This article focuses on the nanotechnology-related research work at Georgina Institute of Technology. The Georgia Institute of Technology’s Multiscale Systems Engineering Research Group is working to integrate the modeling and simulation features of today’s computer-aided design (CAD) with materials design capability. These integrated features would be available at the nano, meso, micro, and macro scales, which is called multiscale CAD. In future CAD systems, engineers will be able to zoom in to specify material morphology and distributions. Offering the capability of designing materials in CAD requires the representation of many different kinds of shapes. The multiscale CAD would also allow engineers to design better functional materials, such as state-change materials. The geometric modeling of microstructures that make up material is still in its infancy. The efficiency and controllability of complex and porous shapes are the most important research topics for the interactive modeling and design of microstructures.
The time has come for engineers to be able to customize their material exactly to the piece they’re designing.
As the advent of flexible electronics attests, the materials with which things are made are at the root of today's product innovations.
The Georgia Institute of Technology's Multiscale Systems Engineering Research Group, where I’m a faculty member, is working to integrate the modeling and simulation features of today's CAD with materials design capability. These integrated features would be available at the nano, meso, micro, and macro scales, which we call multiscale CAD.
Integration would allow engineers to create customized materials (that is, materials that contain pores or voids, or super alloys that have coexisting phases) to meet their needs while performing structural and shape design at the macro scale.
Similar to the conventional CAD as the first tool for virtual prototyping, the primary function of multiscale CAD is to allow the efficient construction and interactive modification of geometric models for microstructures. Existing boundary-representation-based parametric modeling approaches have become inefficient in model construction at nano and meso scales where geometry and topology are highly complex. New modeling and representation techniques are thus needed and this is the goal of our research.
In future CAD systems, engineers will be able to zoom in to specify material morphology and distributions. They’ll be able to combine material design at the nano or micro scales, with geometrical and topological design at the macro scale to optimize the product's performance.
In this way, design engineers will be able to customize materials to their design in much the same way they select and change part geometries today. They’ll be able to simulate the product with the selected geometries and materials in place. These would be available in an all-in-one package so engineers could create specific materials while they are designing a new product.
What we’re envisioning is to allow engineers to define their own materials rather than use those already discovered.
New materials’ future
As it stands now, there's a divide between materials creation and product design. Bringing them together within the same system will allow for all kinds of new material properties and structures that currently haven’t been used in the engineered world.
Integrating those two functions would allow engineers to customize and design material properties on any portion of their CAD design by simply zooming into the specific region, specifying material compositions, designing atomic or crystalline configurations, and simulating material performance within the product to ensure it will work as needed. If it doesn’t, engineers can redesign the material, the part, or both, and test again.
This is of interest because design decisions made at the microscopic level determine a material's properties, which, in turn, determine product behavior.
In today's CAD-enabled design processes, design engineers select available materials from databases. They select the materials they deem best suited to their products’ specifications and their designs.
The conventional material selection approach that design engineers usually take is based on the isolated databases that were built without the input of problem-specific needs. Such a one-directional approach to discover, devise, and deploy new materials has a long development cycle and is not cost-effective. Even the materials science and engineering community has realized that this “lack of design” approach limits the rapid advancement of engineering materials.
To discover new materials, scientists run many experiments. In a process that is analogous to baking a cake, various ingredients are mixed and processing conditions are tried based on scientists’ own experiences. If a cake is too hard or if it falls, a baker will tinker with the recipe, and materials scientists have had a similar practice.
It's only after materials scientists discover a new material through these experiment-driven approaches that product engineers consider how it can be best used within an engineered product. They also enter the new material into the materials database used by CAD designers today.
So from an engineering design perspective, discovery is not design. Design starts with asking the question: “What are the problems I have and what are the materials I need to solve it?”
The existing product development process doesn’t integrate material design for the product into CAD. Say a mechanical engineer wants to design a vehicle from design specifications that call for the car to be light but strong. Currently, the engineer can only play around with different geometric shapes and topologies within the CAD design.
Materials are given. They can only be selected from the existing materials database.
Therefore, the available “degrees of freedom” that design engineers control to optimize the performance of products are restricted to the geometry and topology offered within CAD systems. The addition of material properties in the design space would offer design engineers more degrees of freedom. Customizable materials would provide extra flexibility to realize increasingly intricate product functionality.
In other words, materials selection should be replaced by materials design to better meet customers’ requirements in realizing modern products.
Materials are given. they can only be selected from the existing materials database.
As it stands now, there's a divide between materials creation and product design.
Bringing them together within the same system will allow for all kinds of new material properties and structures that currently haven’t been used in the engineered world.
Compute instead of experiment
Simulation software tools would also be used to analyze the material at multiple scales. The engineer could use them to compute the physical properties instead of experiment. The tools can quickly answer questions to verify the design; for example, what is the modulus of elasticity of the newly designed material? This mechanical property can be directly calculated and predicted from material configurations done at the atomic scale.
Nano scale materials simulation can predict mechanical, thermal, optical, and electrical properties for some particular atomistic structure. The calculated properties from nano scale simulations can be plugged into traditional FEA for overall structure analysis. So it's kind of a chain reaction of calculations. Computation allows engineers to predict material properties and use those numbers to run FEA at the macro scale.
The value of a multiscale CAD environment lies in knowledge shared and used across disciplines. Do mechanical engineers have to learn more about materials in order to use multiscale CAD in the future? No. In fact, today's engineers don’t have to know all details of how to formulate and solve differential equations behind FEA and CFD software in order to run simulations.
Similarly, our goal is to develop integrated software tools that will allow engineers to design materials in the same way they run FEA analyses, without the need of knowing all chemistry and physics behind them. We aim for software tools that simulate materials and predict their properties by simple mouse clicks.
We began working on this concept a few years ago. One challenge has been how to allow computers to represent shapes and structures at the nano or meso scales, which are much more complex than the structures at the macro level, as currently done in CAD. We are also working on a new area called computer-aided nanomanufacturing that can predict whether or not the design of nanostructures and nanomaterials are manufacturable.
Radically different shapes
Offering the capability of designing materials in CAD requires the representation of many different kinds of shapes. Compared to the geometries at the nano scale or in nature, traditionally used geometries in engineering design are very simple and mostly flat, such as buildings, chairs, and computers, because that's what the current CAD can do today.
Natural shapes, like sea urchin, and kale, and geometries at nano scales, like zeolite, and polymer, are very complex. Part of our effort is to see how we can present these atomic configurations and porous structures by computer.
We aim for software tools that simulate materials and predict their properties by simple mouse clicks.
The engineer using the nano CAD software would first design the complex shapes interactively, simulate properties of the design, and optimize toward the need. The engineer could then apply the designed material in the macro scale simulation using FEA software.
The multiscale CAD would also allow engineers to design better functional materials, such as state-change materials. Examples of these materials, used for their state transformation properties every day, include batteries for energy storage, DVDs for information storage, and shape-memory alloys for orthopedic surgery. Charged and discharged batteries are two states of the material, as are burned and erased DVDs, and deformations of shape-memory alloys.
For instance, erasable DVDs depend on special materials with certain optical properties. When a laser pulse burns them, the materials change between transparent and non-transparent. This optical property change is used to store the binary information of zero or one.
The designer of DVD materials needs to decide the speed of burning and the amount of energy required to finish the state transformation. Atomistic simulation can be used to predict the required energy for the transformation. The designer can determine the material compositions and atomic configurations best suited to meet target performance.
The geometric modeling of microstructures that make up material is still in its infancy. The efficiency and controllability of complex and porous shapes are the most important research topics for the interactive modeling and design of microstructures.