Science1 publisher2 min readPublished
Rice theorists cancel a magnetic assembly bias by rotating the field past a full turn
A Rice-led paper in Physical Review Research finds that magnetic lag gives even a symmetric rotating field a preferred axis, and calculates the extra fraction of a turn that cancels it in simulation.
The Scientist · Science desk
What happened
- Tanaka Tatsuya, a visiting Rice research scientist from Kao Corp. in Japan, worked out a way to eliminate a directional bias that appears when magnetic particles are assembled using rotating magnetic fields.
- The bias comes from magnetic relaxation: the particle's magnetic response lags the applied field, so a 360-degree sweep samples some orientations more than others and leaves a preferred axis.
- Rotating slightly beyond 360 degrees before reversing cancels that preference at one precisely calculated angle, leaving two particles with the same average interaction whatever their orientation.
- Numerical simulations left the derived relationship intact after the researchers added interactions between particles and the particles' own physical rotation.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability A lab already running a rotating-field rig can attack the bias by changing a number in the drive program, with no change to the particles themselves.
- constraint The correction is specific to one relaxation time at one rotation rate, so the delay time has to be characterised first, and a sample whose particles relax at different speeds has no single right overshoot.
- precedent The theory names a falsifiable target angle. The next experiment has to count defects at the corrected sweep against a plain 360-degree one.
A field that keeps turning the same way sets particles and their clusters spinning continuously, so the common protocol sweeps one direction and then reverses [3]. The reversal is what makes the sweep angle a choice, and the Rice paper in Physical Review Research turns that angle into a calculated quantity [2].
"When we rotate the magnetic field, the magnetic response of the particle is always slightly behind," said Sibani Lisa Biswal, chair of chemical and biomolecular engineering at Rice and the study's corresponding author [7]. "That tiny delay turns out to matter. It can create a preferred direction for assembly even when the applied field appears perfectly symmetric" [8].
The extra angle is different for every setup. Tanaka's relation fixes it from two quantities: how fast the particle's magnetization responds, its relaxation time, and how fast the field rotates [9]. One of those is an instrument setting and the other is a property of the colloid. That property has to be measured before the drive waveform can be programmed [18]. "Calculate the extra rotation using the delay time of the particle and the rotational speed of the field," said Aldo Spatafora-Salazar, a research scientist in the Biswal lab and one of the authors. "The sweep can then be fixed as an experimental parameter to achieve isotropic interactions" [11].
The work is theory, and the checks are numerical [1]. Rice did not report assembling a crystal at the corrected sweep [16].
One fixed overshoot cancels the bias for one relaxation time at one rotation rate, so particles in a batch that relax at different speeds do not share a single correct sweep angle [17]. The bias itself is the kind the authors say pushes particles in preferred directions and makes uniform crystalline material harder to grow [14].
"If you want particles to form highly ordered crystals, you first need to understand and eliminate unintended biases in how they interact," Biswal said. "This gives us a quantitative recipe for doing that" [13].
What to watch
- An experiment that assembles colloids at the corrected sweep and at 360 degrees, and counts defects in both.
- Whether the relation holds for samples with a spread of relaxation times rather than a single particle type.
- Whether other groups begin reporting the sweep angle as a fixed parameter in published assembly protocols.