不同刺激条件下人外周血淋巴细胞超微结构与粘滞力的AFM分析

应用原子力显微镜（atomic force microscopy,AFM）探测了静息、脂多糖（LPS）或伴刀豆蛋白（ConA）活化的人外周血淋巴细胞的形态结构、超微结构及粘滞力特性。从AFM图像可知,经LPS或ConA刺激活化后的淋巴细胞比静息状态的淋巴细胞有所增大,并且表面分布着大小不一的颗粒状聚合物。利用AFM高空间分辨的力位移曲线测量系统,发现经LPS或ConA刺激活化后淋巴细胞的粘滞力是静息状态淋巴细胞的2~3倍。通过AFM探测淋巴细胞活化状态的可视化数据,可以更好地理解淋巴细胞的行为。     

香花石形态所反映的一种可能的孪晶结构

香花石是我国发现的第一个新矿物,它的形态非常复杂.本文基于香花石形态上出现了左形与右形相聚的现象,对香花石形态进行了重新思考,认为同一单形的左形与右形不能相聚形成单晶体,只能形成孪晶,因此,对原香花石晶体图进行了一些修正,认为原香花石晶体图可能是一个孪晶图.这对深入认识晶体形态及晶体对称理论、澄清长期以来人们并没有足够重视的一些基本概念有重要意义.     

CTAB与SDBS对碳酸锶粒子生长形态调控及机理研究

以表面活性剂CTAB和SDBS为化学添加剂,采用化学共沉淀法对碳酸锶晶体的生长形态进行调控,成功地制备出了实心的树枝状和花瓣为空心的花状碳酸锶粉体,并用X射线衍射(XRD)、扫描电子显微镜(SEM)和傅里叶变换红外光谱(FT-IR)等分析手段对样品进行了表征;最后重点对化学添加剂可能产生的影响机理进行了初步的探讨.结果表明,CTAB和SDBS在晶体生长的过程中能起到显著的影响作用,两者对粒子分散性能的作用效果相反,而且后者对晶体(013)和(213)晶面表面能降低的贡献明显大于前者.     

链状Silicalite-1分子筛的合成

在传统的合成体系中, 加入聚电解质聚二烯丙基二甲基胺盐酸盐(PDDA)自组装合成链状的Silicalite-1分子筛材料, 考察了PDDA的加入对产物的影响, 并对其进行了XRD和SEM表征.     

基于数学形态学的表面原子熔融相的 STM图像识别算法

高温扫描隧道显微镜(HT-STM)可以实时原位地捕捉到表面原子的熔化相变过程.在这一原位变温实验中,快速可靠地识别出每帧STM图像中的熔融相十分关键.传统的手工统计方法存在效率低下、随意性大等问题.我们发展出一套基于数学形态学的算法,来自动快速地识别.与人工方法相比,该算法消除了人为主观误差,使确定的边界更加准确、光滑,处理效率提高了266倍.     

A Simple Hydrothermal Method for the Growth of ZnO Crystals

ZnO microstructures have been grown from zinc chloride (ZnCl₂) and ammonia solution at 100 °C for 1 – 24 hours. X‐ray diffraction, scanning electron microscope and field‐emission scanning microscope were utilized to investigate the structural properties and morphology of the ZnO crystals. Structural investigations show that phase‐pure hexagonal structure ZnO has been successfully synthesized, and the hexagonal structure ZnO can be achieved in solutions with an appropriate range of concentrations. Under our experimental conditions, several different morphologies of ZnO structures were obtained, including flower‐like and bar flower‐like. The relationship between the morphology and experimental conditions are discussed.     

基于形态学水线区域的深度图像分割

深度图像分割是基于特征关系图匹配的曲面物体识别中的关键技术之一,针对已有深度图像分割方法存在的问题,提出了一种基于二值形态学的水线区域增长算法对深度图像进行分割。首先用结构元素迭代腐蚀深度图像形成距离图像;然后根据距离图像计算极限腐蚀的集合,提取出目标的种子点;最后用条件粗化从种子点开始生长回到原尺寸但不使各区域相连,完成深度图像的分割。实验证明,算法的分割结果与人的主观视觉感知具有良好的一致性。     

基于数学形态学的弱点状运动目标的检测

提出了一种新的基于数学形态学的红外图像序列中弱点状运动目标的非参数检测算法。采用数学形态学抑制背景杂波干扰和增强目标,用沿时间轴投影和二维空域搜索代替复杂的时空三维搜索形成组合帧,然后在每条可能的轨迹上将进行目标能量累加,实现了一种快速检测前跟踪(TBD)检测算法。仿真实验表明:在恒虚警概率条件下,该检测算法能高效地检测信噪比约为2的弱点状运动目标,检测性能对噪声分布不敏感,能精确地得到目标的即时位置和速度信息,适合于实时图像处理和目标探测,具有很高的实用价值。     

Predicting Polymorphism of Electrospun Polyvinylidene Fluoride Membranes by Their Morphologies

Although electrospinning of polyvinylidene fluoride (PVDF) has been studied for more than 10 years, the crystalline phase differentiation of the electrospun mats is still normally through the combination of different characterization techniques, and the relationship between polymorphism and morphology of the fibers in electrospun PVDF membranes has never been reported. Here, we show their close relationships by conducting room-temperature electrospinning experiments on various polymer/solvent systems. The electrospun membranes full of bead-free fibers have a very high fraction of β-phase, F(β), over 90%, and high orientation, whereas the membranes comprising beads and/or a large number of beaded fibers most often result in a low fraction of β-phase (F(β) normally below 50%) and low orientation. On the other hand, electrospun membranes consisting of both bead-free fibers and a very limited number of beaded fibers showed a medium high fraction of β-phase, F(β) more than 70% but less than 90%. These findings suggest the feasibility of intuitively predicting the crystalline phase of electrospun PVDF membranes directly by their morphologies, which is obviously simple, inexpensive and convenient for future investigations.