Advanced Composites Spotlight: Long Fiber Thermoplastics

Much attention has been given to the advent of long fiber thermoplastics due to their desirable recycling, manufacturing, and mechanical properties. However, the question remains as to whether current analysis tools and techniques can safely capture this material’s behavior. 


A thermoplastic is a polymer (or resin material) which becomes pliable or moldable above a specific temperature and returns to a solid state upon cooling. This is in contrast to thermosetting plastics which are irreversibly cured. 

When compared to other materials, thermoplastics stack up well: 

  • 60% lighter and 600% stiffer than steel 
  • 30% lighter than aluminum
  • 200% tougher than thermoset composites
  • 60% less scrap during production than sheet goods
  • Recyclable and reusable

A downside is that they are generally more expensive. Epoxy thermosets are typically $1-10/pound, while PEEK thermoplastics can cost anywhere between $10-100/lb. 

One of the most promising uses of this thermoplastic resin is as a matrix material for long, discontinuous fiber advanced composites. Compared to short fiber composites and thermosets, long fiber thermoplastics (LFTs) offer better mechanical properties in terms of elastic stiffness, strength, creep and fatigue endurance, and crashworthiness. 

When aligned along the loading axis, long fiber thermoplastics can withstand up to 70% of the load of continuous fiber composites at a fraction of the manufacturing costs. As techniques improve, the ability to “align” fibers will likely improve and manufacturing costs will continue to decrease. 


LFTs are also favorable because existing injection molding machines used in industry today can be adjusted to mass-produce LFT parts. 

For these reasons, some believe that thermoplastics could be the “missing link” between exotic advanced composite applications and the mass market. 

Whether this is just marketing hype has yet to be seen, but what we do know is that with the advent of this new blend of composites, empiricists need to come up with a whole new set of theories to capture and predict LFT behavior. 

It is a well-disguised fact that most composite analysis tools utilize simplifying assumptions that don’t play nicely with long fiber composites. Analytical theories that assume inclusions are “oblong” or “spherical” in shape are voided with long fiber composites, because the real material deviates so far from the ideal (assumed) geometry. And don’t even get us started on theories that utilize rule of mixtures. 

When a discontinuous short- or long-fiber polymer composite is subjected to monotonic loading or cyclic loading, matrix cracking initiates from existing micro-voids in the matrix material, or cracking will start from fiber ends which are the sites of stress concentrations. This cracking then combines forces with (or is sometimes the cause of) fiber-matrix debonding at the fiber-matrix interfaces. The final cause of failure is often fiber pull-out and rupture. 

Depending on the pre-treatment and manufacturing method used, an LFT composite part can have vastly different values for influential properties such as fiber orientation, material defects, matrix voids, and fiber degradation factors. 

Constructing analytical equations to account for these variables is feasible but very time-consuming. It is far better to have a realistic geometry and physics-based approach for capturing microstructural nuances and predicting material behavior. That is where a tool with advanced micromechanical modeling capabilities, such as MultiMech, can become a huge advantage. 

Long fiber thermoplastics have great potential, but whether or not they will be the material that “catapults” past steel/aluminum is uncertain. What is more probable is that engineers will continue to invent new ways to combine materials, yielding more customized and favorable mechanical properties. 

If we are to keep up with this innovation, engineers must adopt flexible physics-based modeling technologies rather than calibrated analytical equations.