When a polymer solution droplet is deposited on a lyophobic surface, the contact line is moved back to some degree and subsequently pinned. An experimental setup is constructed to investigate not only the receding process but also an internal flow of polystyrene-acetophenone and -anisole solutions. As a result, the time variation of the evaporation rate per unit area during receding does not strongly depend on the initial solute concentration. The average solute concentration at the pinning of the contact line increases as the initial solute concentration increases. A convective circulation flow that is upward at the axis of symmetry is observed. This flow pattern is different from those of pure liquids such as water, acetone, benzene, and so forth, which have been previously reported. Furthermore, the observed flow is enhanced as the initial solute concentration increases, contrary to an increase in the fluid viscosity. To resolve these discrepancies, the mechanism of the flow is numerically investigated using a hemispherical droplet model considering the density and surface tension distributions. The numerical results demonstrate that the circulation flow that is experimentally observed is actually caused. It is also found that the solutal Rayleigh effect initially induces the internal flow, and subsequently the solutal Marangoni effect dominates the flow. Both effects are enhanced as the initial concentration increases because of the evaporative mass balance at the free surface.

The internal flow of an evaporating polymer solution droplet on a substrate is experimentally studied. The flow visualization is carried out. The effect of the initial polymer concentration is further investigated. A polystyrene-acetophenone (PS-Ap) and a polystyrene-anisole (PS-Ani) solution are employed as the droplet. A nylon powder is mixed with the droplet for the visualization by a YAG-laser sheet light. The droplet evaporates after the settlement on the substrate. Without the polymer dissolved in the solvent, complicated flow is observed in both droplets. For the cases with dissolved polymer, the flow pattern is rectified. In the PS-Ap droplet, the source flow is observed for the initial solute mass fraction c0 = 0.005 – 0.20. This convection becomes strong as c0 increases. The mechanism of the flow inside the PS-Ap droplet can be understood by the combination of the natural convection and Marangoni convection due to the differences of the temperature and the solute concentration. As for the PS-Ani droplet, the evaporation process and the flow pattern are affected by c0. For the dilute solution (0 < c0 < 0.03), the contact angle decreases during the contact line receding. The observed flow pattern becomes similar to that in the PS-Ap droplet. At c0 = 0.08 – 0.2, the decline of the contact angle is remarkable and the direction of the internal flow becomes inverse. This flow mechanism cannot be clarified, but it may have the relations with the decreasing contact angle.