硫铁矿生产硫酸的废水治理

电石渣中和－精密微孔过滤技术治理硫酸废水的工艺简洁、可靠，占地面积小，运行费用低，出水优于国家二级排放标准。     

上海崇明岛潮滩重金属的磁学研究

河口潮滩具有独特的环境功能和生态价值，是重金属污染物的一个有效汇库，在河口和近海污染物的迁移转化过程中扮演着十分重要的角色。系统分析了上海崇明岛的表层沉积物中8个样点样品的重金属元素和磁化率（x）、非滞后剩磁（ARM）．非滞后剩磁磁化率（xARM）以及饱和等温剩磁（S／RM）等磁参数之间的关系。结果表明。重金属Pb，Cr、Cu、Zn元素与磁性矿物特别是亚铁磁性矿物之间在赋存形态、沉积特征上存在一定的联系，因此磁性测量可作为沉积物重金属污染记录信息提取的辅助手段，但不同的磁性参数对重金属的指示程度不同。研究得出沉积物中以较粗的多畴亚铁磁性矿物为主，同时x、xARM,／SIRM对重金属元素含量变化比较敏感，因而X、XARM/SIRM可作为重金属含量的替代指标。在重金属元素和磁性参数相关性的基础上，建立了磁诊断线性回归模型。     

Alternating shortcut nitrification-denitrification for nitrogen removal from soybean wastewater by SBR with real-time control

A novel treating technology for nitrogen removal from soybean wastewater was studied. The process for nitrogen removal was achieved by alternating aeration and mixing, combined with real-time control strategies.Results showed that the COD and total nitrogen removal rates are more than 90% and 92% at COD and total nitrogen loads of 1.0- 1.2 kg COD/(kgMLSS. d) and 0.20- 0.27 kg TN/(kgMLSS. d), respectively. In addition, it could improve sludge settling property. SVl value is less than 70 g/ml during the whole cycles. The method not only may be adapted to treat soybean wastewater with high nitrogen, but also may be applied to treat other high nitrogen wastewater.     

微生物法烟气脱硫技术研究

采用微生物（氧化亚铁硫杆菌，简称TF菌）法脱除烟气中的二氧化硫。在相同操作条件下，比较了传统的工业化 方法稀硫酸吸收法（千代田法）与本法的优劣，研究了液气化、二氧化硫浓度、三价铁离子浓度等因素对吸收效率的影响；分析了脱硫反应前后吸收率pH值、三价铁离子浓度的变化。研究结果表明：适宜的液气比（12.5L/Nm^3 以上）、二氧化硫浓度（1000－5000ppm)和三价铁离子浓度（0.6g/L以上）能使该法具有较高（＞98％）的脱硫率。在常温常压操作条件下，微生物法的脱硫率远优于千代田法，具有一定的工业化价值。     

应用分子全息QSAR技术预测硝基芳烃的遗传毒性

分子全息是一种新的分子结构表征技术，基于分子全息的结构．活性关系(HQSAR)技术具有高效、快速、预测能力高等优点，非常适合于分析大量数据样本．本文应用这一最新的QSAR技术，采用偏最小二乘回归技术，研究了54种硝基芳烃的致突变毒性，建立了QSAR定量预测模型．交叉验证和非交叉验证结果表明，模型具有良好的拟合效果和显著的预测能力，进一步的测试数据检验表明，模型具有高度的预测能力和很好的稳健性，可以用来对这一类典型的有毒有机污染物的毒性进行预测．     

社会最优可持续发展模型的建立与分析

通过对能源、环境、经济和社会发展4者关系的分析，建立了社会可持续发展模型，进而找到了社会最优可持续发展轨迹，为国家或地区社会发展的战略决策提供了可资参考的理论工具，同时进一步深化了可持续发展概念，提出了唯一的社会最优可持续发展概念。     

单甲脒等农药的土壤薄层层析行为

用土壤薄层层析法研究单甲脒，单甲脒盐酸盐、灭幼脲和涕灭威等农药在不同土壤中的迁移行为，分别获得了这几种化合物的Ｒｆ值，它们分别为０．１４、０．１７，０．０８，０．７６，研究表明，单脒，单甲脒盐酸盐和灭幼脲不会污染地下水，而涕灭威则会对地下水污染构成严重的威胁。     

Can photoperiod manipulation affect gonad development of a boreo-arctic echinoid (Strongylocentrotus droebachiensis) following exposure in the wild after the autumnal equinox?

A laboratory experiment was conducted in the winter of 2003–2004 to assess the effect of varying photoperiod regime on consumption rate, assimilation rate, absorption efficiency, and gonad development of the green sea urchin, Strongylocentrotus droebachiensis. Adult individuals were collected from the wild after they had been exposed to the ambient autumn photoperiod cue (which is the extraneous trigger thought to elicit gametogenesis in this species) and placed at ambient temperature for 12 weeks under five different photoperiod regimes: (1) 24 h light:0 h dark=“0D”, (2) 16 h light:8 h dark=“8D”, (3) 8 h light:16 h dark=“16D”, (4) 0 h light:24 h dark=“24D”, and (5) ambient photoperiod (range: 10.50–15.25 h dark). Urchins in these five treatments were fed ad libitum with bull kelp, Nereocystis luetkeana. A sixth treatment consisted of starved individuals held under 0D conditions. Various gonad factors including gonad index, percent gonad water, gonad colour (CIE lightness or L*, CIE hue or a*, and CIE chroma or b*), percent area occupation of the gonad by various cell types (nutritive phagocytes, spermatozoa, and secondary oocytes/ova), and stage of development were assessed at the beginning of the experiment and at weeks 4, 8, and 12 of the study. Consumption and assimilation rates were assessed at weeks 4 and 12 and absorption efficiency at week 12 of the experiment. Urchins were predominantly in the growing and premature stages at the beginning of the experiment, but by week 4 at least 20% of individuals in all treatments receiving food were classified as mature. Spawning occurred during all these treatments between weeks 4 and 8, as evidenced by significant decreases in spermatozoa and secondary oocytes/ova and a significant decrease in percent gonad water, but was not accompanied by major declines in gonad indices. Greater than 90% of individuals in all five of the fed treatments were in the recovering and growing stages at the end of the experiment. The 16D treatment had by far the greatest percentage of urchins in the growing stage. In contrast, individuals that were starved were predominantly in the mature stage at weeks 4, 8, and 12, with only ~30% reaching the spawning stage by the end of the experiment. Photoperiod significantly affected gonad indices at the termination of the experiment with gonad index being the highest in the 16D treatment; this was significantly greater than in the 8D and ambient treatments. Photoperiod did not significantly affect gonad percent water, gonad lightness, or gonad hue. Gonad chroma was significantly affected by photoperiod, urchins held under ambient conditions having significantly lower b* readings than individuals in any other treatment. Photoperiod had little or no affect on consumption rate, assimilation rate, or absorption efficiency. Thus, differences among treatments in regards to gonad index, gonad chroma, and stage of development cannot be attributed to variations in feeding, absorption, or assimilation. The results of this experiment indicate that once gametogenesis is initiated, photoperiod manipulation cannot prevent ultimate spawning. However, photoperiod regime can affect the rate at which urchins move through the various stages of the gametogenic cycle. Urchins placed on short days under artificial lighting (16D) moved through the spawning stage into recovering and growing stages the fastest. This photoperiod regime also produced the highest gonad index at the end of the experiment. Since the commercial urchin market prefers large gonads in the growing and premature stages (i.e. before the mature stage is reached and gonads start leaking sperm and eggs), short day-lengths under artificial lighting (16D) appear to be the best photoperiod conditions for optimizing marketability.     

人工神经网络在强化一级处理系统过程模拟与优化中的应用

针对化学强化一级处理系统（CEPT）处理废水时影响因素多，难以进行适当的控制和处理效果的预测等问题，建立起基于BP人工神经网络的CEPT法处理猪场稳定塘废水预测模型，并应用该模型对烧杯试验进行了模拟。结果表明，预测值和实测值吻合较好，模型对COD、总磷、浊度去除率预测的平均相对误差分别为7．5％、4．8％和4．9％。通过对pH值和絮凝剂投药量等可控参数进行优化计算，得到CEPT系统的最佳操作条件和最合理操作条件。该模型的建立为CEPT法处理废水工艺系统实现自动化控制提供了一条简便实用的途径。