Objective To study the effects on compressive strength and rigidity of tibia cortical bone from deep-freezing, freeze-drying or radiation treatments, and to discuss the appropriate method for tibia cortical bone treatment. Methods The cortical bone were collected from the middle part in tibial diaphysis from amputated limbs of trauma patients and made into bone plates with the size of 10 mm×10 mm×5 mm each. The bone plates were then divided into seven groups evenly and randomly: control group (Group A), deep-freezing group (Group B), freeze-drying group(Group C), deep-freezing plus 60Co (25 J/g) radiation group(Group D), deep-freezing plus 60Co (50 J/g) radiation group(Group E), freeze-drying plus 60Co (25 J/g) radiation group(Group F), freeze-drying plus 60Co (50 J/g) radiation group(Group G). The compressive strength and rigidity of allograft cortical bone were tested by mechanical testing machine. Results The largest compressive strength of the tibia cortical bone was in the range of 6.089-9.089 kN. Compared with Group A, the strength in Group B, C, D and F showed no significant difference, and the rigidity in Group B and C showed no significant difference, while the rigidity in Group D and F was decreased by 9.6% (P<0.05) and 8.7% (P<0.05), respectively. Compared with Group A, the strength in Group E and G was reduced by 29.6% (P<0.05) and 33.1% (P<0.05), respectively, and the rigidity was reduced by 16.7% (P<0.05) and 14.8% (P<0.05), respectively. Conclusions The strength and rigidity of tibia cortical bone are not changed significantly after deep-freezing or freeze-drying treatment. Compared with the untreated group, the strength of tibial cortical bone with the small dosage of 60Co treatment is not significantly changed after deep-freezing or freeze-drying, but the rigidity is decreased; the strength and rigidity with the large dosage of 60Co treatment are decreased obviously. For application of cortical bone used in spinal fusion, radiation sterilization dosage should be controlled in the range of 15-25 J/g.