Page 45 - Read Online

Page 16 of 33 Ren et al. Soft Sci. 2026, 6, 6

Heat is highly attractive due to its ease of generation, compatibility with many printable chemicals, tunable
transmission in printed parts, and the ability to be triggered within specific time frames. However, it also has
some drawbacks, including slower response times due to diffusion transmission, fatigue or creep under
repeated cycling, and difficulty matching physiological temperatures in in-vivo applications.

Magnetic field
Magnetic actuation in 4D-printed soft microrobots is achieved by coupling external magnetic fields to
magnetically responsive particles embedded within the material. Torques generated by uniform or rotating
fields induce bending, twisting, folding, or whole-body rotation, while field gradients produce net forces for
translation or lifting [Figure 6A] [119] . Magneto-responsive composites are formed by dispersing micro or
nanoscale superparamagnetic or ferromagnetic particles into soft polymer matrices, enabling programmable
deformations under remote control. Owing to their noncontact operation and deep penetration, these
systems hold significant promise for biomedicine, microfluidics, and soft robotics .
[120]

A magnetic soft microrobot with volume v can be viewed as a composite body possessing effective
magnetization M, which is induced or programmed under an applied magnetic field B. In a spatially varying
field, a net magnetic force F arises and can be expressed in

F = v(M × )B (1)
∇
whereas in a (quasi) uniform field, the robot experiences a magnetic torque τ as expressed in

τ = vM × B (2)

which tends to align the body’s preferred magnetization direction with the field [121] . These relations
emphasize two practical trends. First, for a fixed field strength and gradient, the attainable force and torque
scale approximately with the robot volume v, which explains why actuation strength quickly shrinks as
devices are miniaturized unless stronger gradients or higher effective magnetization are available. Second,
particle loading influences actuation mainly through M. Increasing the magnetic filler volume fraction
generally enhances the effective susceptibility or magnetization, but the improvement can become less than
proportional at high volume fraction due to saturation and interparticle interactions, and it is strongly
affected by dispersion quality [122] . In particular, aggregation, sedimentation, or nonuniform particle
distribution can reduce the effective M even at the same loading, leading to lower force density and larger
variability . The preferred magnetization axis is often dictated by geometry, but can also reflect crystalline
[123]
anisotropy, and it can be intentionally programmed by aligning magnetic nanostructures within the matrix
or by premagnetizing the composite along a prescribed direction.

Chung et al. reported magnetically tunable stiffness metamaterials in which magnetic torque served as the
actuation source . Using DIW of a styrene-isoprene-styrene matrix filled with neodymium microparticles
[124]
and a ternary programming scheme, the metamaterial switched among soft, moderate, and stiff states with
rapid response, and a 3D array enabled multi-layer stiffness control under external fields [Figure 6B].
Building on magnetic actuation, Zhang et al. 3D-printed soft magnetoactive origami films from UV curable
elastomers loaded with up to 75 wt.% ferromagnetic particles . Wireless magnetic fields set polarity and
[31]
drove folding and locomotion for targeted gastric drug delivery and terrain adaptive locomotion [Figure 6C].
To avoid continuous magnetic fields during gripping, Wu et al. developed an electrothermal magnetic shape
memory microgripper based on Fe O -filled polymers with embedded resistive wires . The magnetic field
[125]
3
4
opened the gripper while electrothermal-triggered shape memory effect closed and locked it without power,
yielding about 0.9 s response and a high load-to-weight ratio [Figure 6D].