GOALI: Toward a Fundamental Understanding of Elutriation in Fluidized Beds

摘要CTS-0318999C。科罗拉多大学科罗拉多分校的赫伦亚说，气固两相流动无处不在，尽管还没有被很好地理解。对固体加工厂的一项研究表明，80%的工厂遇到了固体处理问题，如管道堵塞，这些操作的性能通常只有设计的40%-50%。此外，即使是可靠的催化裂化装置的经验关联式也往往是不充分的。拟议的目标努力旨在促进与一种现象相关的基本知识，这种现象通常是通过经验处理的，即淘洗。淋洗是指在鼓泡的气态流态化床中携带细粉。尽管在过去的20年里，文献中已经提出了50多个淋洗关联式，但实验和经验预测之间的不一致通常超过100%，在某些情况下，可能相差100倍以上。此外，即使是定性的淘洗也可能是违反直觉的。例如，已知向流态化床中添加细粉会增加相对粗颗粒(即，终端速度大于床层运行的表观速度的颗粒)的淋洗率。为了更好地预测淋溶现象，提出了一种模型和实验相结合的方法。人们的注意力将集中在由两个在大小和密度上不同的物种组成的二元系统上。这一选择的动机有两方面：(I)该体系对千禧化学公司(该提议的两个工业合作伙伴之一)特别感兴趣，(Ii)二元体系是多分散体系的最简单形式。建模工作将基于运动学理论类比。初步工作表明，不同粒子之间的能量均分是不存在的，这是现有运动论模型的常见假设。初步计算还表明，非均分产生了尺寸/密度分离的驱动力的额外成分；以前没有记录过这种驱动力的存在。由于不同颗粒在床层和干舷段的分离(或解混)将影响系统的流体动力学(从而影响淋洗)，因此需要对非均匀分配效应的影响进行研究。因此，拟议研究的第一部分将涉及推导双星系统的动力学理论，其中包括非均分(和非麦克斯韦)效应。由此产生的连续体理论将被并入两个CFD(计算流体动力学)程序：标准欧拉-欧拉框架(两相都被视为一个连续体)和由Arena-Flow(本提案的工业合作伙伴)开发的新的多相粒子单元(MP-PIC)公式。后一种形式的一个独特特征是结合欧拉和拉格朗日方法来描述粒子相，这允许在计算上有效地描述具有多种粒子类型的系统。与以前的研究不同，拟议工作的实验部分将涉及在由两个大小和密度不同的B组物种组成的床上进行淋洗测量。将获得每个物种的通量的整体和局部测量结果。将这些模型应用到实验系统中，将是第一次没有经验(即没有可调参数)的淋洗预测。由此产生的模型预测和实验数据之间的比较将表明非均匀分配效应对床面流体动力学的影响，以及欧拉-欧拉和MP-PIC框架在预测淋洗方面的可行性，这对两个工业伙伴都非常感兴趣。这项工作的动机是每个合作伙伴的共同和互补的利益：科罗拉多大学(多分散颗粒系统)、千禧化学公司(淋洗)和Arena-Flow(MP-PIC方法)。拟议的努力旨在确保合作伙伴之间的持续互动和交流，每年举行会议，并延长PI和博士生对两个工业现场的访问。这项工作的更广泛的影响包括：(I)对多分散体系有更基本的了解，这是目前的治疗方法可以推广到两种粒度以上之前所需要的；(Ii)对研究生进行颗粒技术领域的培训，这已被确定为国家需要，以及(Iii)与技术界和学生社区分享所学到的信息(通过纳入课程和外联)。

AbstractCTS-0318999C. Hrenya, University of Colorado, BoulderGas-solid flows are ubiquitous, though not well understood. A study of solids processing plants indicates that 80 percent of the plants experienced solids handling problems such as pipe blockage and that the performance of such operations was typically only 40-50 percent of design. Furthermore, even reliable empirical correlations for fluid catalytic cracking units are often inadequate. The proposed GOALI effort is targeted at advancing the fundamental knowledge associated with a phenomenon that is typically handled empirically, namely elutriation. Elutriation refers to the carryover of fines in a bubbling, gas-fluidized bed. Although more than 50 elutriation correlations have been proposed in the literature over the last two decades, the disagreement between experiments and empirical prediction is typically over 100 percent and in some cases may differ by a factor of more than a hundred. Furthermore, even the qualitative nature of elutriation may be anti-intuitive. For example, the addition of fines to a fluidized bed is known to increase the elutriation rate of relatively coarse particles (i.e., particles which have a terminal velocity that is greater than the superficial velocity at which the bed operates). In order to better predict the elutriation phenomenon, a combined modeling and experimental effort is being proposed. Attention will be focused on a binary system composed of two species that differ in both size and density. The impetus for this choice is twofold: (i) this system is of specific interest to Millennium Chemicals (one of the two industrial partners on this proposal), and (ii) the binary system represents the simplest form of a polydisperse system. The modeling effort will be based on a kinetic-theory analogy. Preliminary work has indicated that an equipartition of energy between unlike particles, which is a common assumption of existing kinetic-theory models, does not exist. Preliminary calculations have also indicated that non-equipartition gives rise to additional components of the driving force for size/density segregation; the existence of such driving forces has not been previously documented. Because the segregation (or de-mixing) of unlike particles in both the bed and freeboard section will impact system hydrodynamics (and thus elutriation), an investigation on the impact of the non-equipartition effects is needed. Thus, the first part of the proposed study will involve the derivation of a kinetic theory for binary systems in which non-equipartition (and non-Maxwellian) effects are included. The resulting continnum heory will then be incorporated into two CFD (computational fluid dynamics) codes: a standard Eulerian-Eulerian framework (both phases are treated as a continuum) and a new multi-phase particle-in-cell (MP-PIC) formulation developed by Arena-Flow (an industrial partner on this proposal). A unique feature of the latter formulation is the combined Eulerian and Lagrangian approach to the particle phase, which allows for a computationally efficient description of systems with multiple particle types. Unlike previous studies, the experimental portion of the proposed work will involve elutriation measurements on beds composed of two Group B species which differ in size and density. Both overall and local measurements of fluxes for each species will be obtained. Application of the models to the experimental system will represent the first elutriation predictions without empricism (i.e., no adjustable parameters). The resulting comparisons between model predictions and experimental data will indicate the impact of non-equipartition effects on bed hydrodynamics, and the viability of the Eulerian-Eulerian and MP-PIC frameworks in predicting elutriation, which is of immense interest to both industrial partners. This work is motivated by the combined and complementary interests of each partner: the University of Colorado (polydisperse particulate systems), Millennium Chemicals (elutriation), and Arena-Flow (MP-PIC method). The proposed effort is designed to ensure continual interaction and exchange between the partners, with annual meetings and extended visits to both industrial sites by both the PI and PhD students. Broader impacts of the work include the following: (i) a more fundamental understanding of polydisperse systems, which is needed before current treatments can be extended beyond two particle sizes, (ii) training of graduate students in the area of particle technology, which has been identified as a national need, and (iii) sharing of learned information with both the technical community and the student community (via incorporation into coursework and outreach).