Effect of rare earth content on microstructure and corrosion resistance of hot-dip Zn-5Al alloy coated steel wire for bridge cables

The effects of the content of rare earth elements on the microstructure and properties of hot-dip Zn-5 Al alloy steel wire for bridge cables were investigated.The microstructure of the hot-dip coating was analyzed using an optical microscope and a scanning electron microscope equipped with an energy-dispersive spectrometer.The bonding force between the hot-dip coating and steel wire was determined by the winding test.The corrosion resistance of the steel wire hot-dip coating was tested by the electrochemical workstation.The hot-dip Zn-5A1 alloy coating has a corrosion-resistant structure composed of a zinc-rich phase and an aluminum-rich phase.Due to the enhanced bonding force,the micro structure of the hot-dip coating of the Zn-5A1 alloy with rare earth elements is more compact and uniform than that without rare earth elements.The addition of rare earth elements improves the corrosion resistance of Zn-5A1 alloy coated steel wire.Due to the rare earth segregation,which prevents the corrosion of the grain boundary and enhances the anti-intergranular corrosion performance,steel wire exhibits the optimum corrosion resistance when the content of rare earth elements is 0.08 wt.%.

基于多任务学习多目标优化的稀土元素组分含量与浓度多维度软测量

稀土混合萃取溶液中各元素组分含量的在线软测量是优化连续萃取生产过程、确保产品高纯化的前提.现有软测量方法可独立求解单个稀土元素组分含量,但忽略了多元素组分含量间或组分含量与其它相关因素(如浓度)间的共性.本文为探索多稀土元素组分含量间及组分含量与浓度间的共性,将多任务学习方法用于稀土元素组分含量软测量中.首先,构建多任务深度神经网络,提高模型的泛化能力和鲁棒性.其次,提出基于多目标优化算法的稀土多元素组分含量预测方法,通过搜索Pareto最优以提升各任务的预测精度.经多组对比实验表明,该方法在多元素组分含量或多元素组分含量与浓度同时训练时性能最佳,能满足稀土元素组分含量在线检测的精确性和实时性.

基于卷积神经网络算法判定稀土总量滴定终点技术研究

由于稀土产业发展迅速,市场需求越来越大,应用范围越来越广,需建立一个长期有效且适用于测定各类稀土的方法。针对目前传统稀土总量检测方法效率低、准确度低、滴定终点差异大等问题,难以满足在线检测的需要。卷积神经网络算法通过摄像头记录样品在滴定过程中溶液的颜色变化,对溶液进行图像特征提取和学习,从而有效、准确地实现化学反应过程中溶液颜色的自动化识别,配合步进电机和注射泵等部件实现自动滴定过程,建立了一种基于卷积神经网络的稀土总量浓度在线分析方法。图像识别本质上是对图像信息进行特征提取,而卷积神经网络(CNN)有着传统识别方法不具备的优点,比如能够自行训练、识别速度更快、所需特征更少等,将自动滴定与神经网络相结合,实现了滴定流程的自动化和样品前处理、滴定、终点判定等过程的一体化,且设备内可同时进行五个样品滴定,提高了滴定效率。

REE Contents in Topsoil on the Loess Plateau and Effects of REE Applying on Crops

Eight rare earth elements(REE) in 104 topsoil samples collected from the Loess Plateau has been determined by using of INAA. The regional distribution of REE contents and its effectiveness of REE applying on crops in this area are discussed. The preliminary results showed that the mean concentration of most REE in this area approaches to those of China as a whole; and REE content in main type of soil of the Loess Plateau is as follows: gray cinnamon forest soil, Lou soil(stratified old manured loessal soil)> Heilu soil(dark loessal soil) . loessal soil>sierozem, and similar with the distribution pattern of soil clay contents in the soil which decrease gradually from southeast to northwest. This area may be suitable to apply REE in agriculture production.

Improvement of the microstructure and magnetic properties of(La,Ce)-Fe-B nanocrystalline ribbons

The obstacles of inferior microstructure and poor magnetic properties of(La,Ce)-Fe-B type magnets stand in the way of expanding their applications.In this work,(La,Ce)-Fe-B melt-spun ribbons were prepared with different La/Ce ratios and rare earth contents.The microstructure and magnetic properties of these ribbons were investigated.With the La/Ce ratio increased,the fraction of the CeFe_(2)phase decreased and the β-La phase segregated among matrix grains,which deteriorated the coercivity of the ribbons.Reducing the rare earth content effectively suppressed the β-La phase and improved the proportion of the matrix phase in the ribbons.Here,the optimal alloy composite(La_(15)Ce_(85))_(14)Fe_(80)B_(6) has been obtained with modified microstructure and inhibited secondary phases.Good ribbon performance was tested,with H_(cj)=4.51 kOe,B_(r)=6.29 kG and(BH)_(max)=6.81 MGOe.

基于深度机器视觉的稀土元素组分含量预测

稀土萃取过程中元素含量的快速检测有助于提高稀土产品质量,目前的检测仪器存在检测延时长和维护成本高等问题。基于部分稀土元素独特的颜色特征,如Pr离子、Nd离子,可将机器视觉方法用于元素组分含量的软测量。不同于传统机器视觉方法,本文首次引入卷积神经网络(CNN)提取Pr/Nd混合溶液原始图像抽象表征,同时采用深度神经网络构建回归模型,用于预测Pr/Nd混合溶液中各元素的含量。实验选取1210张Pr/Nd混合溶液图像作为实验数据,相较于已有方法,数据规模提升近12倍。多次独立重复结果表明,预测的组分含量与真实组分含量间的最大相对误差绝对值为2.7738%,满足实际萃取生产中对元素含量分布变化的精度要求,具有一定的实际意义。

2022年珠海市150份水产品中稀土元素的含量水平调查

目的:调查珠海市市售水产品中稀土元素的含量水平。方法:随机采集珠海市5个功能区的农贸市场、商超等共150份水产动物样品,采用微波消解-电感耦合等离子体质谱法测定样品16种稀土元素含量。结果:2022年珠海市售水产品中稀土元素总量为0.01~6679.64μg·kg^(-1),平均含量为176.2μg·kg^(-1),中位值为6.0μg·kg^(-1)。不同类别的水产品种稀土元素含量变化大致为双壳类>腹足、头足类>甲壳类>鱼类。以轻稀土元素富集为主,以轻稀土元素铈(Ce)平均含量最高、其次为镧(La)、钕(Nd),重稀土元素钇(Y)较高。结论:珠海市部分地区及种类水产品中稀土元素相对含量较高,建议加强监管力度。

ICP-AES法测定铁矿石及炉渣中镧、铈、镨、钕、钐

文章研究了铁矿石、炉渣的溶解方法,采用盐酸初步溶解再用碳酸钠-硼酸混合熔剂处理残渣的分解方法,通过试验对镧、铈、镨、钕、钐元素的分析谱线进行优化选择,确定了灵敏度高、干扰少的分析谱线作为分析谱线,采用等离子体发射光谱法同时测定镧、铈、镨、钕、钐五元素含量。结果表明,铁矿石、炉渣中镧、铈、镨、钕、钐元素的检出下限均为0.0050%,铁矿石中各稀土元素测定上限:La测定上限为5%,Ce测定上限为10%,Pr测定上限为1%,Nd测定上限为2%,Sm测定上限为1%;炉渣中各稀土元素测定上限均为0.500%。

稀土火法冶炼回收料中稀土总量的测定

稀土火法冶炼回收料是指稀土熔盐电解冶炼工艺产生的电解渣及金属热还原产生的热还原渣。该类物料经除杂加工后可用于再次提取稀土等有价元素,其中稀土总量是其确定回收利用价值的主要依据。试验采用硝酸-过氧化氢-高氯酸分解,氨水沉淀稀土,继以用盐酸溶解氢氧化物沉淀,在pH 1.5~2.0的酸性条件下,加草酸沉淀稀土,分离铁、铝等。将草酸稀土于950℃灼烧成稀土氧化物,称其质量,计算稀土总量。当样品含重稀土元素镝、钬、铒、铥、镱、镥、钇含量时,需用电感耦合等离子发射光谱法测定草酸沉淀滤液中稀土量,对结果进行补正。标准加入回收率为99.41%~103.10%,统一样品测定结果的相对标准偏差(n=11)为0.14%~0.46%,该方法准确可靠。

白云鄂博矿选矿废水处理技术研究及应用

白云鄂博氧化矿资源综合利用生产线磨矿细度达-0.025 mm75%,多金属矿阶段综合回收导致选矿废水中浮选药剂残留量大,为了改善废水水质,针对选矿废水中悬浮物含量高的情况,采用聚合硫酸铁0.175 mg/L—聚丙烯酰胺0.054 mg/L的加药方式,加药后每步搅拌10 s,水中悬浮物含量由90 000 mg/L降低到200 mg/L以下,水质得到改善,推进了全流程顺利运行,澄清后水质在保持指标稳定的情况下,可降低铁浮选捕收剂22.7%,降低萤石浮选捕收剂13.26%、抑制剂27.14%,水玻璃和硫酸药剂量均降低115.81%。

风化型稀土矿电阻率影响因素分析及模型构建

为了探明影响风化型稀土矿电阻率的主要因素及构建电阻率模型,采用室内电阻率试验和响应曲面法相结合的研究手段,分析含水率、饱和度、干密度、孔隙比以及颗粒级配对电阻率的影响规律,建立主要影响因素与电阻率的关系方程,并探讨其合理性和可靠性。结果表明:含水率、孔隙比和颗粒级配是影响风化型稀土矿电阻率的主要因素,且影响敏感程度顺序为含水率>孔隙比>颗粒级配;其中矿样电阻率随含水率增大而减小,当含水率≤10%时,电阻率下降幅度最大,而当含水率大于10%时,电阻率下降趋势变缓,并逐渐趋于稳定;矿样电阻率随孔隙比增加而增大;矿样电阻率随分形维数增加而减小,且当含水率较大时,分形维数对其影响程度较小。通过多元二次回归分析构建的风化型稀土矿电阻率模型,理论数据与试验数据基本吻合,具有较高的合理性和可靠性,在含水率23.67%、孔隙比1.73、分形维数2.47条件下的电阻率最小。