Predicting Magnetic Janus Particle Assembly with Differential Evolution Algorithm
Eric A. McPherson, Kenneth Kroenlein, Ilona Kretzschmar · 2025
Read Predicting Magnetic Janus Particle Assembly with Differential Evolution Algorithm by Eric A. McPherson, Kenneth Kroenlein, Ilona Kretzschmar (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-10882.txt.
Introduction
Eric McPherson, Kenneth Kroenlein and Ilona Kretzschmar use a differential evolution method to predict how magnetic Janus particles assemble in 2025. These two-faced particles can form complex structures for smart fluids, but their parameter space is far too large for trial and error.
Their simulation searches for low-energy arrangements using a point dipole model of the magnetic patches. Predictions were checked against four published experimental and simulation cases with varied patch shifts and field conditions. The method captured particle orientation and structure in most cases. It struggled only where large clusters had to reorganize wholesale, the authors note.
Catalog record
| CREATOR | Eric A. McPherson, Kenneth Kroenlein, Ilona Kretzschmar |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-10882
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Predicting Magnetic Janus Particle Assembly with Differential Evolution Algorithm Eric A. McPherson, Kenneth Kroenlein, Ilona Kretzschmar Abstract Magnetic Janus particles allow access to complex, nonlinear assembled structures that may enable interesting new magnetorheological (MR) fluids with uniquely engineered field responses. However, the overwhelming size of the parameter space for Janus and patchy particles makes exploration of such systems by experimental trial and error or through detailed simulation impractical. Here, a differential evolution (DE)-based simulation method is explored to predict the assembly of magnetic Janus particles as an alternative method for assembly prediction. Structure predictions from the DE simulation for laterally- and radially-shifted magnetic Janus particles are compared to four published experimental and simulation case studies. The DE simulation captures the orientation and structure of magnetic Janus particles for a range of shifts and a variety of external field conditions using the point dipole approximation. Structural predictions that rely on the reorganization of large clusters of particles were less well represented by the DE predictions. Despite this limitation, the DE simulation method can be used to predict key structural factors for magnetic Janus particle assemblies, as demonstrated by favorable comparison with three of the four model studies. I Introduction Magnetorheological (MR) fluids are colloidal suspensions that change their viscosity in the presence of a magnetic field. Ashtiani, Hashemabadi, and Ghaffari 2015 ; Eshgarf, Ahmadi Nadooshan, and Raisi 2022 The viscosity change is caused by the assembly of magnetic microparticles due to magnetic dipole-dipole and dipole-field interactions. The field-controlled viscosity of MR fluids has applications in mechanical dampers, Olabi and Grunwald 2007 prosthetics, Jonsdottir et al. 2009 braking systems, Karakoc, Park, and Suleman 2008 bottom-up designable materials, Li, Yang, and Yin 2020 and MR finishing for optical surfaces. Kumar et al. 2022 An MR fluid’s response to a field must be characterized to take advantage of it in an engineering context; this necessitates understanding and predicting the assembly of the constituent particles. Janus particles are microparticles that have two different surface materials, giving the particles different functionalities including tunable responses to external fields. Chen et al. 2012 ; Zhang, Luijten, and Granick 2015 ; Zhang, Grzybowski, and Granick 2017 ; Bishop, Biswal, and Bharti 2023 Metallic and metal oxide coatings, including iron, Smoukov et al. 2009 iron oxides, Ren et al. 2012 nickel, Sinn et al. 2011 ; Demirörs et al. 2018 and composite caps made of cobalt/platinum layers, Suzuki 1995 ; Baraban et al. 2008 have been used to create particles that assemble in magnetic fields. Their isotropic interactions have been modeled as a "shifted" magnetic dipole. Kantorovich et al. 2011a The shifted dipole causes an increase in assembly complexity, expanding the range of potential structures and resultant macroscopic properties compared to magnetically isotropic particles. Tan et al. 2024 Several research groups have explored the structures and potential applications of magnetic Janus particles with various cap designs. Zhang, Luijten, and Granick 2015 ; Zhang, Grzybowski, and Granick 2017 Smoukov et al. Smoukov et al. 2009 showed that magnetic Janus particles under the influence of magnetic and electric fields form chain structures that can then be demagnetized and disassembled. Han et al. Han et al. 2017 made patchy magnetic microcubes that had programmed and manipulated folding shapes. Our research group has fabricated ferro- and ferrimagnetic Janus particles and characterized their assembly structures and kinetics in an applied external field illustrating cap-material dependence of their assembled morphology. Ren et al. 2012 ; Long, Córdova-Figueroa, and Kretzschmar 2019 While these investigations have led to a fundamental understanding of the assembly dynamics of magnetic Janus particles, experimental explorations of these materials are limited because of the time it takes to synthesize and test unique particles. It is also nontrivial to characterize the as-manufactured magnetic properties of individual particles such as dipole shift, moment, and direction for magnetic caps. Philipp et al. 2021 In parallel, simulations utilizing an array of particle types with well-defined physical properties have been performed in silico . Eslami, Khanjari, and Müller-Plathe 2019 ; Donaldson, Schall, and Rossi 2021 Kantorovich et al. Kantorovich et al. 2011b explored models of magnetic particles with radially-shifted dipoles using Monte Carlo simulations and found a wide range of structures for different shifts. Vega-Bellido et al. Vega-Bellido et al. 2019 and Victoria-Camacho et al. A. Victoria-Camacho et al. 2020 used Brownian dynamics (BD) simulations to understand the dynamics of magnetic Janus particle assembly for laterally- and radially-shifted particles, respectively. DeLaCruz-Araujo et al. A. DeLaCruz-Araujo et al. 2016 used BD simulations to look at (nonmagnetic) Janus particles in shear flow, finding that they can assemble in a range of structures including micelles, vesicles, lamellae, and worm-like structures. The simulations give insight to a larger structural complexity, show the impact of particle shape and Brownian motion, and validate the point-dipole approximation; however, it still takes significant time and computational resources to explore large particle systems with complex interactions. In BD simulations this problem is further exacerbated as larger dipole shifts are explored because of the shorter time steps required. Vega-Bellido et al. 2019 Variations in size, shape, surface functionalization, and material of the particle core or cap also increase the complexity. Further, external gradients (including chemical, electric, magnetic, and thermal gradients) impact the dynamics for directed self-assembly. Zhang, Grzybowski, and Granick 2017 These drawbacks motivate the search for alternative methods for predicting self-assembled structures. Optimization strategies such as evolutionary algorithms can explore a vast configurational space quickly and predict particle structures without relying on detailed simulations. Deaven and Ho 1995 ; Daven et al. 1996 Evolutionary algorithms have been used to optimize many problems, including the assembly of colloidal crystals, Chremos and Likos 2009 simulated patchy colloidal particles with Lenard-Jones interactions, Doppelbauer, Bianchi, and Kahl 2010 ; Bianchi et al. 2012 and cluster prediction for charged Cruz, Marques, and Pereira 2016 and Lenard-Jones colloids. Schönborn et al. 2009 Evolutionary programming has been use by Brown et al. Brown, Thompson, and Schultz 2010 to predict interatomic potentials, showing that algorithms can be parallelized. Inspired by biological evolution, these optimization methods are easy to interpret and parallelize. Evolutionary algorithms are also widely applicable as long as the solution can be encoded as an array of values termed a "chromosome". Additionally, since they are heuristic-based optimization methods they have the potential to reach cluster predictions with fewer simulation steps and calculations. Because of their ease of use and wide applicability, modifications have been made to create families of similarly related optimization methods. Differential evolution (DE) is a type of evolutionary algorithm Storn and Price 1997 ; Ahmad et al. 2022 that has been shown to work better for continuous-value optimizations compared to genetic algorithms. Storn and Price 1997 Moloi and Ali Moloi and Ali 2005 used DE as an algorithm to minimize potential energy functions for pair and many-body potentials and found that it worked favorably compared to other global minimization methods. This paper tests the ability of a DE-based simulation to predict magnetic Janus particle assembly. Key structure parameters are measured to compare the predictions from the DE simulation to a selected set of published experiments Ren et al. 2012 ; Baraban et al. 2008 and simulations, Vega-Bellido et al. 2019 ; Okada and Satoh 2023 highlighting the advantages and limitations