2, which are in good agreement with those of DLA model (D=5/3) and experimental data. The results also show that the critical transition point in two dimensions seems to be inilependent of the size of lattices and the initial seed patterns. The results seem to belong to the same universality class.