Abstract
本論文では, 差分格子ボルツマン法を用いて二相せん断流れ場における液滴の変形・分裂を数値解析し, メディアレス超薄膜高速回転式分散機分散部における微粒化過程, および微粒化性能に影響する無次元数キャピラリー数Caとレイノルズ数Re関係を考察した。その結果, 一定なレイノルズ数Reで, 微粒化の必要な臨界キャピラリー数Cacが存在することが確認された。これを基に, メディアレス超薄膜高速回転式分散機のナノレベルの微粒化能力を明らかにした。
Introduction: The Challenge of Nano-Dispersal
In industries ranging from cosmetics to fuel cells, the demand for "nano-sized" particles is increasing. Traditional dispersing machines often struggle to reach the sub-micron level consistently. To address this, researchers developed the "medialess ultrathin film high-speed rotary disperser" (such as the Clear SS5). However, optimizing these machines is difficult because the actual "atomization" process—where large droplets break into smaller ones—occurs within a gap only a few microns wide, spinning at extreme speeds.
To understand and optimize this process, numerical simulation is required to bridge the gap between machine design and fluid dynamics.
Methodology: Simulating Shear Flow with FDLBM
The researchers employed the Finite Difference Lattice Boltzmann Method (FDLBM) to model the droplet breakup. Unlike traditional Navier-Stokes solvers, LBM treats fluid as a collection of virtual particles on a grid, making it highly effective for tracking complex moving interfaces between immiscible liquids (like oil in water).
The complex 3D geometry of the disperser was simplified into a 2D "simple shear flow" model. In this setup, a droplet is placed between a stationary top plate and a fast-moving bottom plate, mimicking the high-shear environment within the disperser's micron-scale gap.
Simulation Parameters
The study used physical parameters corresponding to real-world materials:
- Dispersed Phase: Modeled after silicone oil ($\sigma = 20.1 \times 10^{-3}$ N/m).
- Continuous Phase: Modeled after water ($\mu = 10^{-3}$ Pa·s).
- Machine Setup: Rotor diameter of 100mm, 10,000 rpm, and a gap height ($H$) of 1.5 $\mu$m.
Key Insights: The Physics of Breakup
The fate of a droplet depends on the balance between viscous forces (tearing it apart) and surface tension (holding it together), a ratio defined by the Capillary number ($Ca$).
The simulations identified a Critical Capillary Number ($Ca_c$). If the machine's $Ca$ is below this threshold, the droplet merely deforms into an oval. If it exceeds $Ca_c$, the droplet stretches into a long thread and eventually snaps into smaller "daughter" droplets. The relationship between $Ca$ and the Reynolds number ($Re$)—which measures inertia—defines the boundary for successful atomization.
Results and Nano-Level Capability
The FDLBM simulations successfully captured the transition from a stable droplet to a "dumbbell" shape and finally to a string of smaller particles. At a Reynolds number of $Re = 4.9$, the critical Capillary number was determined to be approximately $Ca_c = 0.19$.
Based on these findings, the authors calculated that their specific disperser setup could produce particles with a diameter of roughly 219 nm. This quantitatively confirms the machine's ability to achieve nano-level dispersion through mechanical shear alone.
Limitations and Future Outlook
While the 2D model provides valuable insights, real-world flow is 3D and involves complex spiral trajectories and centrifugal effects. Future research aims to incorporate:
- 3D Geometry: To capture turbulence and spiral flow patterns.
- Surface Chemistry: Modeling "wall wetting" and electrostatic forces to account for re-agglomeration, where particles might clump back together after formation.
Summary updated