Effects of Solution Chemistry Conditions and Adsorbent Surface Properties on Adsorption of Ni(Ⅱ) on Laiyang Bentonite

The effects of solution chemistry conditions and adsorbent surface properties on the adsorption of Ni(II) on Laiyang bentonite were investigated via the batch technique. Potentiometric and mass titration techniques were employed in the batch experimental methods, and the results showed that the point of zero net proton charge(PZNPC) of bentonite at different ionic strength denoted p HPZNPC to be 8.2±0.1. The removal of Ni(II) from the solution increased with an increasing bentonite dosage, with the maximum removal efficiency equating up to 99%. The adsorption of Ni(II) on bentonite increased with an increasing p H value at a p H value of <8.5, and reached a Ni(II) removal efficiency of >99% at a p H value of >10.2. The Ni(II) adsorption performance exhibited different responses to cations(K+, Na+) but was not influenced by the background anions(NO3-, Cl-, and Cl O4-). The adsorption of Ni(II) was dominated by the outer-sphere surface complexation and ion exchange with Na+/H+ on bentonite surface at low p H value, whereas the inner-sphere surface complexation and surface precipitation were the main adsorption mechanisms at high p H value. The adsorption isotherms of Ni(II) on bentonite can be described well by the Langmuir model. The thermodynamic parameters of adsorption, including the Gibbs free energy, the enthalpy change, and the entropy change, at different temperatures indicated that the adsorption of Ni(II) on bentonite was endothermic and spontaneous.

基于正态分布的距离保持哈希图像检索方法

针对图像检索任务中部分监督学习部署困难,以及一般无监督学习没有利用监督信息导致检索性能劣化的问题,提出一种基于正态分布的距离保持哈希的无监督框架,使生成的哈希码保持图像的原始距离关系,在检索结果中尽可能保留相似的图像;距离保持哈希使用正态分布框架约束生成的连续码保持原始特征的距离关系,将图像的语义信息尽可能保留到哈希码中,并使用标准化欧氏距离进行特征向量的相似度计算,解决直接使用传统欧氏距离作为损失时神经网络梯度下降产生的网络波动问题。实验结果表明,与其他模型相比,距离保持哈希在大规模图像检索中检索性能更好。