Fig. 1. Working principle of the device. (a) The scanner consists of a magnet, four springs, and a lens. (b) A pair of coils in an arrangement similar to an anti-Helmholtz coil creates a magnetic field gradient at the magnet location. A change in the current through the coils changes this gradient, which in turn changes the force on the actuator. This change in force ultimately causes a displacement of the lens.

Fig. 2. Mechanical FEA results of the 3D nano-printed MEMS actuator. (a) Simulated displacement of the lens as a function of the applied magnetic force, showing a linear relationship. (b) The simulated force remains constant for all magnet displacements along the optical axis. (c) Resonant modes of the actuator. The first mode is at 303 Hz. (d) Normalized z-component of the simulated magnetic field. The field is approximately zero at the location of the actuator.

Fig. 3. Fabrication process of the complete actuator. A 20 nm layer of PVA is spin-coated onto the substrate (a) prior to 3D nano-printing (b). The structure is developed in PGMEA (c) and rinsed in IPA under UV irradiation (d). PVA is dissolved in water to enable lift-off. (e) After lift-off, the structure is placed in the lower coil (f). Before the evaporation of the solvent, water is exchanged with IPA to lower the capillary forces (g). The micromagnet is placed in the actuator and fixed using UV-curable adhesive (h). The second coil is placed on the actuator and fixed using UV-curable adhesive (i).

Fig. 4. Photographs of the actuator. (a) and (b) Complete actuator with coil pair. (c) Actuator before magnet integration. (d) Mechanical spring before and after the ablation of the safety pins.

Fig. 5. Frequency response of the actuator. (a) Complete frequency response for a peak current of 2 mA. The resonant frequency of the first mode is 347.3 Hz with a quality factor of 17. (b) The resonant peak of the fundamental mode for different drive currents. A shift from 347.3 to 343.8 Hz is observed for an increase of the drive current from 2 to 10.8 mA.

Fig. 6. Quasi-static actuation characteristics of the MEMS scanner under different conditions. (a) Quasi-static displacement of the MEMS actuator as a function of applied current for three peak current levels at 1 Hz. Significant hysteresis is observed. (b) The hysteresis amplitude decreases with increasing frequency, measured in the range from 250 mHz to 20 Hz. (c) Displacement versus current profiles for two selected drive frequencies show reduced hysteresis at higher frequencies. (d) The gain decreases with increasing frequency.

Fig. 7. Step response of the MEMS actuator under different driving currents. (a) The displacement of the actuator over time for different current levels. Viscoelastic creep is observable after the transient response. (b) Initial ringing occurs following the step input and stabilizes over time. The rise time τr is 0.51ms±0.02ms. (c)–(e) Step response for three different currents. The data are fitted with a general Kelvin–Voigt viscoelasticity model with two elements.³⁸

Fig. 8. General Kelvin–Voigt viscoelasticity model of second order³⁸ was used to characterize the viscoelastic material properties. The model was fitted to the step response of the device for three different currents. Errors correspond to the standard deviation of the mean.

Fig. 9. Measurement of the long-term stability. (a) Zoom in to the displacement over time. (b) Peak-to-peak displacement of the actuator over 1000 cycles for three different drive currents. (c) Evaluation of displacement drift and coil temperature over time for a current of 45.1 mA. (d) Linear correlation between displacement drift and coil temperature.

Fig. 10. Results of the shape characterization. (a) A complete profile of the printed lens. (b) Gaussian high-pass filtered section of the lens. The RMS roughness is 37 nm. (c) Shape deviation of the uncompensated, nonreleased lens. (d) Shape deviation of the back and the front surfaces of the compensated lens after lift-off. (e) Total shape deviation (thickness deviation) of the compensated lens. (f) Zernike polynomials of the front and the back surfaces. (g) Combined Zernike polynomials.

Fig. 11. Results of the optical characterization. (a) Profiles of the measured focal spot, the simulated spot based on the measured shapes, and the simulated spot based on the ideal shape. (b) Focal spot of the 3D-nanoprinted actuator. (c) Simulation of the focal spot based on the measured shape. (d) Still images of the focal plane while the actuator translates the lens along the optical axis. Scale bar: 5μm (Video 1, AVI, 0.7 MB [URL: https://doi.org/10.1117/1.APN.4.4.046015.s1]). (e) Still images of the focal spot while the actuator translates bidirectionally. The displacement from the focal plane is estimated from the measured device’s response. Scale bar: 5μm.