Objectives: In this proposal we will provide the enabling science and technical underpinnings that will move sensing with N/MEMS beyond the realm of a scientific curiosity. We will investigate the dissipation of diverse resonant N/MEMS operating in air, leading to optimized performance in air, the carrier gas from which analytes will be sensed. We will also enable and demonstrate sensing of volatile organics.
The intellectual merit rests on our three pronged approach: 1. To reduce radiated sound energy that damps the motion of resonators in air, making the resonant frequency shift a more sensitive transduction mechanism. 2. To investigate tensile stress to increase the Q in polysilicon and other low loss materials and also expose the origin of the enhancement of Q by stress. 3. To explore stress-based sensing where an analyte in the air induces a change in stress in a functionalizing coating on a surface
The broader impacts for the proposed research extend to high Q systems under investigation for quantum computation as mechanical qubits (the Q providing a high fidelity). Our research will enable diverse sensing functions for security and threat recognition from chemical and biological agents. Other benefits will flow from our planned REU and K-12 initiatives. The integration of undergraduates in aspects of the research and links to applications will benefit a broad spectrum of junior scientists, and equip them with practical problem-solving knowledge that extends beyond the classroom or laboratory environment.
The project advanced the state of the art in several important areas. In the technology arena, we conceived of, designed and actualized high tensile stress silicon nitride resonators with varied geometries. We wanted to test recent theoretical models that laid out how the mechanical loss was affected by various geometrical or mechanical factors, for example diameter, thickness and mode number. In general, high stress devices show much sharper resonance frequency (an indication of lower energy loss), and larger devices have lower loss or higher Quality (Q) factor. We showed that the expected diameter dependence is experimentally realized, that the mode dependence is also verified, validating the concept that interference effects from adjacent regions leads to a diminution of mechanical loss. We also came up with an innovative way to fabricate devices whose thickness could be controlled by resorting to a timed etch. Thin devices were shown to also exhibit record high Q factors. These structures will undoubtedly find uses in signal processing and metrology as well as quantum measurement technology in the future. A second technology development was in the area of stress based sensing. In the past stress based sensing of the presence of gasses in ambient air was demonstrated using the familiar bimetallic strip idea where a bilayer is formed of poly silicon and a porous polymer. When exposed to trace gasses in ambient air the polymer swells leading to a bending off the cantilever beam. Our innovation was to take this scheme out of the static case to the dynamic regime where we measure the resonant frequency of an oscillator. In our version, the underlying polysilicon is grown with inbuilt compressive stress sufficient to just induce buckling. Any further change in stress leads to a very rapid evolution in resonant frequency increasing the ability to resolve changes in gas concentrations. We demonstrated the sensitivity to ethanol vapor. Our scientific activities were broad. In addition to the landmark work where we demonstrated an understanding of the source and mechanisms of mechanical energy loss that limit the performance of resonators operating in vacuum, we also successfully modeled the loss of mechanical energy to the air surrounding mechanical resonators used in sensing applications. This can take two forms, squeeze film losses where air trapped in the gap between a shaking structure and the wall is squeezed and provides a loss mechanism. The second mechanism is acoustic radiation. Both are familiar in macroscopic applications but their impact on resonators operating in air was not quantified. Importantly the role of adjacent regions that move out of phase to one another (eg at higher order resonant modes) can provide opportunities for the cancellation of loss, thus improving the sensing capability of these devices. The broader impacts of the research can be considerable. High Q devices are highly prized in quantum measurements because they retain their coherence, and our research lays the way for understanding and ready utilization of these devices. Stress based sensing paves the way for applications involving the sensing of breath gas (for example) whereby human health of an individual can be readily monitored in a non invasive manner by having the individual exhale into an instrument (perhaps an addendum to a cell phone). By monitoring changes in blood-breath chemistry the individual might receive an early warning of an impending health issue. Such delocalized and ubiquitous monitoring might lead to a revolution in the practice (and cost) of medicine. In addition, our students have benefited from international collaborations and have taken up positions in university research (at Columbia University in NY). The PIs mentored REU students and Cornell undergraduates and developed presentations to propagate the concepts of order-of-magnitude estimation to encourage students to carry out simple estimations using basic scientific principles – thus thinking creatively to examine and solve complex problems that they may encounter in all spheres of activity. Research material was also incorporated into a course taught at Cornell.
