Moving the Magnet Off-Center: Designing What a Colloid Builds

Magnetic colloids are a favorite tool for building structures on demand. Disperse magnetic microparticles in a fluid, switch on a field, and they assemble; switch it off and they fall apart. That reversibility is what makes them attractive for programmable materials, microrobots, and targeted-delivery vehicles. The usual picture treats each particle as a tiny bar magnet with its dipole sitting squarely at the center — and a centered dipole, repeated across many particles, almost always produces the same thing: straight chains aligned with the field.

Our latest paper, led by Bioengineering doctoral student Jonathan A. Victoria-Camacho, asks a simple question with a rich answer: what if the magnet is not at the center? In many real magnetic Janus particles — two-faced spheres with a magnetic patch on one side — the effective dipole is displaced toward the surface. We show that this single geometric choice, how far the dipole sits from the particle’s center, reshapes the entire menu of structures the particles can form.

The gap

Most models of field-directed assembly assume an idealized, centered point dipole, and most experiments report the chains that such a model predicts. But field-responsive particles are increasingly fabricated with off-center, anisotropic magnetization, and there has been no systematic map of how dipole placement steers assembly. Without that map, designing a particle to build a specific architecture — a compact cluster, a porous network, a chiral strand — is guesswork.

What we did

Using Brownian dynamics simulations, the work tracks suspensions of magnetic Janus particles in a quasi-two-dimensional layer and sweeps two control parameters. The first is the radial dipolar shift, s — how far the dipole is displaced from the particle center toward its surface. The second is the Langevin parameter, α, which measures the strength of the applied magnetic field relative to the randomizing thermal energy k_BT: small α means thermal jostling dominates, large α means the field commands.

The payoff of this two-knob sweep is a structure diagram — a map that says which morphology to expect for a given combination of dipole shift and field strength. We compress the underlying physics into a single ratio of competing torques, R_Mag, that pits the field’s tendency to align particles against the off-center dipole’s tendency to twist them into other arrangements. That ratio predicts where one structural regime gives way to the next.

What we found

The sweep reveals six distinct aggregation regimes. At low fields, the dipole shift alone selects the outcome: small shifts give disordered loops, intermediate shifts give compact islands, and large shifts give worm-like clusters. Turn the field up into the intermediate range and the off-center dipole expresses itself as chiral and tangled chains — strands with a built-in handedness that a centered dipole simply cannot produce. Push the field higher still and every case converges to the same answer: fully aligned chains marching along the field direction, the off-center detail finally overwhelmed.

The convergence carries a clean quantitative signature. Under strong fields the clusters grow in time with a universal exponent of z ≈ 0.473, independent of the dipole shift — a kinetic fingerprint of complete alignment. The lesson is that the dipole’s position is a powerful design variable in the regime where field and geometry compete, and a negligible one once the field dominates. Knowing exactly where that crossover sits is what lets a designer choose a regime on purpose rather than stumble into it.

How it fits the broader landscape

This work sharpens a theme the colloidal-assembly community has been circling: that the path to programmable matter runs through anisotropy. It complements our companion study on Janus particles with laterally shifted dipoles under shear flow, and it sits alongside a wave of experimental and computational efforts using combined and time-varying fields to widen the catalog of accessible structures. It is also a close cousin of our recent look at how particle size ratio reshapes a magnetic suspension: in both cases a single microscopic knob rewrites the assembled structure. By reducing the design rules to a single torque ratio and a structure diagram, the study turns “what will these particles build?” into a question with a chartable answer — useful for anyone engineering reconfigurable magnetic materials, soft microrobots, or delivery systems that must assemble and disassemble on command.

How to access

Victoria-Camacho, J. A., & Córdova-Figueroa, U. M. (2026). Self-assembly of magnetic Janus colloids with radially shifted dipoles under an external magnetic field. Soft Matter, 22(8), 1717–1731. DOI: 10.1039/D5SM00872G