混合价聚十五钒酸盐[V_(15)O_(36)Cl](NH_4)_3Na_7·30H_2O的合成与晶体结构

以XRD法确定制得的标题晶体属空间群D_(3A)~3,a=13.111,c=11.130。结构由帕特逊-极小函数法结合差值法解出,经精修得一致性因子R_w=0.0344。晶胞中坐落在6m2特殊位置上的是包合1个Cl~-的聚十五钒酸根[V_(15)O_(36)Cl]~(10-),配合键价法确认这是四价钒占主导地位的一类特种混合价同多酸根。     

白酒中游离有机酸的定量测定

本文介绍了一种简单、准确测定白酒中游离有机酸的方法。酒样经中和、浓缩后,未经衍生直接注入毛细管色谱柱,经气相色谱-质谱鉴定并采用三种内标物定量,可准确测定白酒中C2～C18的二十种游离脂肪酸及三种苯羧酸,结果重现性良好。     

无过渡金属催化条件下咪唑衍生物与炔基溴的炔基化反应研究

以炔基溴和咪唑衍生物为反应原料,详细探讨了无过渡金属催化条件下的咪唑类芳香炔胺化合物的合成方法.研究结果表明,以N,N-二甲基甲酰胺为溶剂、碳酸钾为碱,于80℃下反应12 h即可获得中等偏上收率的碳-氮偶联产物.通过上述方法合成得到了系列芳香炔胺衍生物,其结构经1H NMR和13C NMR表征.     

双核烷基咪唑过氧磷钨酸盐相转移催化剂合成及催化烯烃环氧化性能

合成并表征了一类双核长链烷基咪唑阳离子修饰的过氧磷钨杂多酸盐催化剂[Dnmin]1.5PW4O24,考察了催化剂在过氧化氢为氧源的烯烃环氧化反应中的催化活性.研究表明,这类催化剂在反应过程中表现出相转移催化现象,并具有较高的催化活性和选择性.其中,双核十二烷基咪唑杂多酸盐催化剂[D12min]1.5PW4O24的活性最佳,其环己烯转化率和环氧环己烷选择性分别达到97.7%和96.3%.催化剂在经过简单离心分离后可重复使用,重复使用4次后环己烯转化率和环氧环己烷选择性仍可分别达到72.4%和97.2%.催化剂[D12min]1.5PW4O24在其它几种烯烃的环氧化反应中均表现出相转移催化特性,且具有较高的催化活性.     

Fine-tuning the molecular energy levels by incorporating thiophene units onto the π-backbone of core-expanded naphthalene diimides

A series of core-expanded naphthalene diimides (NDI-DTYM) and thiophene-based derivatives (1a-c) were designed and synthesized to investigate the relationship between molecular structures and the highest occupied molecular orbital (HOMO) energy levels but has little impact on the lowest unoccupied molecular orbital (LUMO) energy levels. The results demonstrated that increasing the number of thiophene units can gradually elevate the HOMO energy levels but had little impact on the LUMO energy levels. The n-channel organic field-effect transistors (OFETs) based on 1b and 1c have demonstrated that these almost unchanged LUMO energy levels are proper to transport electrons.     

基于条件生成式对抗网络的油藏单井产量预测模型

为克服机器学习方法在油藏单井产量预测中的过拟合问题,提高油田生产中的产量预测精度,提出一种基于条件生成式对抗网络(CGAN)的油藏单井产量预测模型。该模型使用长短期记忆、全连接等基础神经网络,构建生成和判别网络模型。生成网络模型以产量影响因素为条件输入,生成预测产量数据,利用对数损失函数评价预测数据与真实数据之间的偏差,通过条件生成式对抗网络的博弈训练,并结合贝叶斯超参数优化算法,优化模型结构,综合提高模型的泛化能力。基于Eclipse数值模拟软件建立同一井网条件下不同地质和生产条件下的油藏单井产量数据库,以地质与生产条件等产量影响因素作为模型的条件输入,进行油藏单井产量预测。结果表明：与全连接神经网络(FCNN)、随机森林(RF)以及长短期记忆神经网络(LSTM)模型的预测结果相比,CGAN模型在测试集上的平均绝对百分比误差分别提升了2.59%、 0.81%以及1.72%,并且过拟合比最小(1.027)。说明CGAN降低了机器学习产量预测模型的过拟合程度,提高了模型的泛化能力与预测精度,验证了所提算法的优越性,对指导油田高效开发和保障我国能源战略安全具有重要意义。     

A Hybrid Method Using HAVOK Analysis and Machine Learning for Predicting Chaotic Time Series

The prediction of chaotic time series systems has remained a challenging problem in recent decades. A hybrid method using Hankel Alternative View Of Koopman (HAVOK) analysis and machine learning (HAVOK-ML) is developed to predict chaotic time series. HAVOK-ML simulates the time series by reconstructing a closed linear model so as to achieve the purpose of prediction. It decomposes chaotic dynamics into intermittently forced linear systems by HAVOK analysis and estimates the external intermittently forcing term using machine learning. The prediction performance evaluations confirm that the proposed method has superior forecasting skills compared with existing prediction methods.     

Self‐Assembly and Compartmentalization of Nanozymes in Mesoporous Silica‐Based Nanoreactors

Herein, to mimic complex natural system, polyelectrolyte multilayer (PEM)‐coated mesoporous silica nanoreactors were used to compartmentalize two different artificial enzymes. PEMs coated on the surface of mesoporous silica could serve as a permeable membrane to control the flow of molecules. When assembling hemin on the surface of mesoporous silica, the hemin‐based mesoporous silica system possessed remarkable peroxidase‐like activity, especially at physiological pH, and could be recycled more easily than traditional graphene–hemin nanocompounds. The hope is that these new findings may pave the way for exploring novel nanoreactors to achieve compartmentalization of nanozymes and applying artificial cascade catalytic systems to mimic cell organelles or important biochemical transformations     

Closely packed dense frequency selective surface

In order to realize wideband filtering properties of frequency selective surface （FSS）, FSS of closely packed elements is presented. The Y loop elements are chosen as the graphics elements. Based on the spectral domain method, the frequency response is analyzed for different incident angles and polarizations. The result of the numerical analysis shows that the dense FSS has wide passband with better independence of angle and polarization.