Yupeng Jing

Research on the resonant frequency formula of V-shaped cantilevers

This paper deduces a remarkably precise analytical formula for calculating the fundamental resonant frequency of V-shaped cantilevers using Rayleigh-Ritz method. This analytical formula, which is very convenient for MEMS sensor design, is then validated by ANSYS simulation. This formula raises a new perspective that, among all the V-shaped cantilevers, the simplest triangular cantilever can lead to maximum resonant frequency and highest sensitivity.

Design of a Novel Substrate-Free Double-Layer-Cantilever FPA Applied for Uncooled Optical-Readable Infrared Imaging System

This paper describes the design and performances of a novel focal-plane array (FPA) containing pixels of double bimaterial-layer cantilevers without silicon (Si) substrate for being applied in the uncooled optical-readable infrared (IR) imaging system. The top layer of the cantilever pixels is made of two materials with large mismatching thermal expansion coefficients: silicon nitride (SiNx) and gold (Au), which convert IR heat into mechanical deflection. The bottom layer is SiNx cantilever, which partially serves thermal isolation legs. The top and bottom pads form the resonant cavity, which can dramatically enhance the absorption of incident IR irradiation, and the substrate-free configuration enables reducing the loss of incident IR energy. Responding to the IR source with spectral range from 8 to 14 mum, the IR imaging system may receive an IR images through visible optical readout method. A thermal-mechanical model for such cantilever microstructure is proposed, and the thermal and thermal-mechanical coupling field characteristics of the cantilever microstructure are optimized through numerical analysis method and simulation by using the finite-element method. The thermal-mechanical deflection simulated is 7.2 mum/K, generally in good agreement with what the thermal-mechanical model and numerical analysis forecast. The analysis suggests that the detection resolution of current design is 0.03 K, whereas the noise analysis from FPA indicates the current resolution to be around 100 muK and the limit noise-equivalent temperature difference (NETD) of the IR imaging system can reach to 7 mK.

Simple sticking models and adhesion criterion to predict sticking effects of fixed-fixed beams in RF MEMS switch design

The fixed-fixed beam in RF switches fabricated by MEMS techniques of a sacrificial layer is inclined to stick to the substrate due to the significant capillary force during drying process. In this paper, sticking models of the fixed-fixed beam are investigated to analyze the sticking effect. Based on these models, an adhesion criterion hc for the fixed-fixed beam is derived to determine whether the beam sticks to the substrate or not. The deduced formula shows that hc is determined by the material property of the beam and its length L and thickness t. It is also found that the width of the beam has no effect on hc. The theoretical models and adhesion criterion are simulated and verified by Finite Element Analysis. The effects of residual stress are also discussed to meet the uncertainty of real fabrication processes.

Two microthermal shear stress sensors: surface micromachined and bulk-bonding micromachined

We describe two fabricated microthermal shear stress sensors by antiadhesion surface technology and anodic bulk-bonding technology. Two sensors are based on thermal transfer principles with adiabatic structures. The thermal sensor element is a titanium—platinum alloy resistor sputtered on the top of a low pressure chemical vapor deposited (LPCVD) silicon nitride diaphragm with an adiabatic vacuum cavity underneath. The surface micromachined thermal shear stress sensor uses microbumps on the silicon substrate in the sacrificial layer technology to prevent the silicon nitride diaphragms stiction to the substrate. Microbumps formed by isotropic silicon etching in HNA (the system HF, HNO3, and HC2H3O2) are arrayed in several points on the silicon substrate with distances of 147 µm in the (200×250)-µm2×1.5-µm vacuum cavity. This cavity is formed by LPCVD silicon nitride film sealing with 30-Pa vacuum degree. The anodic bulk-bonding micromachined thermal shear stress sensor uses bulk silicon substrate etching and anodic bonding to form the (200×250)-µm2×400-µm high aspect ratio cavity with 5×10−2 Pa vacuum degree. The titanium platinum alloy resistor, (5×150)-µm2×0.2 µm, sputtered on the top of the 1.5-µm-thick LPCVD silicon nitride diaphragm with this bonding chamber, has a temperature coefficient of resistance (TCR) value of 0.33%/°C. According to the comparison of the adiabatic characteristics among three cases—a titanium platinum alloy resistor located over the high aspect ratio 5×10−2 Pa vacuum cavity, over the 30-Pa vacuum cavity, and directly on top of the substrate—the first case has the best adiabatic characteristic: the titanium platinum alloy resistor located over the 5×10−2-Pa vacuum cavity has the maximum thermal resistance of 5362 °C/W. Besides the sensor sensitivity performances, it has a comparatively short time constant with value of 0.1 ms under the constant current (CC) mode driving circuit.

Circuit models applied to the design of a novel uncooled infrared focal plane array structure

This paper describes a circuit model applied to the simulation of the thermal response frequency of a novel substrate-free single-layer bi-material cantilever microstructure used as the focal plane array (FPA) in an uncooled opto-mechanical infrared imaging system. In order to obtain a high detection of the IR object, gold (Au) is coated alternately on the silicon nitride (SiNx) cantilevers of the pixels (Shi S et al Sensors and Actuators A at press), whereas the thermal response frequency decreases (Zhao Y 2002 Dissertation University of California, Berkeley). A circuit model for such a cantilever microstructure is proposed to be applied to evaluate the thermal response performance. The pixels thermal frequency (1/τth) is calculated to be 10 Hz under the optimized design parameters, which is compatible with the response of optical readout systems and human eyes.

Failure analysis of uncooled infrared focal plane array under a high-g inertial load

This paper describes the failure analysis of an uncooled infrared focal plane array (IRFPA) under a high-g inertial load system using finite element simulation and experimental validation methods. The uncooled IRFPA, responding to a source of infrared (IR) radiation with spectral range from 8 µ mt o 14µm, is a cantilever array, which consists of two materials with mismatched thermal expansion coefficients. The radiance distribution of the IR source could be obtained by measuring the thermal–mechanical rotation angle distribution of every pixel in the cantilever array using a visible optical readout method. Based on this principle, room-temperature infrared imaging was developed under a static gravity environment, as described in our previous paper (Li C et al 2006 Meas. Sci. Technol. 17 1981–6). But under a dynamic inertial load, the rotation angle of every pixel includes not only the thermal–mechanical part but also a part induced by the inertial load. In the elastic deformation range, with a linearly increasing acceleration, the deformation angle induced by the inertial load increases linearly, which is validated by finite element simulation. This linear change in deformation, which can be subtracted from the total rotation angle in the optical readout using certain arithmetic, will not influence the imaging result. It is noteworthy that failure stress will occur when the deformation angle induced by the inertial load moves into the plastic deformation range, and the optical readout cannot image the IR object. Through finite element simulation the critical load resulting in IRFPA failure is 2715g, and this can be validated through impact using a Hopkinson bar after the IRFPA is placed in vacuum. By finite element simulation, the initial IRFPA surface profile without IR radiance after the 2715g load showed a conicoid characteristic. Simulation of the failure analysis of the uncooled IRFPA under 2715g acceleration predicts the military application of IRFPAs for an uncooled infrared imaging system in the high-g tactical range.