Study On Design and Fabricate a Mems Piezoresistive Accelerometer with Very Low Cross-Axis Sensitivity for Aircraft Sensing Application

INTRODUCTION

―Small is Beautiful.‖ The fact of this statement is discussed in economic and communal circles, but when it comes to technology, there is no debate; small is beautiful because small is fast, small is cheap, and small is profitable. Feynman termed the word ―miniaturization‖ in science and technology for our time. The technology which combines miniaturization in electronics and mechanical engineering structures is micro-electro mechanical systems (MEMS). MEMS is the technology to realize very small devices and made up of components between 1 to 100 micrometers in size. MEMS devices generally range in size from 20 micrometers to a few millimeters. These devices started finding practical applications once they could be fabricated using modified semiconductor device fabrication technologies such as bulk micromachining, surface micromachining, wafer bonding and electroplating to name a few (Madou M J et al., 2012). The application of of MEMS spans over a wide range of devices like accelerometers, gyroscopes, microphones, wearable devices and biomedical devices. An accelerometer measures the acceleration it experiences relative to a freefall. Accelerometer devices detect both the magnitude and the direction of the acceleration as a vector quantity, and is used to sense orientation, vibration and shock of various device (Mac Donald G. A., 1990). An accelerometer behaves as a damped mass on a spring and experiences an external force such as gravity, the mass is displaced until the external force is balanced by the spring force. An accelerometer fabricated by a MEMS process are mechanically similar to traditional accelerometers, but only fabricated on a micrometer scale. Advantages of MEMS sensors are small in size, light weight, low cost and low power and their ability to monolithically fabricate signal conditioning circuitry on the same die, resulting in improved sensor performance and reduced sensor costs. Accelerometers are in great demand for many applications including biomedicine,

It is desirable that accelerometers have high sensitivity and fine resolution. Various sensing techniques like capacitive, piezoelectric, tunneling, piezoresistive are used for acceleration measurement. Performance of a capacitive accelerometer is superior due to its high sensitivity, good dc response, low temperature sensitivity and low power dissipation. Capacitive accelerometers have several attractive features: In most micromachining technologies no or minimal additional processing is needed, capacitors can operate both as sensors and actuators. However, the structural configuration of the capacitive accelerometer is much more complicated than the piezoresistive devices. Moreover, capacitive accelerometer needs on-chip integration with associated signal conditioning circuits and reduces noise. Irrespective of the base material, capacitive accelerometer relies on the variation of capacitance when the geometry of the capacitor is changing. Acceleration is detected by a change of distance or change of overlapping area of the two electrodes due to the inertial force (Rocha L A et al., 2011, Sankar A R et al., 2011, Zhou X et al., 2015). Figure 1.1 shows the model of a capacitive accelerometer.

Piezoelectric accelerometers are made up of piezo ceramics (e.g., lead zirconate titanate) or single crystals (e.g., quartz, tourmaline) which produce an electrical charge when stressed. When the seismic mass experiences a vibration, piezoelectric accelerometer generates an electric charge signal proportional to the induced acceleration (Wang T et al., 2017). A model of a piezoelectric accelerometer is shown in Figure 1.2. Piezoelectric accelerometers are widely accepted as the best choice for measuring absolute vibration. They are unmatched in terms of their upper frequency range, low packaged weight and high temperature range. Tunneling accelerometers are sought to measure very fine perturbations such as detecting and identifying submarines. Tunneling tips have been utilized to increase the resolution and sensitivity of the device. By utilizing a feedback control circuit to al., 2001).

The system to measure the acceleration is shown in figure 1.3. Though the tunneling accelerometer offers high sensitivity, the damping mechanisms are very complex due to the complexity of the structure. In piezoresistive accelerometers, miniature silicon resistors are embedded at maximum stress locations in a suspended beam to sense the input acceleration. The sensing mechanism utilizes the change of resistivity of the embedded resistors to sense the acceleration (Rockstad H K et al., 1996). Required damping of a piezoresistive accelerometer can be obtained by changing the gap between the moving proof mass and top or bottom glass lids that are static. No complex process steps are required in achieving the required damping, since changing the gap between moving and fixed cover plates does not affect the device sensitivity. However, the main drawbacks of the piezoresistive sensing method are large temperature sensitivity and low overall sensitivity.

used without any signal loss. Figure 1.4 shows the classification of various accelerometers with their performance features.

Piezoresistive accelerometer was the first silicon accelerometer realized using micro machined technology (Roylance L M et al., 1979). Piezoresistive accelerometers are the most popular and widely used method of acceleration sensing due to their simplicity in fabrication, packaging and inherent ruggedness.

The first batch fabricated accelerometer made use of the piezoresistive property and was initiated by Roylance and Angell in 1979. The device is 3mm long with a mass of 0.02g. The accelerometer was able to measure accelerations ranging from 0.001g to 50g. His work motivated several researchers to develop various silicon based micro-machined piezoresistive accelerometers (Roylance L M et al., 1979).

Piezoresistive accelerometer consists of a heavy seismic mass (M) attached to a fixed frame through a spring of stiffness (K). When acceleration is applied, the mass stays in rest and the spring stretches to get sufficient energy which makes the mass to move. When there is a steady acceleration, the mass tends to move either away or towards the frame; while in oscillation the mass will tend to move in both damping force. Acceleration acting on the mass causes damping force to act, which in turn causes the mass to move up or down until the restoring force of the beam suspension equals the inertial force. In Figure 2.1, ‗M‘ represents the mass of the system; ‗K‘ represents the spring Constant ‗X‘ represents displacement and ‗D‘ represents damping Constant. The desired properties of micro-accelerometers are high sensitivity, high resolution, good linearity, low offset, low drift, low cross-axis sensitivity, low temperature sensitivity, wide operation range and frequency response, high accuracy, precision and high signal/noise ratio (Roylance L M et al., 1979).

Kampen et al. reported the mechanical behavior of the static and dynamic properties of bulk micro-machined accelerometers (Van Kampen R V et al., 1998). The static characteristics are the properties of the system after all the transient effects have settled to their final or steady state; and is determined by the mass and the stiffness of the beam suspension which includes sensitivity, accuracy, precision, linearity, etc. The stiffness of a material in a particular direction depends on (i) the number of flexures (ii) geometrical dimension on the flexures and (iii) the position of the flexures (Sankar A R et al., 2012). Equation 1.1 represents the vertical displacement of the mass subjected to normal acceleration.

The properties of the system transient response to the input are called dynamic characteristics of the device. The resonant frequency and damping determine the dynamic characteristics. It includes zero, first and second order systems (Van Kampen R V et al., 1998). The resonant frequency is governed by the Eigen values of the system and it is given in Equation (1.4),

where ‗M‘ is the mass of the proof mass, ‗az‘ is acceleration along the z-axis, ‗ay‘ is acceleration along the y-axis, ‗ax‘ is acceleration along the x-axis, ‗Zc‘ is the distance between the ‗Kɵx‘ is the spring constant of the flexure along ‗X‘ axis, ‗Kɵy‘ is the spring constant of the flexure along ‗Y‘ axis, ‗Kz‘ is the spring constant of the flexure along ‗Z‘ axis.

This research work aims to design and fabricate a MEMS piezoresistive accelerometer with very low cross-axis sensitivity for aircraft sensing application. It also focuses to solve the undercut issues which happen during micromachining and the problems related to wet anisotropic etching.

A single cantilever proof mass structure shown in figure 3.1 (a), was introduced by Roylance et al., in 1979. It deflects the 200µm thick proof mass for accelerations acting perpendicular to the proof mass plane. The stress was measured using p-type resistors diffused in the beam. However, the cross axis sensitivity measured was around 10% and the device is used for medium g accelerometers. In 1991, Tschan et al., developed a two beam accelerometer in which the beams are located at one side of the proof mass as shown in figure 46 3.1 (b). Though the cross axis sensitivity is slightly less as compared to single cantilever proof mass structure, the prime axis sensitivity is reduced by 50% and not suitable to be used for high performance placing two beams in a balanced manner at the centre of two opposite edges.

two proof mass. The figure of merit is high compared to other structures nevertheless it occupies more space. An eight beam by Jun Hwan Sim et al. has eight beams with four beams placed at the top side of the proof mass and other four beams placed in the bottom surfaces. Though the cross axis sensitivity is reduced by 50% compared to quad beam structure, micromachining the device without any undercut is quite difficult. Figure 3.1(v), shows the configuration of an eight beam structure. The device specifications considered for study is given in Table 3.1,

The above results are analyzed using ACES simulator to determine the etch time consumption for different etchants at its corresponding etchant concentration and temperature.

From the experimental analysis made by Seidal H et al., it is observed that the etchant concentration is inversely proportional to the etch rate and the temperature is directly proportional to the etch rate (Seidal H et al., 1987, Madou M J et al., 2012). Xuefeng D et al. analyzed the characteristics of square mask file which has 300µm length and 300µm breadth using TMAH etchant (Xuefeng D et al., 2005). He used a triangular strip (a small pad for corner compensation structure) along (100) direction for 60 minutes by simulating the etching process using a cellular automata method. The present study focuses to study the etching time of different compensation structures using KOH and TMAH etchants using Anisotropic Crystalline Etching Simulator (ACES). ACES takes the values of etchant concentration, temperature, the mask file, and the mode of etchant as its input and the processing time is evaluated as soon as perfect corner is obtained at the 84 edges of the corner. The size of the mask has

This present research work aspires to study and design very low crossaxis sensitivity MEMS piezoresistive accelerometer. This thesis has addressed the following topics of research (i) the simulation analysis of etch time optimization in bulk silicon MEMS devices using novel structures, (ii) realization issues of the sensor using wet anisotropic etching based bulk micromachining, and (iii) investigation of fixed quad beam accelerometer and its damping analysis.

1. Li-Xia Yu., Li Qin., and Ai-Da Bao., (2015) ―Reliability Prediction for MEMS Accelerometer under Random Vibration Testing‖ 18, pp. 41-46. 2. Hensel. J. C., and Feher. G., (1963) ―Cyclotron resonance experiments in uniaxially stressed silicon: valence band inverse mass parameters and deformation potentials‖ Physical Review 129, pp. 1041-1062. 3. Kavitha. S., Joseph Daniel. R. and Sumangala. K. (2013) ―A Simple analytical design approach based on computer aided analysis of bulk micromachined piezoresistive MEMS piezoresistive accelerometer for concrete SHM applications‖ Measurement 46, pp. 3372-3388. 4. Mac Donald. G. A. (1990) ―A review of low cost accelerometers for vehicle dynamics‖ Sensors and Actuators A 21, pp. 303-307. 5. Madou. M., J. (2012) ― Fundamentals of microfabrication‖ The science of miniaturization - CRC press. 6. Mathew. R., and Sankar R. (2015) ―Design of a triangular platform piezoresistive affinity microcantilever sensor for biochemical sensing applications‖ Journal of Physics D: Applied Physics 48, pp. 1-14. 7. Messina. M., Njuguna. J., Dariol. V., Pace. C. & Angeletti. G., (2013) ―Design and simulation of a novel biomechanic piezoresistive sensor with silicon nanowires‖ IEEE/ASME Transactions on Mechatronics 18, pp. 1201-1210. 9. Reithmiller. W., Benecke. W., Schnakenberg. U., and Wagner. B., (1992) ―A smart accelerometer with on chip electronics fabricated by a commercial CMOS process‖ Sensors and Actuators, A, 31, pp. 121-124. 10. Robert Kuells, Siegfried Nau, Manfred Salk, Klaus Thomas, (2012) ―Novel piezoresistive high-g accelerometer geometry with very high sensitivity bandwidth product‖ Sensors and Actuators A 182, pp. 41-48. 11. Rocha. L. A., Dias. R. A., Cretu. E., Mol. L., and Wolffenbuttel. R. F., (2011) ―Auto-calibration of capacitive MEMS accelerometers based on pull-in voltage‖ J. Microsystem technologies 17, pp. 429-436. 12. Rockstad. H. K., Tang. T. K., Reynolds. J. K., Kenny. T. W., Kaiser. W. J., and Gabrielson. T. B., (1996) ―A miniature, high-sensitivity, electron tunneling accelerometer‖ Sensors and Actuators A. 53, pp. 227-231.

Associate Professor, Department of Electronics, Electrical and Communications, Galgotias University, Greater Noida, Uttar Pradesh, India