Search

Resonant sensors can detect chemical and biological analytes by measuring shifts in the resonant frequency due to adsorption-induced mass changes. This work describes a sensing approach that provides the sensitivity of nanoscale devices without sacrificing capture area through the use of highly porous surface so as to increase the adsorbed mass. Three strategies for exceptional capturing area are studied to achieve high sensitivity gas sensors; (1) porous-etched silicon resonator, (2) ZIF (zeolitic imidazolate framework)-agglomerated resonator by drop casting, and (3) ZIF-coupled resonator by dielectrophoresis. Microscale silicon resonators with nanoscale pores are developed for increased surface area. Increased mechanical stability and detection performance are also achieved by keeping parts of the resonating device nonporous and adding a receptor coating. Partially-porous silicon resonators and receptor-coated partially-porous silicon resonators are improved up to 165% and 654% in resonator sensitivity, respectively, as compared to nonporous silicon resonators. This work also presents ZIF-agglomerated resonators whose sensitivity shows an improvement up to 78 times (780% improvement) over the silicon resonators with identical dimensions while additionally utilizing the inherently selective adsorption properties of ZIFs. ZIF nanoparticles provide previously unattainable surface area as well as the ability to tailor crystal structure for inherent selectivity. A unique fabrication technique in combination with a drop casting method is presented, enabling deposition of a wide variety of materials on released MEMS devices by providing a temporary support layer of photoresist.Lastly, ZIF nanoparticles are coupled to resonators using dielectrophoresis (DEP) to maximize adsorption of ZIFs so as to gain further sensitivity enhancement. By utilizing an inherently sensitive and selective adsorption property of ZIFs, amplitude of frequency shift shows a sensitivity improvement up to 158 times over the silicon resonator. Also, consistent decay constant of the frequency shift provide significant chemical recognition ability of the ZIF-coupled resonant sensor.

Using available digital seismic stations deployed since the 1980’s, the largest data set based on broadband waveforms among studies on body-wave attenuation (t*) and quality factor (Q) are used in this thesis. The use of nearly 300,000 measurements of body-wave spectral ratio from globally distributed stations renders better constraints of t* and Q variations with higher spatial and depth resolutions in the mantle than have been previously available. The maps of body-wave t* correlate well with the variations of t* computed from the most recent surface-wave Q model QRFSI12 indicating that body-wave and surface-wave t* reflect the same intrinsic attenuation even though these waves sample the upper mantle entirely differently. The high correlation between body-wave t* maps and the t* inferred from a thermal interpretation of shear-wave velocity tomography S20RTS suggests that temperature controls both variations in attenuation and velocity in the upper mantle. The distance variations of P- and S-wave t* (t*P and t*S) are inverted for a radial profile of the quality factor Qμ in the lower mantle. On average, t*P and t*S increase by about 0.2 s and 0.7 s, respectively, between epicentral distances of 30 degree and 97 degree. The body-wave spectra are explained best if Qμ increases in the lower mantle with the rate of 0.1/km. The relatively strong increase of t*S compare to t*P (t*S ~ 4 t*P ) suggests that intrinsic attenuation is the cause of the overall trend in our data. The ratio of P- and S-wave quality factor determined in this thesis (QP/Qμ = 2.27) confirms that intrinsic attenuation occurs mostly in shear and that bulk attenuation is negligible in the mantle. Finally, the delay of seismic waves which traversed numerical mantle plumes are calculated in this thesis for the first time. High-resolution numerical simulations of mantle plume are used to investigate the effects of numerical plumes on waveforms. The measurements of wave front delay demonstrate that the delay of shear-waves by plume tails at depths larger than 1000 km are immeasurably small (< 0.2 s) at seismic periods commonly used in waveform analysis.