Glass ceramics were made from coal bottom ash by adding CaO and $Li_2O$ as glass modifiers and $TiO_2$ as a nucleating agent in a process of melting and quenching followed by a thermal treatment. The surface of the glass ceramics has 1.6 times more $Li_2O$ compared to the inner matrix. When $TiO_2$ was not added or when only 2 wt% was added, the surface parts of the glass ceramics were crystalline with a thickness close to $130{mu}m$. In addition, the matrixes showed only the glass phase and not the crystalline phase. However, doping of $TiO_2$ from 4 wt% to 10 wt% began to create small crystalline phases in the matrix with an increase in the quantity of the crystalline. The matrix microstructure of glass ceramics containing $TiO_2$ in excess of 8 wt% was a mixture of dark-gray crystalline and white crystalline parts. These two parts had no considerable difference in terms of composition. It was thought that the crystallization mechanism affects the crystal growth, direction and shape and rather than the existence of two types of crystals.