Microcantilever-based force sensing, control and imaging.

摘要

This dissertation presents a distributed-parameters base modeling framework for microcantilever (MC)-based force sensing and control with applications to nanomanipulation and imaging. Due to the widespread applications of MCs in nanoscale force sensing or atomic force microscopy with nano-Newton to pico-Newton force measurement requirements, precise modeling of the involved MCs is essential. Along this line, a distributed-parameters modeling framework is proposed which is followed by a modified robust controller with perturbation estimation to target the problem of delay in nanoscale imaging and manipulation. It is shown that the proposed nonlinear model-based controller can stabilize such nanomanipulation process in a very short time compared to available conventional methods. Such modeling and control development could pave the pathway towards MC-based manipulation and positioning.;The first application of the MC-based (a piezoresistive MC) force sensors in this dissertation includes MC-based mass sensing with applications to biological species detection. MC-based sensing has recently attracted extensive interest in many chemical and biological applications due to its sensitivity, extreme applicability and low cost. By measuring the stiffness of MCs experimentally, the effect of adsorption of target molecules can be quantified. To measure MC's stiffness, an in-house nanoscale force sensing setup is designed and fabricated which utilizes a piezoresistive MC to measure the force acting on the MC's tip with nano-Newton resolution. In the second application, the proposed MC-based force sensor is utilized to achieve a fast-scan laser-free Atomic Force Microscopy (AFM). Tracking control of piezoelectric actuators in various applications including scanning probe microscopes is limited by sudden step discontinuities within time-varying continuous trajectories. For this, a switching control strategy is proposed for effective tracking of such discontinuous trajectories. A new spiral path planning is also proposed here which improves scanning rate of the AFM. Implementation of the proposed modeling and controller in a laser-free AFM setup yields high quality image of surfaces with stepped topographies at frequencies up to 30 Hz.;As the last application of the MC-based force sensors, a nanomanipulator named here MM3ARTM is utilized for nanomanipulation purposes. The area of control and manipulation at the nanoscale has recently received widespread attention in different technologies such as fabricating electronic chipsets, testing and assembly of MEMS and NEMS, micro-injection and manipulation of chromosomes and genes. To overcome the lack of position sensor on this particular manipulator, a fused vision force feedback robust controller is proposed. The effects of utilization of the image and force feedbacks are individually discussed and analyzed for use in the developed fused vision force feedback control framework in order to achieve ultra precise positioning and optimal performance.