Hall and colleagues experimentally characterize two cylindrical polymer bellows microsprings fabricated by two-photon polymerization in IP-Dip. Compression tests reveal specimen-dependent linear and nonlinear stiffness regions, continued displacement under held loads, and altered load–displacement responses after months in ambient laboratory storage. Low-frequency cyclic loading reveals a displacement lag and increasing effective damping with frequency. A simplified finite-element model connects deformation to the folded geometry. These measured behaviors motivate attention to time, aging, and

TWO-PHOTON POLYMERIZATION

A printed spring just 100 micrometres wide bends, creeps and falls behind the force driving it. Discover what changes when you press, wait and repeat.

Two specimens. Three experiments. A nonlinear response that unfolds over time.

Small enough for a sensor. Its response unfolds over time.

THE VIRTUAL MICROSPRING BENCH

Meet the spring.

Choose an experiment. Change one thing. Watch what the spring tells you.

Does twice the force mean twice the compression?

The shape is an educational interpolation of the reported linear and nonlinear regions; it is not a reconstruction of the measured curve.

Print a hollow shape. Give it room to flex.

Two-photon polymerization cures the IP-Dip photoresist only near a focused laser spot. Moving that spot through the resin builds a three-dimensional spring with a folded wall, a top cap and a fixed base.

The ports let uncured resin leave the hollow interior. They can also let air move as the cavity changes volume.

Write

48 mW laser power 10,000 µm/s scan speed

Build

200 nm slices 300 nm hatch spacing

Develop

15 min PGMEA 15 min isopropyl alcohol

Post-cure

365 nm UV for 2 min No heat treatment

Follow the response, across force and time.

The researchers characterized two springs with a flat-punch nanoindenter.

LOAD MATTERS

The slope changes.

The initial low-load stiffness was 1.75 mN/µm for A and 1.70 mN/µm for B. At higher loads, the springs passed through different nonlinear regions. A single stiffness cannot describe the entire response.

TIME MATTERS

Still force. Moving spring.

While the force stays constant, compression continues to accumulate. In 10 mN holds lasting 5, 10 and 20 seconds, Spring B’s creep rate fell over time: motion slowed without stopping.

RHYTHM MATTERS

The response falls behind.

Cyclic loading at 0.5–1.5 Hz revealed greater lag at higher frequency. Force–displacement loops express energy dissipation. The authors attribute much of the damping to air through the ports, alongside the polymer’s viscoelastic response.

AGE AND LOAD INTERACT

Months change the story.

At 10 mN, A’s end compression increased by about 1 µm in the later characterization; B stayed similar. At 14 mN, A’s end compression instead decreased by about 9% after 11 months. The effect of aging depends on the specimen and load.

LOOK CLOSER

Inside the evidence.

Follow the measurements and the supplementary visuals behind the three experiments.

Design for the push. And for the pause.

The bellows geometry offers a way to manage mechanical contact in tiny optical-fibre sensors. The paper points toward springs that soften impacts while preserving a compact sensing structure.

To design with a printed polymer spring, its loading history and age become part of the specification. These measurements motivate characterization and preconditioning of the resist before use. The sensor applications are proposed directions rather than demonstrations in this study.

The bench uses teaching models, not fitted experimental reconstructions. Press curves interpolate qualitative response regions and the published 10 mN endpoints. Hold creep follows a smooth illustrative expression shaped by Figure 7; model responses at other held forces are illustrative. The cycle uses exact tabulated loss metrics and normalized sine waves with displacement phase lag φ = arctan |δ|; the paper’s δ is not an angle in degrees. Deformation is exaggerated for visibility.

The study characterized two specimens. Its simplified finite-element geometry omits the cap and ports. The contributions of air and material to damping were not individually isolated, and port geometry was not parametrically varied. Aging effects on cyclic damping were not examined.

THE PAPER

Hall, Grunzke, Campbell & Chandrahalim. A nonlinear viscoelastic 3D bellows microspring fabricated with two-photon polymerization. Scientific Reports, 2026 · Article in Press version.

FIGURES & ARTWORK

Paper and supplementary figures reproduced under CC BY 4.0. Images extracted and resized; scientific content unchanged.

The graphical abstract is an AI-generated conceptual illustration adapted from Hall et al.’s Figure 1, Figure 8 and supplementary Figure S1(a). Geometry and traces were simplified, recoloured and conceptualized.