I suppose the effect of a solvent on the morphology of a submerged solid is a surface phenomenon. I know from catalysis that surfaces have active sites depending on the crystal structure of the surface. The activity at the active sites for catalytic reactions depends on the environment for the process in question. I know that the activity of such sites depend on both electric potential and chemical potential. Could it be that the change in concentration of the solvent, by changing the chemical potential (of the solute ?) relative to the surface of your MnSe, in this way strips off MnSe material from the MnSe surface in different patterns depending on the concentration because the favorable patterns - those with minimized surface energies - depend on such chemical potentials? I know that one can use acids to structure silicon surfaces in inverse pyramid shapes to better trap light for electricity generation, just as an example of what I came to think about when reading your question.