There are some very basic problems with the existing definition of momentum operators in quantum mechanics and need to be further improved. For example, the kinetic energy operator and the momentum operator are used to calculate the kinetic energy of microscopic particles, and the results obtained are generally different. This article redefines the momentum operator of quantum mechanics and proposes the concept of a universal momentum operator to solve the above problems well. The universal momentum operator only changes the direction of the particle's momentum, but does not change the particle's kinetic energy and energy. Since the calculated values u200bu200bof these two momentum operators differ in the direction of particle motion, additional momentum and additional angular momentum result. This article gives the relationship between the additional angular momentum and the spin of microscopic particles, and explains the nature of the spin of microscopic particles. It is proved that spin is related to the part of angular momentum that cannot be described by the angular momentum operator of quantum mechanics, and is consistent with the inference of the Dirac equation. The description using the Schrödinger equation is unified with the description of the Dirac equation. The so-called spin of an electron is not the rotation around itself, but the rotation around the outermost orbit of the atomic nucleus. The spin operator of microscopic particles is a quantum number, which can be used to make calculations conveniently. On this basis, it is proved that the real reason why Bell's inequality is not supported by experiments is that quantum mechanics has a wrong understanding of the concept of spin projection, and the formula used in deriving Bell's inequality does not hold.