Ultrafast cross‐sectional imaging of gas‐particle flow in a fluidized bed
Ultrafast cross‐sectional imaging of gas‐particle flow in a fluidized bed
复制标题
流化床中气体-颗粒流的超快横截面成像
DOI：
10.1002/aic.12121
复制
发表时间：
2009
期刊：
影响因子：
3.7
通讯作者：
U. Hampel
中科院分区：
文献类型：
--
作者：
M. Bieberle;F. Fischer;E. Schleicher;H.;H. Mayer;U. Hampel
气固流化床广泛应用于化学和加工工程，例如流化床催化裂化、颗粒固体干燥、聚烯烃生产、煅烧、煤气化，以及最近的碳纳米管生产和富氧燃烧。流化床中的气体-颗粒流动通常复杂且难以观察，但迫切需要有关空隙分布和固体传输的准确信息来评估、监测和优化流化床操作。通常，流化的特征在于气液流动中的流态。因此，气泡流、湍流流和环形流是已知的。气固分布及其动力学强烈影响运行过程的性能。例如，气态反应物的转化取决于气泡尺寸和气泡内的固体含量。因此，需要在给定的流动机械和热力学条件下恢复流化床的内部结构和动力学，这些条件由质量流量、压力、颗粒尺寸、床几何结构等定义。为此需要测量技术，以高空间和时间分辨率提供动态空隙率信息。目前，流量测量技术很多，但适用于密相流化床的流量测量技术很少。 1 光学成像、基于激光的方法和超声技术通常无法揭示流化床内的空隙率分布，因为颗粒对可见光不透明并且不可预测地散射超声波。这种测量技术最多能够捕获固相的外围结构。相反，局部光学和电容探头2-4被广泛用于以高时间分辨率测量流化床中选定点的局部相分数。然而，它们没有给出空隙分布的横截面视图，甚至在相当程度上干扰了流动本身。更合适的是电容断层扫描(ECT)，5 已成功应用于流化床。 ECT 的时间分辨率高达每秒 1000 帧 (fps)，但空间分辨率仅限于管道直径的 10% 左右。因此，它既不能可视化单个颗粒，也不能可视化小空隙体积。正电子发射断层扫描6是另一种成像技术，能够在约1秒的时间范围内对示踪粒子的分布进行成像，空间分辨率约为5毫米。放射性粒子跟踪方法7提供了更好的时间信息，可用于测量单个粒子的速度和轨迹。但粒子的空间分布仅作为时间平均值获得。磁共振成像8 (MRI) 已适用于流化床中轴向固体含量的超快测量，其中时间分辨率达到 1 至 2 毫秒。传统的 X 射线和伽马射线计算机断层扫描可以达到毫米范围的空间分辨率，但速度较慢，因此只能测量时间平均密度分布。 9 近年来，人们提出了一些提高时间分辨率的建议。
Gas–solid fluidized beds are widely used in chemical and process engineering, for instance in fluid catalytic cracking, drying of particulate solids, polyolefin production, calcination, coal gasification, and more recently for carbon nanotube production and oxycombustion. The gas-particle flow in fluidized beds is generally complex and difficult to observe, but exact information on voidage distribution and solid transport is urgently needed for assessment, monitoring, and optimization of fluidized bed operation. Commonly, fluidization is characterized by flow regimes as in gas–liquid flow. Thus, bubbly, turbulent, and annular flow is known. The gas–solids distribution and its dynamics strongly influence the performance of the running process. As an example, the conversion of gaseous reactants depends on bubble sizes and solids content inside the bubbles. It is thus desirable to recover the inner structure and dynamics of fluidized beds at given flow mechanical and thermodynamic conditions, defined by mass flow rates, pressure, particle sizes, bed geometry, etc. For this purpose measurement techniques are needed, which provide dynamic void fraction information at high spatial and temporal resolution. Currently, there are many flow measurement techniques but only few are applicable to dense fluidized beds. 1 Optical imaging, laser-based methods and ultrasound techniques generally fail to disclose the voidage distribution inside a fluidized bed, since the particles are opaque to visible light and scatter ultrasound waves unpredictably. Such measurement techniques are at the most able to capture the peripheral structure of the solids phase. Instead, local optical and capacitance probes2–4 are widely used to measure local phase fractions at selected points in the fluidized bed with high temporal resolution. However, they give no crosssectional view on the voidage distribution and even disturb the flow itself to a considerable degree. More suitable seems electrical capacitance tomography (ECT), 5 which has already been successfully applied to fluidized beds. ECT reaches high temporal resolution of up to 1000 frames per second (fps), but is limited in spatial resolution to about 10% of the pipe diameter. It can therefore neither visualize single particles nor small void volumes. Positron emission tomography6 is yet another imaging technique which has the capability to image the distribution of tracer particles in the temporal range of about 1s and with spatial resolution of about 5mm. Radioactive particle tracking methods7 provide even better temporal information and can be used to measure velocities and trajectories of single particles. But the spatial distribution of particles is obtained as a time-average only. Magnetic resonance imaging8 (MRI) has been adapted for ultrafast measurements of the axial solids content in a fluidized bed and therein reaches a temporal resolution between 1 and 2ms. Conventional X-ray and gamma ray computed tomography can reach a spatial resolution in the millimeter range, but are slow and can therefore only measure timeaveraged density distributions. 9 In recent years, some proposals have been made to increase the time resolution of