Statement of problem: A phenomenon of screw-loosening in implant abutment is frequently occurred in a single and multiple implant restoration. Purpose: This study was performed to evaluate an effect of abutment material on screw-loosening before and after a cyclic loading. In a single-tooth implant, different materials of abutment, Type III Gold alloy and Zirconium composite$(ZrO_2/Al_2O_3)$ were used. Material and method: The Gold alloy(Type III) and Zirconium composite$(ZrO_2/Al_2O_3)$ were used to make a superstructure of implant, the one of types of UCLA, Each group was constituted of 5 sample with a 30-degree offset angulated loading platform. The external hexagonal fixture was rigidly hel d in a special holding zig to ensure solid fixation without rotation during the tightening and a cyclic loading. A Titanium-alloy screw was used to connect and controlled to be tighten in 20Ncm torque by a digital torque gauge. A 20 times of consecutive closing/opening cycle were performed to evaluate the immediate torque loss. In 5 sample of each material group, an initial opening torque was recorded during 3 closing/opening cycle, then 2Hz, 200N, 1,000,000 cyclic loadings were performed, then a opening torque was evaluated. Result & Conclusion: 1. In this limited study, titanium alloy screw tightened in 20Ncm, a cold-welding phenomen on was not observed during the 20 times of closing/opening cycle(p=0.11, p=0.18). 2. In titanium alloy abutment screw, repeated opening and closing of the screw caused to progressive decrease of opening torque(p=0.014). 3. The difference in preload of screw between gold alloy abutment and ceramic$(ZrO_2/Al_2O_3)$ abutment was not significant(p=0.78). 4. The difference in torque loss of screw between gold alloy abutment and ceramic$(ZrO_2/Al_2O_3)$ abutment was not significant after 2Hz,200N, 1,000,000 cyclic loading(p=0.92). 5. In titanium alloy abutment screw tightened by 20Ncm, the screw loosening was not significant on each group after 2Hz, 200Ncm, 1,000,000 cyclic loading(p=0.59).