【作者】 穆春丰；

【导师】 刘天庆；

【作者基本信息】 大连理工大学 ， 化学工程， 2008， 博士

【摘要】 滴状冷凝初始液滴形成机理一直是悬而未决的问题。从形成新相的热力学计算可知,初始液核的尺寸应该在纳米尺度,所以要彻底解决这个难题,必须在纳米尺度上对初始冷凝过程进行研究。由于目前的仪器设备尚不能在线观察到纳米尺度液滴初始形成的过程,所以本文采用间接的方法来实现这个过程。以能与热水(冷凝液)反应的镁表面作为冷凝表面,在实验中通过控制过冷度和冷凝时间来实现水蒸汽在镁表面的初始冷凝。由于镁与冷凝液反应后必然会留下产物“痕迹”,可以事后应用电子探针和扫描电镜分析冷凝前后试件表面化学成分及其分布的变化,用于推断蒸汽的初始冷凝状态是核状还是薄膜状。本文首先以机械抛光和磁控溅射两种方法制备镁表面,用原子力显微镜检测两种方法制备的冷凝材料表面的形貌和粗糙度,结果表明机械抛光表面粗糙度在几百纳米左右,磁控溅射所得镁膜表面平均粗糙度小于23nm。通过电子探针检测磁控溅射制备的镁膜厚度为21μm,这说明衬底完全被镁覆盖。由于初始液滴的尺寸在纳米级,所以对冷凝材料表面的光洁度要求很高,因此,磁控溅射法制备的镁膜适宜作为本实验用的冷凝表面,并能满足冷凝实验后电子探针面扫描的要求。进而在初始冷凝实验中,通过控制过冷度和冷凝时间来实现镁表面初始冷凝液核的形成,然后利用电子探针和扫描电镜进行检测,结果显示初始冷凝后镁表面上氧元素的含量随着过冷度和冷凝时间的增加而增加,并且氧元素在表面上的分布是不均匀的。而且电子探针和扫描电镜两种方法所检测的得结果基本吻合,充分说明了本研究方法和结果的可靠性,即凝液与镁表面反应所留下的“痕迹”能真实的反应初始凝液的形成状态。因此,本研究的结果一致表明,初始凝液只在部分区域、而不是在整个冷凝表面形成,并且得到初始液核的尺寸在3到10纳米,从而首次在纳米尺度上证实滴状冷凝初始液滴形成的机理符合固定成核中心假说。之后,鉴于滴状冷凝表面的传热性能与表面几何结构特征有密切联系这一事实,而前人尚未研究滴状冷凝初始液核数与材料表面形貌之间的定量关系这一现状,本文进一步利用扫描电子显微镜拍摄不同磁控溅射条件下制备的镁膜表面的形貌图象,运用分形理论对镁膜表面冷凝前表面形貌的复杂程度进行量化研究,选用差分计盒维数法计算镁膜表面冷凝前表面形貌的分形维数。然后在相近的冷凝条件下,通过控制过冷度和冷凝时间而实现水蒸汽在这些镁膜表面的初始冷凝,再应用电子探针分析冷凝前后试件表面化学成分的变化,运用数字化图象处理技术对冷凝后镁膜表面上的氧元素分布图进行分析,从而获得表面上的液滴成核密度,并关联出冷凝材料表面形貌特征与滴状冷凝成核中心密度之间的定量关系式。关联结果表明,冷凝试件的表面形貌特征对滴状冷凝的成核密度有较大影响,其分形维数不同,冷凝后镁表面初始液核数也不同。同时理论和实验结果均表明,冷凝表面的分形维数越大,相应的滴状冷凝热通量也越大。最后,针对液固界面相互作用对滴状冷凝传热的影响,以液滴数平衡概念为基础,考虑表面形貌和表观接触角对冷凝传热的作用,建立了新的滴状冷凝传热模型。模型计算结果表明,表面形貌和表观接触角对滴状冷凝传热均有很大影响。在相同过冷度下,表面分形维数越大,成核中心密度越大,热通量也越大,随着过冷度的增加,分形维数对成核中心密度和热通量的影响更加明显;相同半径液滴的传热速率随接触角的改变而有明显变化,并且均存在适宜接触角使该尺度的液滴传热量达到最大。从单个液滴传热分析得到,对于半径小于1微米的单个液滴,其适宜接触角随着液滴半径的增加而略有减小,但变化极不明显,均在120度左右。液滴半径等于1微米时,适宜接触角为119.1度。但当液滴半径大于1微米后,对应的适宜接触角开始随液滴半径的增加而明显降低。对于考虑了表面液滴脱落的动态滴状冷凝过程,整个冷凝表面的热通量仍随接触角而变化,并且对应热通量最高的适宜接触角为87.6度。本文从传热学角度分析和证明了滴状冷凝传热过程并不是接触角越大越好。更多还原

【Abstract】 The mechanism of the formation of initial condensate droplets for dropwise condensation is still in suspense. To solve the problem, we must understand if the initial condensation is in nucleus or in thin film in nanometer scale, since the calculation of the thermodynamics for new phase formation indicates that the size of the initial condensate nucleus is in nano-scale. Magnesium was applied as the condensation surfaces in this study since it can react with hot water (condensate) and thus leave marks of initial condensate state on the surface. In the experiment, the subcooling and reaction time were controlled to realized the initial condensation on the surfaces, and then an electron probe microanalyzer (EPMA) and scanning electron microscope(SEM)were used to scan the variations of the chemical compositions as well as their distribution on the surfaces before and after the initial condensation. These consequences were used to deduce whether the initial condensate state is in nuclei or in film.In this paper, mechanically polishing and magnetic-control sputtering (MCS) were used firstly to prepare the magnesium surfaces. The topography and surface roughness of the magnesium surfaces were characterized with the atomic force microscope and the results show that the average roughness of the mechanically polishing surface is about 100nm, while that of the MCS surface is about 23nm. The thickness of the magnesium film plated with MCS method was also measured with the electron probe microanalyzer, and the thickness of the plated film is 23μm, which meant that silicon substrates were covered by magnesium completely. So the films of magnesium were feasible for this condensation experiment, and they can meet the requirements of surface scan by electron probe microanalyzer.In the followed experiment, the initial condensation process was realized by controlling the subcooling and reaction time, and then, EPMA and SEM were used to scan the variation of the chemical compositions on the magnesium surfaces. The results showed that the oxygen contents on the test surfaces increased with subcooling and condensation time obviously after the initial condensation, and the oxygen on the test surface distributed non-uniformly. Moreover the results from the scan of EPMA and SEM were nearly identical, which implys that the method presented in the paper is reliable. Therefore, the reaction marks on the surface are the true display of initial condensate state. Meanwhile, the reaction dynamic equation of magnesium and water was set up and used to calculate the area occupied by initial condensate, which was well agreed with the measured results of EPMA. All these consequences indicate that the initial condensate forms in the nucleus state on solid surfaces, not in the state of thin film, and the size of the initinal droplet is 3 to 10nm. This is the first investigation that the initial condensation droplets in nanometer scale were displayed and the mechanism of initial droplet formation for dropwise condensation in nano size was confirmed.Furthermore, in view of the fact that there’s close relation between the geometry structure characteristic of a condensation surface and its heat transfer performance for dropwise condensation, but there is no qutantative relation yet between surface topography and nucleation site density, SEM was used again to obtain the topography photographs of magnesium surfaces with different surface characteristics prepared with distinct MCS parameters. Then fractal theory was applied to describe the irregularity and complexity of magnesium surfaces quantitatively. And the differential box-counting was used to calculate the fractal dimension of these magnesium surfaces before condensation experiment. The initial dropwise condensation on these magnesium surfaces was then achieved by controlling the subcooling and the contacting time between the steam and the magnesium surfaces. The nucleation site density can then be obtained after the oxygen element distribution was analyzed with EPMA and the image processing technology. And the quantitative relation between the nucleation site density and surface fractal dimension was thus correlated.The results show that the surface topography effects the initial dropwise condensation significantly. When the surface fractal dimensions are different, the number of nucleation sites for dropwise condensation on the surfaces is distinct. And both the theoretical and the experimental results indicated that the larger the surface fractal dimension, the greater the heat flux of dropwise condensation.Finally, in consideration of the interfacial interaction between liquid and solid, a new model for dropwise condensation heat transfer was proposed by using the population balance concept and considering the effect of surface topography and contact angle. The calculation results indicate that dropwise concdensation can be influenced considerably by surface topography and contact angle. Under the same subcooling, the greater the surface fractal dimension, the larger the nucleation site density and the higher the heat flux of dropwise condensation. Also the effect of fractal dimension on nucleation site density and heat flux becomes more obvious as subcooling increases. Meanwhile the heat transfer rate through a single droplet increases with contact angle firstly and then decreases, i.e. there is an optimum contact angle to make the heat transfer rate of the droplet maximum. The optimum contact angle is about 120°for the droplets with radius less than 1 micron, and the value almost does not vary with the increase of radius. The optimum contact angle is 119.1°for the droplet with radius 1 microns. When the droplet radius is larger than 1 micron however, the optimum contact angle decreases with the increase of droplet radius obviously. Moreover, the heat flux of dropwise condensation on an entire surface also varies obviously with contact angle when the departure of condensed drops is considered. And the optimum contact angle value becomes 87.6°in this case. Our resutls show that there exists a proper contact angle for dropwise condensation, and it is not true that the larger the contact angle, the higher the heat flux of dropwise condensation.更多还原