← All news

Articles & applications

Soft actuators that bend. But how much force do they exert?

Polymer actuators that bend under electrical voltage, but whose force remains difficult to measure. With Sara Beauquis and her partners, we characterized the forces they produce to compare their actual performance with mechanical models.

Illustration of a soft polymer actuator bending under electrical voltage and exerting a force at its tip
Illustration SensiTips

The partners

This work builds on the research of Sara Beauquis at the Laboratoire de l'Intégration du Matériau au Système (IMS, University of Bordeaux), with Sara Zanchi, Cédric Ayela and Damien Thuau.

Their goal: to develop soft actuators, fully fabricated by screen printing, that deform under an electric field. This approach opens up interesting prospects for miniaturized devices that are simple to fabricate and could potentially be integrated into soft systems.

The study also involves Sébastien Roland and Eva Héripré, from the PIMM laboratory (Arts et Métiers, CNRS, Cnam), for the characterization of the materials and structure.

It was in this context that Sara Beauquis came to FEMTO-ST, in Besançon, with her samples. We worked with her on the SensiTips microforce measurement platform to quantify the forces actually produced by her actuators.

Why this problem is so interesting

An actuator can deform a great deal… without exerting a useful force.

This is the whole challenge of electroactive polymers, and in particular of electrostrictive materials, which deform when an electric field is applied to them. These devices are of interest for soft robotics, delicate manipulation and, eventually, some biomedical applications.

But to design a useful actuator, observing its bending is not enough. You also need to know how much force it can exert. It is this trade-off between deformation and force that determines its real value in an application.

The measurement challenge

The structures studied here are very thin: the active layers are only 5 to 16 µm thick, deposited on flexible substrates of 25 or 50 µm.

Under these conditions, measuring the blocking force — that is, the force developed when the actuator is prevented from moving — is far from straightforward. You have to make contact with the moving tip without disturbing the measurement, while remaining sensitive to forces of the order of a few hundred micronewtons.

That is too low for many conventional instruments. The measurement therefore had to be sensitive, controlled and repeatable enough to be compared with a model.

How we measured it

Sara came to carry out the tests with us, bringing her samples. We provided the SensiTips platform and supported the experiments to measure the forces developed during actuation.

The setup relied on a microforce sensor with a 2 mN full scale and a resolution better than 1 µN. The sensor had been calibrated against a reference linked to a mass standard, with a relative expanded uncertainty of 0.44 % (k = 2) on its sensitivity.

In practice, the sensor was brought into contact with the tip of the actuator at rest. When an electric field was applied for 5 or 10 s, the actuator tried to bend, but its displacement was blocked by the sensor. The force could then be recorded over time.

This approach gave Sara quantitative data on the force actually available, where simply observing the deformation would not have been enough.

What our partners demonstrated

From these measurements and their models, Sara Beauquis and her colleagues showed that the actuator geometry directly governs the trade-off between deformation and blocking force.

Increasing the length-to-width ratio favours bending but reduces the force. Conversely, increasing the active area improves both. Substrate thickness also plays an important role: a thicker substrate can increase the force but limits the deformation.

For a single-layer actuator, the authors report a mean force of 0.61 ± 0.01 mN over five successive cycles, with a drop of only 7 % after 30 actuations.

The comparison between theory and experiment also highlights some limitations. For a three-layer architecture, a 29 % discrepancy is observed between model and measurement, which the authors attribute in particular to leakage currents.

Finally, the study highlights the value of multilayer architectures, with up to six electroactive layers. Calculations show that increasing their number significantly increases the available force.

What about you?

Are you developing microactuators, soft materials or miniaturized structures that are difficult to characterize? We can help you obtain reliable, SI-traceable microforce measurements on objects that conventional methods struggle to test.

Discover our measurement services →

Reference

Sara Beauquis, Sara Zanchi, Eva Héripré, Valentin Reynaud, Joël Agnus, Cédric Ayela, Sébastien Roland and Damien Thuau, “Optimization of Fully Screen-Printed Electrostrictive Multilayer Actuators”, Macromolecular Materials and Engineering, vol. 311, e70331, 2026. DOI: 10.1002/mame.70331

Read the publication ↗ (opens in a new tab)

← All news