Publications

Abstract not provided.

In this paper we describe a method for controlling both the residual stress and the through-thickness stress gradient of aluminum nitride (AlN) thin films using a multi-step deposition process that varies the applied radio frequency (RF) substrate bias. The relationship between the applied RF substrate bias and the AlN residual stress is characterized using AlN films grown on oxidized silicon substrates is determined using 100 nm-1.5 μm thick blanket AlN films that are deposited with 60-100 W applied RF biases; the stress-bias relationship is found to be well described using a power law relationship. Using this relationship, we develop a model for varying the RF bias in a series of discrete deposition steps such that each deposition step has zero average stress. The applied RF bias power in these steps is tailored to produce AlN films that have minimized both the residual stress and the film stress gradient. AlN cantilevers were patterned from films deposited using this technique, which show reduced curvature compared to those deposited using a single RF bias setting, indicating a reduction of the stress gradient in the films.

Abstract not provided.

Abstract not provided.

Substitution of Al by Sc has been predicted and demonstrated to improve the piezoelectric response in AlN for commercial market applications in radio frequency filter technologies. Although cosputtering with multiple targets have achieved Sc incorporation in excess of 40%, industrial processes requiring stable single target sputtering are currently limited. A major concern with sputter deposition of ScAl is the control over the presence of non-c-axis oriented crystal growth, referred to as inclusions here, while simultaneously controlling film stress for suspended microelectromechanical systems (MEMS) structures. This work describes 12.5% ScAl single target reactive sputter deposition process and establishes a direct relationship between the inclusion occurrences and compressive film stress allowing for the suppression of the c-axis instability on silicon (100) and Ti/TiN/AlCu seeding layers. An initial high film stress, for suppressing inclusions, is then balanced with a lower film stress deposition to control total film stress to prevent Euler buckling of suspended MEMS devices. Contour mode resonators fabricated using these films demonstrate effective coupling coefficients up to 2.7% with figures of merit of 42. Furthermore, this work provides a method to establish inclusion free films in ScAlN piezoelectric films for good quality factor devices.

Substitution of Al by Sc has been predicted and demonstrated to improve the piezoelectric response in AlN for commercial market applications in radio frequency filter technologies. Although cosputtering with multiple targets have achieved Sc incorporation in excess of 40%, industrial processes requiring stable single target sputtering are currently limited. A major concern with sputter deposition of ScAl is the control over the presence of non-c-axis oriented crystal growth, referred to as inclusions here, while simultaneously controlling film stress for suspended microelectromechanical systems (MEMS) structures. In this paper, we describe 12.5% ScAl single target reactive sputter deposition process and establishes a direct relationship between the inclusion occurrences and compressive film stress allowing for the suppression of the c-axis instability on silicon (100) and Ti/TiN/AlCu seeding layers. An initial high film stress, for suppressing inclusions, is then balanced with a lower film stress deposition to control total film stress to prevent Euler buckling of suspended MEMS devices. Contour mode resonators fabricated using these films demonstrate effective coupling coefficients up to 2.7% with figures of merit of 42. Finally, this work provides a method to establish inclusion free films in ScAlN piezoelectric films for good quality factor devices.

Abstract not provided.

Biasing a MEMS switch close to static-pull in reduces the modulation amplitude necessary to achieve resonant pull-in, but results in a highly nonlinear system. In this work, we present a new methodology that captures the essential dynamics and provides a prescription for achieving the optimal drive waveform which reduces the amplitude requirements of the modulation source. These findings are validated both experimentally and through numerical modeling.

Abstract not provided.

Abstract not provided.

Two dimensional SiC-air phononic crystals have been modeled, fabricated, and tested with a measured bandgap ranging from 665 to 693 MHz. Snowflake air inclusions on a hexagonal lattice were used for the phononic crystal. By manipulating the phononic crystal lattice and inserting circular inclusions, a waveguide was created at 680 MHz. The combined insertion loss and propagation loss for the waveguide is 8.2 dB, i.e., 39% of the energy is guided due to the high level of the confinement afforded by the phononic crystal. The SiC-air phononic crystals and waveguides were fabricated using a CMOS-compatible process, which allows for seamless integration of these devices into wireless communication systems operating at microwave frequencies.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Recent literature has focused on improving piezoelectric coupling coefficients by alloying aluminum nitride (AlN) with scandium (Sc). Akiyama et al. showed the highest d-33 piezoelectric coefficient increase of >4x at a 41% Sc substitution for Al. Thus far, studies mainly focus on material measurements such as x-ray diffraction or piezoelectric constants to assess the material quality. Although these measurements are useful to assess the improvement in the piezoelectric performance of the material, they do not address improvements in the figure-of-merit (FOM) of resonators (i.e., coupling coefficient times quality factor). Resonator structures are needed to directly extract these key performance parameters for film assessment. Fabrication integration, however, must be minimized to avoid obscuring film performance by extrinsic device effects.

Abstract not provided.

Abstract not provided.

Abstract not provided.

This work reports the utilization of a recently developed film, ScAlN, as a silicon etch mask offering significant improvements in high etch selectivity to silicon. Utilization of ScAlN as a fluorine chemistry based deep reactive ion etch mask demonstrated etch selectivity at 23 550:1, four times better than AlN, 11 times better than Al2O3, and 148 times better than silicon dioxide with significantly less resputtering at high bias voltage than either Al2O3 or AlN. Ellipsometry film thickness measurements show less than 0.3 nm/min mask erosion rates for ScAlN. Micromasking of resputtered Al for Al2O3, AlN, and ScAlN etch masks is also reported here, utilizing cross-sectional scanning electron microscope and confocal microscope roughness measurements. With lower etch bias, the reduced etch rate can be optimized to achieve a trench bottom surface roughness that is comparable to SiO2 etch masks. Etch mask selectivity enabled by ScAlN is likely to make significant improvements in microelectromechanical systems, wafer level packaging, and plasma dicing of silicon.

Abstract not provided.

With the advent of the internet-of-things, sensors that are constantly alert yet consuming near-zero power are desired. Remote sensing applications where sensor replacement is costly or hazardous would also benefit. Piezoelectric micro-electro-mechanical systems (MEMS) convert mechanical or acoustic energy into electrical signals while consuming zero power. When coupled with low-power complementary metal-oxide-semiconductor (CMOS) circuits, a near-zero power sensing system is formed. This work describes piezoelectric MEMS microphones based on aluminum nitride (AlN). The microphones operate as passive acoustic filters by placing their resonant response within bandwidths of interest. Devices are demonstrated with operational frequencies from 430 Hz to greater than 10 kHz with quality factors as large as 3,000 and open-circuit voltages exceeding 600 mV/Pa.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Many piezoelectric microelectromechanical systems (MEMS) measure or generate acoustic signals via the motion of radially non-uniform, thin film composite plates. The composite layers provide piezoelectric actuation, structural support, electrode metallization, passivation, etc. Often, the layers are non-uniform over the plate and contain residual stresses introduced during the fabrication process. Accurate models of non-uniform composite plate mechanics are crucial for predicting and optimizing device performance. In this paper, an analytical solution for a radially non-uniform, piezoelectric, circular composite plate incorporating residual stress is derived. The analytical solution is compared to experimental measurements of a MEMS piezoelectric diaphragm. The results show the improved accuracy of the analytical model when including film stress, the speed of the analytical solution as compared to finite element analysis, the sensitivity of device performance to residual stress and the importance of accurate film stresses as model inputs. Finally, the analytical model presented is useful as a design optimization tool given the efficiency of the computational time, shown to be 275 times less than a comparable finite element analysis.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Localized stress variation in aluminum nitride (AlN) sputtered on patterned metallization has been monitored through the use of UV micro-Raman spectroscopy. This technique utilizing 325 nm laser excitation allows detection of the AlN E2(high) phonon mode in the presence of metal electrodes beneath the AlN layer with a high spatial resolution of less than 400 nm. The AlN film stress shifted 400 MPa from regions where AlN was deposited over a bottom metal electrode versus silicon dioxide. Across wafer stress variations were also investigated showing that wafer level stress metrology, for example using wafer curvature measurements, introduces large uncertainties for predicting the impact of AlN residual stress on the device performance.

Abstract not provided.

A single-axis MEMS wall shear stress sensor with differential capacitive transduction method is produced. Using a synchronous modulation and demodulation interface circuit, the system is capable of making real time measurements of both mean and fluctuating wall shear stress. A sensitivity of 3.44 mV/Pa is achieved, with linearity in response demonstrated up to testing limit of 2 Pa. Minimum detectable signals of 340 μPa at 100 Hz and 120 μPa at 1 kHz are indicated, with a resonance of 3.5 kHz. Multiple full scale wind tunnel tests are performed, producing spectral measurements of turbulent boundary layers in wind speeds ranging up to 0.5 Ma (18 Pa of mean wall shear stress). The compact packaging allows for minimally invasive installation, and has proven relatively robust over multiple testing events. Temperature sensitivity, likely due to poor CTE matching of packaged materials, is an ongoing concern being addressed. These successes are being directly leveraged into a development plan for a dual-axis wall shear stress sensor, capable of producing true vector estimates at the wall.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

After years in the field, many materials suffer degradation, off-gassing, and chemical changes causing build-up of measurable chemical atmospheres. Stand-alone embedded chemical sensors are typically limited in specificity, require electrical lines, and/or calibration drift makes data reliability questionable. Along with size, these "Achilles' heels" have prevented incorporation of gas sensing into sealed, hazardous locations which would highly benefit from in-situ analysis. We report on development of an all-optical, mid-IR, fiber-optic based MEMS Photoacoustic Spectroscopy solution to address these limitations. Concurrent modeling and computational simulation are used to guide hardware design and implementation.

Abstract not provided.

Abstract not provided.