For decades, shape-shifting liquid-metal robots that can morph into new forms and heal their own wounds have been strictly confined to the realm of science fiction blockbusters. However, a new engineering breakthrough has just made them a reality.
According to a study published in the journal Science Advances, a joint research team from Seoul National University (SNU) has successfully developed a radically new type of robotic muscle using a “phase-transitional ferrofluid” (PTF).
Unlike traditional robots that are permanently locked into a single, pre-programmed motion, this highly adaptive, slime-like material allows a single soft robot to tackle completely new tasks, survive catastrophic physical damage, and introduce a highly sustainable model for the future of electronics.
The problem with soft robotics
The new technology aims to solve a major bottleneck in the commercialisation of soft robotics. Currently, engineers rely on dielectric elastomer actuators (DEAs) — soft transducers that convert electrical energy into mechanical motion. Because they move rapidly and precisely, they are often referred to as “artificial muscles.”
These lightweight DEAs are increasingly used for delicate tasks, such as robotic grippers that handle fragile fruit, or haptic vibration components in smart wearables.
However, they suffer from a fatal flaw: once their internal electrode pattern is printed and manufactured, their shape is permanently fixed. If a factory robot needs to grasp a slightly different object, engineers must design and fabricate an entirely new robot from scratch, driving up manufacturing costs and killing efficiency.
A shape-shifting breakthrough
To overcome this rigid limitation, the SNU researchers, led by Professor Jeong-Yun Sun and Professor Ho-Young Kim, combined advanced materials science with mechanical engineering to create the PTF electrode.
This unique material behaves as a stable solid at standard room temperatures. However, when exposed to external stimuli such as heat, it melts into a highly flexible, low-viscosity fluid. Because it is a ferrofluid, this liquid state is highly magnetically responsive, meaning its shape can be actively manipulated and dragged into new three-dimensional configurations using magnetic fields.
This phase-transitional ability unlocks three leaps forward for robotics:
Real-time reconfiguration: Even while the artificial muscle is operating, the electrode can be melted and repositioned. It can be split apart, bridged across severed circuits, or moulded into complex 3D architectures. This allows a single robot to essentially “learn” entirely different motions, switching from bending to expanding on the fly.
Autonomous self-healing: If the robotic muscle is slashed by a sharp object or suffers an electrical breakdown due to high voltage, the system is not destroyed. By simply heating the damaged area to a liquid state, the ferrofluid flows to reconnect the broken circuit or actively bypasses the ruined dielectric, fully restoring the robot’s capabilities.
Environmental reusability: Currently, when a soft robot reaches the end of its lifespan, the entire device is thrown away. With the PTF system, the electrode can simply be melted down, extracted as a liquid, stored, and injected into a brand-new device. The researchers demonstrated that even after multiple complete reuse cycles, the fluid maintains a staggering recovery rate of approximately 91 per cent alongside consistent physical performance.
The end of disposable machines
The research team believes this material will spark a transformative shift away from passive, disposable machines, paving the way for sustainable, adaptive robotics capable of self-repair in extreme industrial environments.
“This study represents a breakthrough in transforming traditionally static and passive electrodes into ‘living, programmable elements’ through innovations in particle and polymer design,” Professor Sun stated. “This self-healing and shape-reconfigurable electrode technology will serve as a key foundation for sustainable next-generation soft robotics.”
