固体超强酸S_2O_8~(2-)/SnO_2-SiO_2催化合成环己酮1,2-丙二醇缩酮

采用沉淀-浸渍法制备了固体超强酸S2O28-/SnO2-SiO2,以它为催化剂催化环己酮和1,2-丙二醇合成了环己酮1,2-丙二醇缩酮;考察了带水剂种类及用量、酮醇摩尔比、催化剂用量、反应时间对产品收率的影响,并用正交实验对反应条件进行了优化。实验结果表明,适宜的反应条件为:n(环己酮)∶n(1,2-丙二醇)=1∶1.6、催化剂用量为反应物料总质量的2.0%、带水剂环己烷用量5.0mL、反应时间50min。在此条件下,环己酮1,2-丙二醇缩酮收率达到91.47%;催化剂的稳定性良好,在重复使用5次后环己酮1,2-丙二醇缩酮收率为82.20%,活性下降的主要原因为催化剂表面积碳和吸附了有机物;经傅里叶变换红外光谱和气相色谱质谱分析表明,产物为环己酮1,2-丙二醇缩酮,纯度为100%。     

SW233 PIN驱动器自动测试系统的研制

介绍了一种集成电路自动测试系统，该系统采用计算机并口作通信接口，用VB6编程，实现了对外围测试电路的控制，用IEEE-488接口卡控制测试仪器，可对SW233电路的36个参数进行自动测试，并将测试结果自动保存在数据库中。该测试系统具有自动化程度高、操作方便、测试结果精确等特点。     

金刚石光电导探测器研制及在软X光测量中的初步应用

介绍了我们自己设计制造的天然Ⅱａ型金刚石光导探测器的结构，原理，制造工艺，实验标定及其在ＩＣＦ实验中软Ｘ光测量中的应用等。     

智能测温仪表的研究与实现

文章介绍了一种以8098单片机为核心、以铂铑一铂为测温元件研制的智能化测温仪表，它与传统的接触式温度传感器相比具有价格低、性能高、应用范围广、人机界面好等特点。     

我国车用无铅汽油的质量状况及分析

对1998－2001年间生产领域和流通领域的近500批次汽油产品检验数据进行了总结和分析，表明我国汽油的实物质量正在不断提高，但从环保要求出发还需不断向国际水平靠扰，尤其应进一步降低烯烃含量，硫含量。     

RADIOACTIVE OR NATURAL TRACER。TECHNIQUES FOR LEAK DETERMINING OF DAM ABUTMENT

Infiltration and localization of preferential infiltration zones at the dam abutment are measured using radioactive tracer tests of flow in boreholes,meanwhile interconnection between boreholes and the energing water points is analysed.The theory and practice of radioactive tracer synthetic detective method are described to give methods and calculation formulae used under the condition of stable flow in single well to measure permeability coefficient and hydrostatic heads.Major single hole techniques including measurement for seepage line,velocity,rate of seepage flow and relationship of recharge of groundwater in aquifers are introduced briefly.The possibilities offered by natureal tracers are analysed,uincluding electric-conduct,pH-value and temperature of water as well as stable isotopes(D,^18O) and tritium.Furthermore,the sensibilities of this theory and methods were confirmed by detecting seepage flow field of Xinanjiang Dam.     

Protect privacy of deep classification networks by exploiting their generative power

Machine Learning - Research showed that deep learning models are vulnerable to membership inference attacks, which aim to determine if an example is in the training set of the model. We propose a...