Jenny Lei

Institution: 
UC Santa Barbara
Major: 
Mechanical Engineering
Year: 
2010

Microcantilevers as Mass Sensors

Microcantilevers are among the most versatile microelectromechanical systems (MEMS). They can be used in numerous applications: atomic force microscopes, biosensors, chemical sensors, flash memory, and signal filters. Cantilevers oscillate at a natural frequency (ω0), which depends on the cantilever’s mass. As mass is added, ω changes; thus, cantilevers can be used as mass sensors. This study focuses on increasing detectability of microcantilevers by exploiting nonautonomous dynamics and feedback.

Parametric amplification and Q control are the amplification techniques analyzed. Parametric amplification is the same process as “pumping” one’s legs on a swing to increase the amplitude. Q control performs a method known as “self-excitation.” The cantilever response signal runs through a filter to isolate ω0, it is then amplified and phase shifted before being used to actuate the cantilever.

This work concentrates on implementing these techniques, which are tested at the macro scale. A function generator provides a response signal to the shaker, exciting the cantilever. An accelerometer measures the cantilever response viewed with the oscilloscope and spectrum analyzer. A MATLAB code implements parametric amplification. The product of a forcing term at 2ω0 with the displacement effectively modulates the spring stiffness; we expect 5x gain. Q control involves a circuit with variable potentiometers; we expect 10x gain.  Parametric amplification requires a frequency sweep, whereas, Q control does not. The signal to noise ratio for Q control is orders of magnitude larger than that of parametric amplification. Because of this, we predict Q control will provide both better and quicker results.

UC Santa Barbara Center for Science and Engineering Partnerships UCSB California NanoSystems Institute