The tensile strength of glass fiber is many times higher than that of bulk glass, but studies have shown that the structure of glass fiber is the same as glass. The hypotheses about the glass structure so far can reflect the actual situation are the "microcrystalline structure hypothesis" and the "network structure hypothesis".
The microcrystalline structure hypothesis believes that glass is composed of silicic acid blocks or "microcrystals" of silicon dioxide, and the "crystallites" are filled with a supercooled solution of silicic acid blocks.
The network structure hypothesis believes that glass is an irregular three-dimensional network of tetrahedrons, alumina trihedrons or boron-oxygen trihedrons of silica, and the gaps between the networks are filled with cations such as Na, K, Ca, Mg, etc. The three-dimensional network structure of silica tetrahedron is the basis for determining the properties of glass, and the filled cations such as Na and Ca are called network modifiers.
A large amount of data proves that the glass structure is approximately orderly. The reason is that there are a certain number and size of regularly arranged regions in the glass structure. This rule is caused by a certain number of polyhedrons following the regular arrangement of similar crystal structures. However, the ordered regions are not strictly periodic like the crystal structure. They are not uniform on the microscopic level, but are uniform on the macroscopic level, reflecting the isotropic properties of the glass.