Objective This in vitro study is to evaluate the effect of ultrasonic power settings and cooling methods on fracture behavior of the separated nickel titanium instruments during ultrasonic removal. Methods 40 discarded and 10 new nickel titanium instruments were used to establish the separated instrument removal model. The discarded instruments were divided into four groups according to two ultrasonic power settings（6 or 10） and different cooling methods（using water coolant or without using water coolant）, and all the new instruments were put into the group in which ultrasonic vibration was used at a higher power setting（10） without using water coolant. The time required for fracture of the separated instruments was recorded. Results The time was longer with statistical significance at a lower power setting compared with that at a higher power setting regardless of the use of water coolant. The time was longer with statistical significance by the use of water coolant compared with that without using coolant regardless of power settings（6 or 10）. The time required for fracture of the new instruments was longer than that of the discarded instruments with statistical significance. Conclusion During ultrasonic removal of separated nickel titanium instruments, water coolant could decrease the occurrence of fracture. When removing separated nickel titanium instruments with ultrasonic vibration, an attempt without coolant shouldn’t be made at a higher power setting.
