To find strategies for improved cleaning with microbubbles (MBs), we used an MB cleaning and microscopic observation system to investigate (a) radius distributions of MBs generated by a hydrogen bubble method, (b) the behavior of MBs adsorbed on a polymer black ink (contact angle to water = 60°), and (c) the performance of the MBs in removing the ink from a substrate. The radius distribution of MBs was not affected by changing the cathode diameter or the applied voltage between the anode and cathode. The number of MBs increased when we used thicker cathodes and applied higher voltages. MBs adsorbed onto the ink surface increased in radius during the cleaning process. The rates of increase in adsorbed MB diameters on the ink surface were lesser than those for MBs that were not adsorbed, because of ink movement on the adsorbed MB surfaces. We determined the response of ink weight loss from the substrate due to changes in height from the cathode, applied voltage, and cathode diameter. The ink weight loss increased with decreased height between the ink and cathode and increased applied voltage and cathode diameter. These results suggest that increases in weight loss were strongly influenced by the number of MBs (10–100 μm in diameter). MBs adsorbed onto the substrate removed trace amounts of ink from the substrate; after removal of the ink, adsorption of other MBs was induced. The diameter distribution of MBs had a lesser effect on weight loss than the number of MBs generated. Important processes for cleaning the ink are as follows: (1) adsorption of MB onto the surface of the ink, (2) increasing diameter of the adsorbed MB, and (3) removal of the adsorbed MB from the ink surface. We also would like to suggest that the cleaning efficiency for hydrophobic ink could be improved by increasing the number of adsorbed MBs, and MBs must be generated closest to the surface of the ink.