1 October 2015 in Physics of Fluids
A thick ridge of a film arising along the periphery of a substrate, which is the so-called edge-bead, is investigated experimentally and numerically for the case where the bead becomes double-peaked shape in direction away from the periphery of the substrate. It is clarified, by the optical measurement of thickness variations, that the double-peaked bead is separated from a single bead during the drying process, and after the separation, the inner bead moves inward. This motion of the bead is modeled by a lubrication-approximated flow equation, where the Laplace pressure and the solutocapillary effect are taken into account, coupled with the convective and diffusive mass transfer. The numerically simulated motion of the bead shows a good agreement with that obtained by the experiment, and it is revealed that the mechanism how the double-peaked edge-bead is formed. At the early stage of the drying process, a single bead is formed by the Laplace pressure due to the curvature along the substrate periphery. Along this single bead, the concentration gradient is caused by both the spatial distribution of the evaporation rate and the thickness effect on the change rate of the concentration, due to the fact that the resin concentration increases at a faster rate in the thin region. Consequently, the concentration distribution has a local minimum in the middle of the single bead, which drives oppositely directed solutocapillary flows and separates the bead. After the separation, the inner bead moves inward by a kind of a feed-forward mechanism, until the viscosity of the film becomes large enough to suppress the flow.
1 April 2007 in Physics of Fluids
The thermocapillary flow in cylindrical half-zones is investigated numerically for low-Prandtl-number fluids. We consider a configuration in which part of the cylindrical surface is covered by a very thin rigid film and another part is uncovered, supporting thermocapillary stresses. The steady axisymmetric flow is calculated by finite differences. Besides the aspect ratio, the structure of the toroidal vortex flow depends strongly on the location and relative size of the cylindrical free-surface strip. The stability with respect to arbitrary three-dimensional perturbations of the steady two-dimensional flow is investigated by a linear stability analysis. The critical Reynolds numbers and the physical mechanisms are analyzed depending on the aspect ratio of the liquid bridge, the axial position of the uncovered part of the surface, and the free-surface fraction , i.e., the ratio of the height of the free surface to the full height of the liquid zone. When only a small area of the liquid zone is covered by the solid film, i.e., , the well-known stationary instability for the conventional unconfined half-zone is recovered, which is driven by a combination of the elliptic and the centrifugal mechanism. Decreasing , new instabilities can appear. It is found that all critical modes are likewise driven by the elliptic mechanism, the centrifugal mechanism, or a combination of both. When the unconfined free surface is located adjacent to the hot wall, the linear stability boundary shows an intricate behavior. Four qualitatively different instability modes can arise depending on the free-surface fraction and the aspect ratio. One of the time-dependent modes that is strongly driven by the elliptic mechanism resembles an oscillatory Kelvin mode on a strained vortex. The other oscillatory modes show a different behavior as they are mainly driven by centrifugal mechanisms. Quite generally, the importance of the elliptic mechanism diminishes as is further decreased until, for a very small free-surface fraction , an instability appears that is purely centrifugal and stationary. When the free surface is located next to the cold wall, the types of instabilities are the same as for the unconfined half-zone, and the azimuthal wave numbers increase with decreasing .
1 January 2008 in Advances In Space Research
19 April 2022 in International Journal for Numerical Methods in Fluids
A physics‐informed neural network (PINN), which has been recently proposed by Raissi et al. [J Comput Phys. 2019;378:686–707], is applied to the partial differential equation (PDE) of liquid film flows. The PDE considered is the time evolution of the thickness distribution owing to the Laplace pressure, which involves fourth‐order spatial derivative and fourth‐order nonlinear term. Even for such a PDE, it is confirmed that the PINN can predict the solutions with sufficient accuracy. Nevertheless, some improvements are needed in training convergence and accuracy of the solutions. The precision of floating‐point numbers is a critical issue for the present PDE. When the calculation is executed with a single precision floating‐point number, the optimization is terminated due to the loss of significant digits. Calculation of the automatic differentiation dominates the computational time required for training and becomes exponentially longer with increasing order of derivatives. By splitting the original fourth‐order single PDE into lower‐order coupled PDEs, the computational time for each training iteration is greatly reduced. The sampling density of training data also significantly affects training convergence. For the problem considered in this study, improved convergence was obtained by allowing the sampling density of training data to be greater in earlier time ranges, where the rapid flattening of the thickness occurs.
9 October 2008 in Heat Transfer
28 June 2018 in Flow, Turbulence and Combustion