A dual-frequency laser interferometer has been developed based on a low-performance commercial interferometer. An optical resolution of 1.24 nm and a nanometer-scale accuracy have been achieved by using unique techniques to obtain an optical subdivision factor of 1/8. A method for reducing static positioning errors was also shown. The measurement of a free-falling body was performed to test the maximum achievable target velocity of the device. The experimental setup for measuring the static positioning errors was also given. The new interferometer could be widely used in nanometer-scale fabrications and measurements.

A measurement interferometer available for wafer stage metrology of lithography has been investigated by means of resolution-extending of optical subdivision based on commercial interferometers. Factors that determine the accuracy, linearity and repeatability of nanometer-scale measurements of displacements exceeding hundreds of millimeters and oftarget velocities exceeding hundreds ofmillimeters per second are also discussed.

No abstract

No abstract

Only the first-order Doppler frequency shift is considered in current laser dual-frequency interferometers; however; the second-order Doppler frequency shift should be considered when the measurement corner cube (MCC) moves at high velocity or variable velocity because it can cause considerable error. The influence of the second-order Doppler frequency shift on interferometer error is studied in this paper, and a model of the second-order Doppler error is put forward. Moreover, the model has been simulated with both high velocity and variable velocity motion. The simulated results show that the second-order Doppler error is proportional to the velocity of the MCC when it moves with uniform motion and the measured displacement is certain. When the MCC moves with variable motion, the second-order Doppler error concerns not only velocity but also acceleration. When muzzle velocity is zero the second-order Doppler error caused by an acceleration of 0.6g can be up to 2.5 nm in 0.4 s, which is not negligible in nanometric measurement. Moreover, when the muzzle velocity is nonzero, the accelerated motion may result in a greater error and decelerated motion may result in a smaller error.