【作者】 夏法锋；

【导师】 贾振元； 吴蒙华；

【作者基本信息】 大连理工大学 ， 机械电子， 2009， 博士

【摘要】 复合电沉积技术是在镀液中加入不溶性微米或纳米固体粒子（如TiN、AlN、SiC和Al₂O₃等）,通过电沉积的方式,使固体粒子与基质金属离子共同沉积的一种方法。纳米复合镀层由于具有良好的光学非线性、磁性以及机械性能（如高强度、高硬度和良好的耐磨性能）,近年来成为广泛研究的热点。目前,纳米复合镀层的研究还处于探索阶段,纳米粒子和基质金属晶粒的共沉积理论还不完善;纳米复合镀层的制备工艺、纳米粒子在镀液和镀层中的均匀分散问题以及纳米粒子在镀层中的作用机制等方面的研究还显得比较薄弱,严重制约着有关纳米复合镀层的进一步研究和应用。本论文以镍基TiN纳米复合镀层为研究对象,采用理论分析与实验研究相结合的方法,对超声-电沉积镍基TiN纳米复合镀层的制备工艺和主要性能进行研究,并利用BP神经网络对镍基TiN纳米复合镀层的粒子复合量和显微硬度建立网络预测模型。本论文从镍基TiN纳米复合镀层的制备机理入手,首先采用超声-电沉积法制得镍基TiN纳米复合镀层。研究分析了超声波功率、脉冲电流密度、脉冲占空比、TiN粒子悬浮量以及表面活性剂对镍基TiN纳米复合镀层显微组织、显微硬度以及镀层中TiN粒子复合量的影响。同时,利用正交试验法对超声-电沉积镍基TiN纳米复合镀层的工艺参数进行优化。实验结果表明,超声-电沉积的主要工艺参数对镍基TiN纳米复合镀层显微硬度和TiN粒子复合量的影响主次顺序为:纳米TiN粒子的悬浮量＞超声功率＞电流密度＞占空比＞表面活性剂浓度。采用超声-电沉积方法制备镍基TiN纳米复合镀层的最佳工艺参数为:镀液温度50℃,pH=4.5,复合电沉积时间50min,TiN粒子悬浮量8g/l,电流密度5A/dm²,占空比40%,表面活性剂浓度100mg/l,超声功率200W。其次,本论文首先简述了超声波在介质中的作用机理。然后,研究了不同的搅拌方式对镍基TiN复合镀层表面形貌、显微组织结构、镀层中TiN粒子复合量以及镀层磨损量的影响,探讨了超声波在电沉积镍基TiN纳米复合镀层中的作用。并对镍基TiN纳米复合镀层进行表征。研究结果表明,超声波在电沉积镍基TiN复合镀层过程中,具有明显的搅拌分散作用、强化促进作用以及细化TiN粒子和镍晶粒作用等。XRD分析表明,在机械搅拌-电沉积制得的复合镀层中,镍晶粒和TiN粒子的平均粒径分别为119.3nm和56.7nm:而在超声-电沉积制得的复合镀层中,镍晶粒和TiN粒子的平均粒径分别为74.6nm和34.8nm。对镀层表面XPS分析表明,在镍基TiN纳米复合镀层表面存在零价态的镍和Ti元素。AFM分析表明,电沉积镍镀层的表面均方根粗糙度、平方粗糙度和最大高度粗糙度分别为Rms=48.213nm、Ra=39.567nm和Rmax=339.28nm;机械搅拌-电沉积镍基TiN纳米复合镀层的表面均方根粗糙度、平方粗糙度和最大高度粗糙度分别为Rms=27.427nm、Ra=21.857nm和Rmax=174.74nm;超声-电沉积镍基TiN纳米复合镀层的表面均方根粗糙度、平方粗糙度和最大高度粗糙度分别为Rms=19.242nm、Ra=15.719nm和Rmax=125.53nm。这说明超声波对纳米粒子的搅拌分散作用明显超过机械搅拌所能达到的剧烈程度。为了研究镍基TiN纳米复合镀层的主要性能,分别采用机械搅拌-电沉积法和超声-电沉积法制得镍基TiN纳米复合镀层和镍镀层。利用显微维氏硬度计、涂层附着力划痕仪、腐蚀法、中性盐雾试验法（NSS）以及磨损试验机对镍基TiN纳米复合镀层和镍镀层的显微硬度、结合力、孔隙率、耐腐蚀性能和耐磨性能等进行对比研究分析。在此基础上,探讨了镍基TiN纳米复合镀层的耐腐蚀性和耐磨性机理。研究结果表明,镍基TiN纳米复合镀层不仅具有较高的显微硬度、耐腐蚀和耐磨损性能,还具有较低的孔隙率和摩擦系数。其显微硬度大约为镍镀层的2倍左右;其孔隙率约为镍镀层的1/3左右;其磨损量为镍镀层4倍左右。但其结合力与镍镀层相差不大,结合力都在60～70N之间。耐腐蚀性能实验表明,试样的平均腐蚀速率大小依次顺序为:45钢＞机械搅拌-电沉积镍镀层＞超声-电沉积镍镀层＞机械搅拌-电沉积镍基TiN纳米复合镀层＞超声-电沉积镍基TiN纳米复合镀层。最后,利用BP神经网络对超声-电沉积制备镍基TiN纳米复合镀层的TiN粒子复合量和显微硬度进行建模预测。预测结果表明,所建BP网络预测模型的预测误差相对较小,最大相对误差为5.43%。说明BP网络预测模型的拟合效果较好,准确率较高。本BP网络预测模型的建立,为试验数据的处理提供了一种新途径。更多还原

【Abstract】 Composite electrodeposition technology is a method to obtain composite coatings by adding insoluble micrometer- or nanometer-sized solid particles（such as TiN,AIN,SiC, Al₂O₃ and so on） to the electrolyte to co-deposit the particles and metal matrix with electrodeposition.In recent years,nanocomposite coatings have become the focus of widespread research owing to favorable properties such as optical non-linearity,magnetism and mechanical performance such as high hardness,wear resistance and corrosion resistance. At present,the research of nanocomposite coatings is still in the exploratory stage,and the codeposition theory of nanoparticles and the matrix grains is not perfect.The study on such as the preparation of nanocomposite coatings technology,the evenly dispersed issue of nanoparticles in the bath and coating,and the role of mechanisms of nanoparticles in coatings is weak relatively.This will restrict the further study and application of nanocomposite coatings seriously.In this thesis,the technological preparation and the main performance of Ni-TiN nanocomposite coatings are studied by combining theoretical analysis with experimental verification,the forecast models for TiN particles content and microhardness of Ni-TiN nanocomposite coatings are established by using the BP neural network.Firstly,based on the preparation of Ni-TiN nanocomposite coatings mechanism,the Ni-TiN nanocomposite coatings are prepared by ultrasonic electrodepostion,then the effects of ultrasonic power,current density,duty cycle,TiN nanoparticles concentration and surfactants on the surface morphologies,microhardness and TiN particles in coatings are studied. The technological parameters of Ni-TiN nanocomposite coatings prepared by ultrasonic electrodeposition are optimized with the orthogonal test.The constituent,surface morphologies and surface roughness of Ni-TiN nanocomposite coatings are characterized.The results show that the technological parameters of Ni-TiN nanocomposite coatings prepared by ultrasonic electrodeposition have great effects on the microhardness and TiN particles of coatings.The order from large to small is:TiN nanoparticles concentration＞Ultrasonic power＞Current density＞Duty cycle＞Surfactants concentration.And the optimal parameters are as follows: Temperature 50℃,pH 4.5,Time 50min,TiN nanoparticles concentration 4g/l,Current density 5A/dm²,Duty cycle 40%,Surfactants concentration 80mg/l,ultrasonic power 200W. By using XPS,TEM,EPMA and AFM,the Ni,Ti and N elements are investigated systematically on the surface of Ni-TiN nanocomposite coatings prepared by ultrasonic electrodeposition method,The Ni crystal and nano TiN particles are measured as approximately 55 nm and 35 nm.Moreover,the surface of Ni-TiN nanocomposite coatings is evener,the fluctuation is smaller and the exterior protuberance is also small,and the roughness of the coating are Rms =19.242 nm,Ra =15.719 nm and Rmax=125.53 nm.And they are smaller than that of Ni-TiN nanocomposite coatings prepared by mechanical stirring electrodeposition obviously.Secondly,the effect mechanism of the ultrasonic in media is outlined.The effects of different stirring ways on the surface morphologies,microstructure,the TiN particles content and the abrasion of Ni-TiN nanocomposite coatings are studied.The role of ultrasonic is discussed in the Ni-TiN nanocomposite coatings prepared by the electrodeposition method. Then the Ni-TiN nanocomposite coatings are characterized.The results show that the ultrasonic has obvious stirring scattered role,strengthen and promoting role and refining the nickel grains and particles TiN role.The XRD shows that the average sizes of nickel gains and TiN particles are 119.3nm and 56.7nm,separately,in the Ni-TiN nanocomposite coatings prepared by mechanical stirring electrodeposition.While the average sizes of nickel grains and TiN particles are 74.6nm and 34.8nm,separately,in the Ni-TiN nanocomposite coatings prepared by ultrasonic electrodeposition.The XPS indicates that there are nickel and Ti element in the Ni-TiN nanocomposite coatings surface.The AFM shows that the surface roughness of the nickel coating prepared by electrodeposited method are Rms = 48.213nm, Ra = 39.567nm and Rmax = 339.28nm.The surface roughness of the Ni-TiN nanocomposite coatingprepared by mechanical stirring electrodeposited method is Rms=27.427nm, Ra=21.857nm and Rmax=174.74nm.And the surface roughness of the Ni-TiN nanocomposite coating prepared by ultrasonic electrodeposited method is Rms=19.242nm, Ra=15.719nm and Rmax=125.53nm.This shows that the mixing nanoparticles by the ultrasonic dispersed more than that by mechanical agitation to achieve the degree of dramatic.In order to study the main performance of the coatings,the Ni-TiN nanocomposite coatings and nickel coatings are prepared by mechanical stirring electrodeposition and ultrasonic electrodeposition.The microhardness,binding force,porosity,the anti-corrosive and wear resistant performances are studied by using Vickers microhardness microscope,scratch tester, corrosive test,NSS and wear testing machine.On this basis,the corrosion resistance and wear resistance mechanism of Ni-TiN nanocomposite coatings are discussed.The result indicates that the Ni-TiN nanocomposite coatings not only have the high microhardness,anti-corrosive and the wear resistant performances,but also have lower porosity and the friction coefficient. The microhardness of the coatings is about 2 times for nickel coating;the porosity approximately 1/3 for about nickel coating,and the abrasion is about 1/4 for nickel coating. But the binding force is close to that of nickel coating between 60～70N.The order of the specimen average corrosion rate is:45 steel＞Nickel coating prepared by mechanical stirring electrodeposition＞Nickel coating prepared by ultrasonic electrodeposition＞Ni-TiN nanocomposite coating by mechanical stirring electrodeposition＞Ni-TiN nanocomposite coating prepared by ultrasonic electrodeposition.Finally,the forecast models of the TiN particles content and microhardness of the Ni-TiN nanocomposite coatings are established by using the BP neural network.The results indicate that the prediction error of the BP network models is small relatively,and the largest relative error is 5.43%.The fitting effect of the BP network forecast model is explained perfectly,the rate of accuracy is high.The establishment of the BP network forecast models has provided a new way for processing the tentative data.更多还原