Functionally Graded Materials Are More Important Than Ever—and Digital Tools are Making Them Easier to Design

What You’ll Learn:

How additive manufacturing advanced the development of functionally graded materials.
Why compositionally graded materials present a greater challenge to materials engineers.
How computational modeling and machine learning expand engineering capabilities.

Increasingly, the aerospace, energy, defense and medical device industries are demanding more of their components. This doesn’t just mean a component needs to be stronger, lighter or have higher temperature resistance. It means the component may need to be all of those things at different times in different locations, while also being as inexpensive as possible.

When selecting materials for these components, design engineers are confronted with a difficult reality: These various factors often involve conflicting tradeoffs in material properties. There is usually no single material that can meet all necessary requirements.

Historically, engineers have solved this problem by decomposing their designs into multiple parts, each made with a different material for its own intended functionality. However, the complexity of a design tends to increase with the number of its constituent parts, as do the points of failure and costs associated with manufacturing, assembly, qualification and maintenance.

Fortunately, recent advancements in materials science like additive manufacturing and computational modeling have opened the door for a new solution: Functionally graded materials (FGMs). These materials have properties that change as a function of location in the material and can be used to create multifunctional components. Components made with FGMs are often more optimal than their traditional counterparts as they replace complex assemblies with a continuous structure that satisfies the same design requirements.

The Birth and Boom of FGMs

One of the earliest uses of FGMs came in the 1980s when the National Aerospace Laboratory of Japan manufactured metal-ceramic composites to withstand the intense environments of space flight. These FGMs were created to account for the twin challenges of incredibly high temperatures and thermal gradients and they achieved superior thermal performance by gradually transitioning from ceramics with ultraheat resistance to metals with higher strength and thermal conductivity. However, these advancements were difficult to replicate elsewhere; the methods used to embed ceramic particles in a metal matrix could not be easily extended to other material classes.

It wasn’t until the rise of additive manufacturing, or 3D printing, in the 21st Century that FGM-related research boomed. 3D printing allowed for increased control of component geometry, which enabled property grading by way of changes in the material’s structure.

For example, the design of lattice structures can be manipulated in additive manufacturing to make the lattice denser (and stronger) in some areas and less dense in others, saving on material costs. Similarly, increased control of material deposition enabled polymer 3D printers to print components made of multiple polymer compositions with varied strengths and densities. Such materials have been used in medical implants to tailor their stiffness and compliance to mimic joints or bones and optimize patient comfort.

Yet, many of the most challenging engineering applications today require high-performance alloys, which are the only materials with the capability to meet design requirements while surviving intense environmental conditions. Gradients between such alloys, commonly called compositionally graded alloys, can couple resistance to high temperatures or corrosive environments with superior mechanical performance or thermal management.

However, compositionally graded alloys are much more difficult to manufacture than other FGMs. Unlike polymer composites, which typically consist of distinctly separate materials, metal alloys tend to diffuse into one another, creating a continuous blend of composition and phases, many of which can have problematic properties or cause cracking or material failure. While these alloys will always present some challenges, advancements in digital modeling promise more success in the future.

How Computational Modeling Can Help

In recent years, successful gradient alloys have been manufactured with the help of integrated computational materials engineering (ICME) techniques, like calculation of phase diagram (CALPHAD) modeling. CALPHAD models can predict which phases will form in a linear gradient between two alloys, enabling designers to know, prior to manufacturing, which gradients will be problematic. Furthermore, CALPHAD can be used to create maps between alloys that show where embrittling phase regions lie.

Other ICME models can predict relevant properties like strength, density, cost, printability or corrosion resistance as a function of alloy composition. Combined with CALPHAD maps of brittle phase regions, these techniques can be used to choose which alloys can be graded to produce the desired multifunctionality with high confidence in manufacturability.

At QuesTek, coupled ICME techniques have been used to computationally assess hundreds of potential FGMs before they are manufactured, saving materials costs and time that might otherwise have been wasted on gradients that are destined to fail.

For example, QuesTek Europe has had success joining nickel (Ni) to titanium (Ti) alloys for lightweighting of space propulsion components using an ICME FGM design framework. Other successful applications include high-performance turbines, hypersonic flight and fusion energy. In all cases the designed gradient materials show superior performance to single-material solutions, enabling designs that push the boundaries of what is possible.

Machine Learning and the Future of FGMs

As new additive manufacturing methods have been developed, the ability to control both geometry and material has steadily increased. But while maps of predicted phases and properties are becoming standard in the design of compositionally graded alloys, they are inherently limited by human ability to visualize and plan in higher-dimensional spaces. These maps typically display properties for two or maybe three material variables at a time. As such, graphical techniques will struggle to utilize the freedoms afforded by future manufacturing processes.

One way to overcome the limitations of graphical approaches is the integration of path-planning algorithms. The same algorithms used by autonomous vehicles to plan routes through their physical environment can be used to design gradient alloys through material spaces. With this approach, undesirable phases can be avoided like obstacles on a road. And just as vehicle routes can be optimized for driving time or fuel savings, gradient alloys can be optimized for manufacturability or smooth property transitions.

Combining artificial intelligence and machine learning with ICME in this way can lead to the generation and evaluation of design concepts far beyond human capabilities. These emergent tools can also shorten the time needed to evaluate ICME models for new compositions.

Another exciting development on the horizon of the FGM landscape is the simultaneous optimization of component geometry and material placement. Topology optimization is a computational design technique that has seen wide adoption in additive manufacturing. With topology optimization, the geometry of a component can be optimized to save cost or increase performance. Muli-material topology optimization (MMTO) is an evolution of the traditional approach that also optimizes the location of material within the component.

The future of FGMs in material design has just begun, and as additive manufacturing and FGM technologies mature, advanced computational design tools and strategies will be essential to realizing the unprecedented design freedom of multifunctional manufacturing.