In this work, a new three-dimensional sessile droplet freezing model, involving the ice fraction and ice distribution after the droplet recalescence, is established based on the many-body dissipative particle dynamics with the energy conservation method for the first time. The proposed model is verified by comparing it with experimental results, and the accuracy of this model increases as the ice distribution becomes more non-uniform after recalescence. Furthermore, the effects of surface contact angle, droplet volume, surface temperature, and droplet supercooling degree on the freezing process are investigated in detail. The numerical results demonstrate that the angle of ice tips maintains a constant under various conditions. The upper and lower limits of solidification time under specific conditions are derived, and the droplet solidification time decreases linearly with the increase in supercooling. In addition, the average droplet solidification rate decreases with the increase in droplet volume, contact angle, and surface temperature, and the surface temperature is demonstrated to have the greatest influence on the solidification rate. Emphatically, we put forward an empirical formula, as a function of droplet volume, contact angle, droplet supercooling degree, and surface temperature, to predict the freezing time of a sessile supercooled droplet.

Increasingly heart failure patients need to use Ventricular Assist Devices (VADs) to keep themselves alive. During treatment, hemolysis is an inevitable complication of interventional devices. The most common method for evaluating mechanical hemolysis is to calculate Hemolysis Index (HI) by the power-law formula. However, the HI formula still has obvious flaws. With an intention of further understanding the phenomenon of mechanical hemolysis in non-physiological flow, our study developed a coarse-grained erythrocyte destruction model at the cellular scale and explored the mechanism of the single erythrocyte shear destruction utilizing the Dissipative Particle Dynamics, including the erythrocyte stretching destruction process and the erythrocyte non-physiological shearing destruction process. In the process of stretching and shearing, the high-strain distribution areas of erythrocytes are entirely different. The high-strain areas during stretching are concentrated on the central axis. After the stretch failure, the erythrocyte changes from fusiform to shriveled biconcave. In the shear breaking process, the high strain areas are focused on the erythrocyte edge, causing the red blood cells to evolve from an ellipsoid shape to a plate shape. In addition to the flow shear stress, the shear rate acceleration is also an important factor in the erythrocyte shear damage. The erythrocyte placed in low shear stress flow is still unstably destroyed under high shear rate acceleration. Consequently, the inclusion of flow-buffering structures in the design of VADs may improve non-physiological hemolysis.

The mechanical properties (compressibility or deformability) of cells are closely related to their death, migration, and differentiation. Accurate separation and manipulation of bioparticles based on these mechanical properties are still a challenging in the field of acoustofluidics. In this work, based on surface acoustic waves (SAW) and divergent microchannels, we developed a new method for separating and detecting particles or cells with different compressibility. The difference in acoustic radiation force (F r) caused by compressibility are gradually amplified and accumulated by decreasing the flow velocity, and they are finally reflected in the particle migration distance. During the transverse migration process, the alternating dominance of the acoustic radiation force and the Stokes resistance force (F s) drives the particles to create three typical migration patterns: intermittent migration, compound migration, and near-wall migration. In the present tilted SAW device, a 91% separation success rate of ∼10 μm polystyrene (PS) and polydimethylsiloxane (PDMS) particles can be achieved by optimizing the acoustic field input power and the fluid velocity. The application potential of the present divergent microchannel is validated by separating the myelogenous leukemia cell K562 and the natural killer cell NK92 that have similar densities and sizes (∼15 μm) but different compressibility. The results of this work are expected to provide valuable insights into the acoustofluidics separation and detection of the cells that are with different compressibility.

Erythrocyte dynamics and hemorheology exist inextricably connection. In order to further explore the population dynamics of erythrocytes in non-physiological shear stress flow and its microscopic hemolysis mechanism, this study improved the coarse-grained erythrocytes damaged model and established the hemoglobin (Hb) diffusion model based on the transport dissipative particle dynamics. The whole blood hemolysis simulation results showed that the red blood cells near the active shear side were more likely to be damaged, and most of the escaping cytoplasm was also concentrated in this side. After the destruction and relaxation of erythrocytes, the cell membrane presents a pathological state of relaxation and swelling. Moreover, we built a deep learning network for recognizing erythrocyte morphology and analyzing the erythrocyte population change rule in non-physiological shear stress flow. In this study, the clues of the blood shear-thinning effect were found from erythrocyte dynamics and coarse-grained simulation. After the shearing starts, the coin-stacked erythrocytes are depolymerized. Then, the overturned double concave erythrocytes changed into multilobe erythrocytes. When the flow shear stress gradually increases, most erythrocytes show an ellipsoidal tank-treading movement along the shear direction. Changes in erythrocyte morphology can reduce flow resistance, showing a phenomenon of the whole blood shear-thinning effect.

No abstract