Predicting Fiber Attrition during Processing of Long-Fiber Reinforced Composites using a Mechanistic Model Approach

由于其重量轻和上级强度，长纤维增强热塑性塑料（LFT）被汽车和航空航天工业用于制造关键的承载结构。LFT的主要问题之一是增强塑料的纤维在成型过程中断裂，从而影响最终部件的强度。纤维断裂还影响制造商在制造这些部件时的其他问题：纤维如何排列（纤维取向）以及它们如何聚在一起，留下没有纤维的部件部分（纤维密度分布）。问题就在这里--是纤维赋予了零件强度。今天，制造公司花费大量的时间和资源试图控制过程，以防止纤维断裂、聚束或过度定向。他们通过反复制作原型来做到这一点，直到得到正确的结果（试错法）。该项目中的计算机模拟可以帮助工程师可视化成型过程中的纤维运动，从而在实际制造零件之前解决潜在的问题。该项目将产生一种工具，工业界可以用来控制不连续纤维增强聚合物复合材料结构的制造过程，使工程师能够制造出他们知道可以信任的部件。随着信任度的提高，这些用于轻质部件的节能生产方法将得到更广泛的接受。轻型部件的增加将提高运输部门的燃油效率，大幅减少二氧化碳排放，并直接支持重要的全球气候变化最小化战略。此外，能够充满信心地设计LFT部件将释放出具有成本竞争力的环保、轻质和坚固复合材料部件的生产潜力，并在需要能源效率和创新的时候为汽车和航空工业提供技术优势。本项目中提出的建模方法旨在提供理解和预测纤维增强复合材料的成型，特别是流动过程中纤维磨损和纤维密度分布的发展，目前的模拟程序无法处理。该项目中的模拟使用纤维弯曲和纤维断裂的单颗粒模拟方法模拟聚合物加工浓度下的纤维悬浮液行为。该模型将系统中的每根纤维表示为通过节点互连的梁单元的柔性链。对柔性纤维进行建模对于正确理解纤维堵塞和纤维断裂等行为至关重要，而在使用常见的刚性纤维假设时没有考虑这些行为。在这里研究的模型包括影响，如流体动力，纤维的灵活性，并排除体积力，由于纤维纤维和纤维壁接触。通过该模拟工作获得的结果将通过PI实验室在受控实验装置中进行的测量进行验证，最终将与PI工业合作伙伴制造的现实零件进行比较。在项目结束时，机械模型方法将与商业软件包相结合。最终产品将允许工艺工程师在设计阶段预测工艺引起的纤维断裂以及最终部件的性能。此外，还可以对工艺进行修改，以找到最佳条件、螺杆和浇口几何形状，从而最大限度地减少纤维磨损并实现理想的纤维长度分布。最终的工具将是第一个模型，它结合并耦合了所有三个纤维特性及其相互作用：纤维取向，纤维密度和纤维长度分布。 随着对成型过程中纤维运动现象的更高水平的理解，最终将有可能大规模生产具有更高性能和受控质量的聚合物复合材料部件，从而使轻质聚合物复合材料可用于更广泛的应用。

Due to their light weight and superior strength, Long Fiber-Reinforced Thermoplastics (LFTs) are used by the automotive and aerospace industries to manufacture critical load bearing structures. One of the major problems with LFTs is that the fibers that reinforce the plastic break during the molding process, compromising the strength of the final part. The fiber breakage also influences other concerns that manufacturers have when making these parts: how the fibers align (fiber orientation) and how they bunch up, leaving portions of the part without fibers (fiber density distribution). Therein lies the problem - it is the fibers that give a part strength. Today, manufacturing companies spend an enormous amount of time and resources trying to control the process to keep the fibers from breaking, bunching up or excessively orienting. They do this by repeatedly making prototypes until getting it right (trial-and-error). The computer simulation within this project can help the engineer visualize the fiber motion within the molding process, and thus solve the underlying problems before actually making a part. This project will result in a tool that industry can use to control the manufacturing process of discontinuous fiber-reinforced polymer composite structures, allowing engineers to make a part they know they can trust. With this increase in trust, these energy-efficient production methods for light weight parts will have a much wider acceptance. An increase in light parts will result in fuel efficiency in the transportation sector, significantly reducing CO2 emissions and directly supporting important worldwide climate change minimization strategies. Furthermore, being able to design LFT parts with confidence will unleash the potential of cost-competitive production of environmentally friendly, light weight and strong composite parts, and provide a technical edge to the automotive and aeronautical industries at a time when energy efficiency and innovation are needed.The modeling approach presented in this project is aimed at providing a tool required to understand and predict defects that arise in the molding of fiber reinforced composites, which today's simulation programs are not able to handle, in particular fiber attrition and fiber density distribution development during flow. The simulation in this project models the behavior of fiber suspensions at polymer processing concentrations using a single particle simulation approach for fiber bending and fiber breakage. The model represents each fiber in the system as a flexible chain of beam elements interconnected by nodes. Modeling flexible fibers is essential to properly understand behavior such as fiber jamming and fiber breakage, which is not accounted for when using the common rigid fiber assumption. The model researched here includes effects such as hydrodynamic forces, fiber flexibility, and excluded volume forces due to fiber-fiber and fiber-wall contacts. Results obtained with this simulation work will be validated with measurements conducted in controlled experimental set-ups at the PI's laboratories, and ultimately comparisons will be made with realist parts made by the PI's industrial partners. At the end of the project, the mechanistic model approach will be coupled to commercial software packages. The final product will allow the process engineer to predict process-induced fiber breakage as well as the properties of the final part during the design phase. Additionally, the processes can be modified to find optimal conditions, screw and gate geometries in order to minimize fiber attrition and achieve ideal fiber length distributions. The final tool will be the first model that incorporates and couples all three fiber properties and their interactions: fiber orientation, fiber density and fiber length distributions. With a higher level of understanding of the fiber motion phenomena during molding it will eventually be possible to mass produce polymer composite parts with increased properties and controlled quality making light weight polymer composites available to a wider range of applications.