Title: INTEGRATED COMPUTATIONAL MATERIALS ENGINEERING DEVELOPMENT OF CARBON FIBER COMPOSITES FOR LIGHTWEIGHT VEHICLES

Abstract

This project, Integrated Computational Materials Engineering Development of Carbon Fiber Composites for Lightweight Vehicles, develops Integrated Computational Materials Engineering (ICME) techniques for Carbon Fiber Reinforced Polymer (CFRP) composites and uses these ICME tools within a multi-disciplinary optimization scheme to design a carbon-fiber intensive front subframe for a passenger sedan. The subframe is a structural automotive component with stringent requirements for vibration, strength, durability and safety performance. CFRP composites, with a density of 1.55 g/cm³ and a tensile strength of up to 2,000 MPa in the fiber direction, have a high strength-to-weight ratio, making them one of the most promising candidates to replace the metals currently used for automotive structural components. By increasing the accessibility of CFRP component designs, reducing development-to-deployment lead time of CFRP components, and improving the robustness of initial designs, this project supports weight reduction in light-duty vehicles. These efforts can help mitigate greenhouse gas emissions from passenger vehicles and have the potential to improve national energy independence. The four-year project concluded in early FY2019 and leveraged a total budget of $8.58 million dollars. The U.S. Department of Energy funded the project under award DE-EE0006867 for $6M and industry partners supplied the remaining $2.58M. Researchers from Ford Motormore » Company led the project team with colleagues from the Dow Chemical Company, Northwestern University, the University of Maryland and partners from the following software companies: Livermore Software Technology Corporation (LS-DYNA), ESTECO (modeFRONTIER), HBM Prenscia (nCode) and Autodesk, Inc. (Moldflow).The project focused on three key architectures of epoxy-resin CFRP composites: chopped carbon fiber, unidirectional (UD) carbon fiber and woven carbon fiber. The chopped carbon fiber was investigated in the sheet molding compound (SMC) form. All three architectures, chopped SMC, UD and woven were investigated in the compression molding manufacturing process. The mechanical properties of CFRP are highly direction-dependent as the initial fabric quality, material layout, preforming and molding processes all determine the final local orientation of the fiber. As the local fiber orientation significantly affects the properties and performance of a CFRP-intensive component, achieving the optimal component design requires tools that are capable of predicting both the microstructure and performance of the component based on fiber architecture, molding process, and curing history. Strong consideration should also be made of the uncertainty inherent in each process, and the probabilistic nature of materials and manufacturing processes. The ICME tools developed in this project meet these design challenges. The new, validated and integrated ICME tools significantly advance the capabilities of the industry and include a novel non-orthogonal material model for preforming analysis, improved compression molding simulations of chopped carbon fiber SMC, industry-first multiscale models of constitutive behavior of chopped SMC, UD and woven continuous fiber composite, refined crash analysis models, and fatigue models. These models are incorporated for the first time into a multi-disciplinary optimization workflow for the design of a CFRP composite component. modeFRONTIER serves as the platform for ICME tool integration and design optimization. Existing capabilities and added scripts enable a seamless integration of manufacturing simulation and component performance analysis. The project results improve the infrastructure for CFRP composite design for automotive structures. The data from each of the three considered epoxy-resin CFRP composite materials (UD, woven and chopped SMC) add to the publicly available materials characterization database housed at the National Institute of Standards and Technology (NIST). In total, more than 700 coupon test results provide the data necessary for input and validation of the material models for CFRP, including those currently available and developed in this project. The project contributes improved and novel models for CFRP composites in both manufacturing and performance simulations. Newly defined models generate fiber orientation during the preforming process for continuous and chopped fiber tows. Model improvements in the compression molding simulations add speed and accuracy to manufacturing simulations. These updated manufacturing models pass information to novel local property models for the crash, strength, fatigue, and vibration performance simulations of CFRP components. These local material models embrace a stochastic multiscale framework that leverages quantified measures of uncertainty and its propagation across length scales. modeFRONTIER connects the ICME-based models into a multi-disciplinary optimization workflow that modifies geometry, part thickness, material selection, and composite layup from the manufacturing process through various engineering performance simulations. Finally, the design exercise combines a component weight and variable cost estimate into the ICME-based multi-disciplinary optimization for an industry-first end-to-end design tool for CFRP components and systems. This ICME-based multi-disciplinary design This ICME-based multi-disciplinary design process is an industry first. A representative hat section is modeled and compared with physical molded parts to validate the full ICME-based multi-disciplinary optimization. A similar modeFRONTIER workflow produces an initial design for a subframe. The multi-material subframe design meets critical engineering requirements and achieves at least a 25% weight reduction with a cost increase of less than $4.27 per pound of weight saved when compared to the baseline stamped steel subframe. The final phase of the project produces an initial subframe design that saves 30% weight and costs an additional $4.01 per pound saved compared to a stamped steel baseline subframe. The multi-material design utilizes 79 wt% (percentage by weight) steel with reinforcements of 16 wt% chopped carbon fiber SMC and patches of UD carbon fiber composite 5 wt%. This initial design meets critical stiffness, strength and durability targets, and one key safety metric. For the first time, multiple diverse design variables are concurrently investigated in an automated procedure processed in batch mode on a high-performance computing platform to produce a design that meets multiple engineering performance metrics and variable cost.The ICME models are all accurate within 15% of experimental values, except in the case of specific dynamic, high loading-rate behaviors that require ongoing investigation to refine predictive capabilities. The inclusion of a variable cost estimate into the multi-disciplinary optimization adds tremendous value for users in the automotive industry by expanding the objectives in the optimization scheme to include cast as well as weight while adding little cost or complexity to the computational procedures. This combination of ICME-based models and tools showcases the advantage of producing an initial design of a complex automotive part with a methodology that concurrently considers materials, performance and costs. The public benefits from improved CFRP composite materials testing methods and a plethora of novel data, simulation tools applicable to many industries, and a multi-disciplinary optimization scheme that includes performance and variable cost in a holistic component design. The ICME-based multi-level, multi-disciplinary design optimization developed in this project produces an initial design that meets the selected engineering performance metrics. While a production-ready final design must also address other considerations such as manufacturability, serviceability, corrosion performance and repair, the initial design criteria integrated through ICME methods in this project are a critical subset of factors to produce a robust initial design. This exercise produces an initial design that could be used as a robust starting point for the more detailed final design. Experience shows that moving from a robust initial design to a final design typically results in small changes in weight and variable cost.« less

Authors:

Su, Xuming; Wagner, David

Publication Date:

2019-03-25

Research Org.:

Ford Motor Company, Detroit, MI (United States)

Sponsoring Org.:

USDOE

OSTI Identifier:

1502875

Report Number(s):

DE-EE0006867

DOE Contract Number:  

EE0006867

Resource Type:

Technical Report

Country of Publication:

United States

Language:

English

Subject:

36 MATERIALS SCIENCE; 42 ENGINEERING; 99 GENERAL AND MISCELLANEOUS