Additive manufacturing, such as 3D printing, provides an excellent opportunity to design metamaterials: materials with an engineered structure that leads to desired properties such as, for instance, resistance to vibrations. However, a major challenge was that the predicted metamaterial response often failed to match real-world behaviour. Researchers at the University of Groningen have now shown that the unexpected behaviour of 3D-printed metamaterial structures is not due to structural defects, as was commonly believed, but that the material simply needs to be properly characterized to obtain models with high predictive accuracy. The results were published in Materials Horizons on 3 June 2026.
3D printers build up objects layer by layer. This additive process can potentially introduce structural defects, including weak planes and direction-dependent properties. These were thought to be the culprit of unpredicted behaviour in 3D-printed metamaterials. 'Our study shows that this is not true,' says PhD student Sidharth Beniwal. 'In fact, it is perfectly possible to manufacture a 3D-printed object on a low-cost machine and design it in such a way that it exhibits excellent vibration attenuation – and even more exotic properties, such as localizing vibrations in certain parts or allowing their propagation in one direction while prohibiting it in the opposite direction.'
Under the supervision of Ranjita Bose and Anastasiia Krushynska, Beniwal used a combination of numerical and experimental methods, testing various materials commonly used for 3D printing. He compared the vibration-attenuation properties of 3D-printed parts of different shapes in various environmental conditions. Beniwal: 'We observed almost no differences in vibration control between structures manufactured with low-cost and more expensive 3D-printing techniques, indicating that manufacturing defects can safely be neglected.'
This reveals that the long-standing mismatch between numerical predictions and experimental observations in 3D-printed vibration-controlling structures must have a different origin than the structural defects, as was previously assumed. Instead, the researchers show that if the material that is used for 3D printing is properly characterized in the predictive model, it is perfectly possible to design 3D-printed structures that behave as predicted. This paves the way for various useful applications, such as vibration isolation, noise reduction, structural health monitoring, sensing, signal processing, and energy harvesting, as well as next-generation wave-controlling materials and devices.
Beniwal: 'The most exciting part of the project was obtaining the first experimental results and seeing how closely they matched our numerical predictions. Initially, this was demonstrated for relatively simple structures, which was already encouraging. However, the real excitement came when we extended the approach to more complex structures and observed the same level of accuracy. We noticed that we were, in fact, able to predict the metamaterial behaviour quite accurately, contrary to what is commonly reported in the literature. So, it became clear that we were addressing a fundamental challenge in the field.'
