In this paper, the optimum coverage of 4.44% and the optimum adsorption sites were determined for the Be3N2 adsorption system of O atoms at different coverages based on density functional theory. The electronic and optical properties of the model were investigated by applying bending deformation to the model at these coverage and adsorption sites. Adsorption of O atoms disrupts the geometrical symmetry of Be3N2, resulting in orbital rehybridization and lowering its band gap. Bending deformation causes the band gap of the adsorbed O atom structure of Be3N2 to first increase and then decrease, resulting in the modulation of its band gap. With increasing bending deformation, the adsorbed system is redshifts, and the degree of redshift increases with increasing bending deformation. All calculations in this paper were performed using the first-principles-based CASTEP module of Materials Studio (MS). The generalized gradient approximation (GGA) plane-wave pseudopotential method and the Perdew-Burke-Ernzerhof (PBE) Perdew et al. Phys Rev Lett 77:3865, 1996 generalized functional were used in the geometry optimization and calculation process to calculate the exchange-correlation potential between electrons. The effect of coverage on the electronic and optical properties of the Be3N2-adsorbed O atom system was investigated by adsorbing different numbers of O atoms on a monolayer of Be3N2. The Be3N2 protocell contains two N atoms and three Be atoms with a space community of P6/MMM (No.191). The original cell was expanded 3 times along the direction of the base vectors a and b in the Be3N2 plane to create a 3 × 3 × 1 monolayer Be3N2 supercell system. A vacuum layer of 15Å is set in the direction of the crystal plane of the vertical monolayer Be3N2 supercell to eliminate interactions between adjacent layers. In the overall energy convergence test of the Be3N2 supercell, the plane wave truncation energy was set to 500eV, and the energy difference between the calculations given in the literature Reyes-Serrato et al. J Phys Chem Solids 59:743-6, 1998 using 550eV was less than 0.01eV, verifying the reliability of the data at a truncation energy of 500eV. The Monkhorst-Pack special k-point sampling method Monkhorst et al. Phys Rev B 13:5188, 1976 was used in the structural calculations, and the grid was set to 3 × 3 × 1. The geometric optimization parameters are set as follows: the self-consistent field iteration convergence criterion is 2.0 × 10-6eV, and the iterative accuracy convergence value is not less than 1.0 × 10-5eV/atom for the total force of each atom and less than 0.03eV/Å for all atomic forces. In addition the high-symmetry k-point path is taken as Γ(0,0,0) → M(0,0.5,0) → K(- 1/3,2/3,0) → Γ(0,0,0) Chen et al, AIP Adv 8:105105, 2018.