炼油污水的深度处理

对炼油污水深度处理常用技术——过滤、膜分离、化学氧化技术进行了详细的论述，并结合中试试验，对动态砂滤系统、PAN膜处理系统和高级氧化系统处理炼油厂二级排放水的适用性进行了论证。     

Stable isotope on the evaluation of water quality in the presence of WWTPs in rivers

Environmental Science and Pollution Research - We investigated the distribution of nitrogen compounds in Han River as well as two tributaries of Tancheon and Jungrangcheon. Particularly, we...     

广州夏季办公室内细颗粒中多环芳烃污染特征研究

大部分的都市办公人群每天在办公室至少度过8 h。而室外环境的渗透、办公室内吸烟、办公设备使用和中央通风系统均可能导致细颗粒物及多环芳烃(PAHs)在室内积聚而造成微环境污染。2015年5—6月,在广州市3种不同功能区(商住区、高新产业区、工业区)共选取了14间不同类型的办公室,对其室内外PM_(2.5)和多环芳烃进行同步监测。结果表明,(1)14间中有12间办公室内的PM_(2.5)浓度水平高于世界卫生组织(WHO)的推荐值25μg·m-3;(2)与国内外类似研究相比,办公室内外∑16PAHs及Ba P-eq的监测浓度水平均较低,并呈现一致规律:文印>室外>吸烟>多人>单人>无窗(无人),其中Ba P-eq低于欧盟规定的安全限值1 ng·m-3;(3)文印工作和吸烟行为与室内PM_(2.5)和PAHs浓度升高有密切关系,分别对5环和4环PAHs贡献明显;(4)其他无明显内源的办公室的细颗粒中PAHs污染在监测期间主要来源于室外贡献。     

气相色谱-高分辨双聚焦磁质谱测定水中超痕量多氯萘

采用固相萃取法处理水样，气相色谱-高分辨双聚焦磁质谱法测定水中超痕量多氯萘，同位素内标法定量，并对样品前处理条件和仪器条件进行优化。试验表明：方法在0．500 ng/L～500 ng/L范围内线性良好；当取样体积为1 L时，方法检出限为0．005 ng/L～0．01 ng/L；对实际水样进行2个质量浓度水平的加标回收试验，平行测定6次的 RSD为2．7％～8．7％，回收率为70．2％～110％。方法适用性试验表明，水样中复杂的基质对测定无影响。     

High CO2 adsorption on improved ZSM-5 zeolite porous structure modified with ethylenediamine and desorption characteristics with microwave

Journal of Material Cycles and Waste Management - Synthesized zeolite Socony Mobil-5 (ZSM-5) with large pore surface structures and modified with ethylenediamine were used to achieve high CO2...     

兰坪铅锌矿区不同污染梯度下优势植物的重金属累积特征

对兰坪铅锌矿区污染程度不同的3个样地(云南松林、魁蒿群落和马桑灌丛)进行植被调查,选择其中9种共有的自然生长的优势草本植物为研究对象,测定了土壤及植物体的重金属含量。结果显示,3个样地土壤Cu、Cd、Pb和Zn等4种重金属含量均表现为马桑灌丛>魁蒿群落>云南松林;植物体内重金属积累呈现出随着土壤污染程度增加而增加的趋势;所选择的9种植物均不符合超富集植物的标准,依据不同的耐性机制将9种植物分为3类,野棉花能较强吸收土壤中重金属,并转移到地上部分,属于富集型植物;西南金丝梅、倒提壶、长籽柳叶菜、魁蒿、翻白叶和四脉金茅吸收的重金属主要积累在根部,属于根部囤积型植物;尼泊尔蓼和中华山蓼体内重金属含量较低,属于规避型植物。讨论了利用这些植物进行矿山治理。     

Study on the denitrification performance of FexLayOz/activated coke for NH3-SCR and the effect of CO escaped from activated coke at mid-high temperature on catalytic activity

Currently, activated coke is widely used in the removal of multiple pollutants from industrial flue gas. In this paper, a series of novel Fe_xLa_yO_z/AC catalysts was prepared by the incipient wetness impregnation for NH₃-SCR denitrification reaction. The introduction of Fe-La bimetal oxides significantly improved the denitrification performance of activated coke at mid-high temperature, and 4% Fe_0.3La_0.7O_1.5/AC exhibited a superior NO_x conversion efficiency of 90.1% at 400 °C. The catalysts were further characterized by BET, SEM, XRD, Raman, EPR, XPS, FTIR, NH₃-TPD, H₂-TPR, et al., whose results showed that the perovskite-type oxide of LaFeO₃ and oxygen vacancies were produced on the catalysts’ surfaces during roasting. Fe-La doping enhanced the amount of acid sites (mainly Lewis and other stronger acid sites) and the content of multifarious oxygen species, which were beneficial for NO_x removal at mid-high temperature. Moreover, it was investigated that the effect of released CO from activated coke at mid-high temperature on the NO_x removal through the lifetime test, in which it was found that a large amount of CO produced by pyrolysis of activated coke could promote the NO_x removal, and long-term escaping of CO on the activated coke carrier did not have a significant negative impact on catalytic performance. The results of the TG-IR test showed that volatile matter is released from the activated coke while TG results showed that the weight loss rate of 4% Fe_0.3La_0.7O_1.5/AC only was 0.0015~0.007%/min at 300–400 °C. Hence, 4% Fe_0.3La_0.7O_1.5/AC had excellent thermal stability and denitrification performance to be continuously used at mid-high temperature. Finally, the mechanisms were proposed on the basis of experiments and characterization results.

     

基于人工神经网络的气体泄爆最大超压预测研究

为解决传统经验公式在预测气体泄爆中最大超压出现时的较大偏差或过于保守的问题，提出使用人工神经网络预测气体泄爆最大超压。基于124组实验数据，采用BP与RBF神经网络，通过优化算法计算与迭代循环对泄爆样本中的影响因素进行降维与选择，并确定2类神经网络本身在学习与计算气体泄爆样本时的相关参数。结果表明:PCA（主成分分析法）在当前样本条件下的降维效果较差，而通过迭代对比确认气体泄爆样本中的5类特征全部保留时神经网络的训练模拟效果最好；通过对124组实验数据进行随机挑选训练集与测试集的训练模拟结果发现，神经网络对气体泄爆中最大超压的预测效果较好；通过对比Molkov提出的和经Fakandu等改进的NFPA 68经验公式以及2类神经网络的预测结果表明，神经网络相比于传统气体泄爆经验公式具有明显优势。     

Structure optimization and annual performance analysis of the lens-walled compound parabolic concentrator

The primary lens-walled compound parabolic concentrator (lens-walled CPC) has a significant advantage of a larger half acceptance angle as a static solar concentrator, but it also has a drawback of a low optical efficiency. In order to overcome this drawback, in this article, series of structure parameters were investigated and compared to further improve the optical efficiency within the half acceptance angle combined with the material properties. The average optical efficiencies of the improved lens-walled CPCs could achieve more than 82% within the half acceptance angle of 35?. Experiments were adopted to verify the credibility and validity of the simulation. Moreover, annual performance of the lens-walled CPCs comparison with that of the mirror CPC for Nottingham was analyzed. Results show that the improved lens-walled CPC has a higher optical performance for actual building application.