MEMS Mechanical Sensors - Stephen Beeby Part 8 doc

maximize the sensitivity of the gauge. Metal gauges can be incorporated onto the diaphragm face by bonding foil gauges or by depositing and patterning insulator and metal materials using thin-film techniques such as sputtering or CVD [5]. Another resistive approach is the use of screen printed thick-film strain gauge resis - tors. These can be printed on the top surface of a metal diaphragm, previously coated with a printed dielectric layer, and offer improved sensitivity compared with bonded strain gauges. Maximum resistive strain gauge sensitivity can be achieved by bonding a silicon strain gauge to the metal diaphragm. This approach utilizes the piezoresistive nature of silicon, which increases the output of the strain gauge for a given deflection. The relative merits of these resistive methods and their associated gauge factors are discussed in Chapter 5. Other transduction techniques include capacitance, inductance, reluctance, and piezoelectric. The capacitive approach uses the diaphragm as one electrode of a parallel capacitor structure. Diaphragm displacement causes a change in capaci - tance between it and a fixed electrode. Inductance can be used to monitor the displacement of the diaphragm by mechanically linking it to the core of a linear vari - able differential transformer (LVDT). This consists of a symmetrical arrangement of a primary coil and two secondary coils. Movement of the magnetic core causes the mutual inductance of each secondary coil to vary relative to the primary. Variable reluctance transducers remove the mechanical link to the core and use the perme- ability of the diaphragm material itself to alter the inductance within two coils posi- tioned on either side of the diaphragm. The coils are typically wired in an inductive half bridge, and a change in inductance alters the impedance of each coil unbalanc- ing the bridge. Unbalances result in the ac drive signal being coupled across to the output, and the physical arrangement is suitable for differential pressure-sensing applications. Piezoelectric pressure sensors utilize a piezoelectric sensing element mechanically linked to the diaphragm. Movements in the diaphragm induce a strain in the piezoelectric and hence a charge is generated. These sensors are only suitable for measuring dynamic pressures and are not suitable for static applications because piezoelectric materials only respond to changing strains. 6.5 MEMS Technology Pressure Sensors Research into solid-state pressure sensors began as far back as the 1960s [6–8]. Since then there have been many developments both in micromachining and sensing tech - niques, which have enabled MEMS pressure sensors to mature into a commercially successful solution for many sensing applications. The mechanical sensor element is typically (but not exclusively) a micromachined diaphragm. This section commences with a brief analysis of rectangular silicon diaphragms. The different sensing princi - ples employed to date will be introduced and illustrated with both commercially available and research based devices. Finally, the state of the art in micromachined pressure sensor technology will be discussed. 6.5.1 Micromachined Silicon Diaphragms MEMS pressure sensors typically employ a diaphragm as the sensor element. This is because of its compatibility with a range of bulk and surface silicon micromachining 130 Pressure Sensors processes. The most common fabrication method is anisotropic wet silicon etching, which allows good control over diaphragm dimensions and is a batch process capa - ble of producing hundreds of devices simultaneously across a group of wafers. When combined with a (100) wafer orientation, a wet potassium hydroxide (KOH) etch, for example, produces a rectangular diaphragm with sloping side walls that follow the (111) planes. A cross-section of a typical diaphragm is shown in Figure 6.14. Diaphragm thickness can be controlled by timing etch duration, or more precisely by using boron doping or electrochemical etch stops. Surface micromachining techniques are becoming increasing applied since they offer the opportunity for reduced device size and compatibility with integrated electronics. When modeling complex micromachined structures, finite element (FE) pack - ages such as those described in Chapter 3 are normally employed. Diaphragms represent one of the few MEMS structures that can be modeled analytically. Since the diaphragm is rectangular, the characteristic equations will differ from those describing the circular case above. The characterizing equations for a rectangular diaphragm, where a is the length of the shorter side, and with rigidly clamped edges and small deflections are given next. ()y Pa Eh 0 4 3 2 1=         −αν (6.34) σβ=         Pa h 2 2 (6.35) For a rectangular diaphragm, the coefficients α and β depend upon the ratio of the lengths of the diaphragm sides and the position of interest. Assuming a square diaphragm, α equals 0.0151, and β equals 0.378 for the maximum stress that occurs along the edge of the diaphragm and 0.1386 for the maximum stress at the center of the diaphragm. Bossed diaphragms can also be fabricated using both anisotropic and isotropic etching. Such structures are typically modeled using FE techniques [9]; however, Sandmaier has presented a set of analytical equations enabling basic optimization of diaphragm design [10]. Corrugated silicon diaphragms have been discussed in the papers by van Mullem et al. [11] and Jerman [12]. The analytical equations presented in Section 6.4.5 provide an adequate approximation to the silicon case. The dynamics of a micromachined diaphragm can be adequately characterized by linear plate theory. The undamped resonant frequency f n of a clamped square diaphragm of uniform thickness and homogenous material is given by [13] 6.5 MEMS Technology Pressure Sensors 131 54.7 o a/2 h Figure 6.14 Anisotropically etched silicon diaphragm. ()f E ha n =−       1654 1 22 1 2 . ρ ν (6.36) The amount of damping present will depend not only on the diaphragm design but also its packaging and surroundings. As a rough guide, resonant frequencies of typical diaphragms should range between ~80 kHz for a 1-bar device to 575 kHz for a 40-bar device [14]. Higher frequency devices have been developed; for example, the Entran EPIH Micro Miniature range high-frequency pressure sensor series offers a maximum resonant frequency of 1.7 MHz for the 20-bar device [15]. For this series, the pressurized media is in direct contact with the micromachined silicon structure, and therefore it is suitable only for dry gas or some noncorrosive fluid applications. The introduction of a stainless steel barrier diaphragm lowers the reso - nant frequency to 45 kHz for a 17-bar device [16]. 6.5.2 Piezoresistive Pressure Sensors The piezoresistive nature of silicon makes the use of diffused or implanted resistors an obvious and straightforward technique for measuring the strain in a micromachined silicon diaphragm. The piezoresistive effect of silicon was first exploited by bonding silicon strain gauges to metal diaphragms [7], but this is an unsatisfactory approach given the thermal mismatch between the metal, adhesive layer, and silicon. Diaphragms were first micromachined into the silicon itself by mechanical spark erosion and wet isotropic etching [8]. This was not a batch approach and therefore device costs were high. The use of anisotropic etching, anodic and fusion bonding, ion implanted strain gauges, and surface micromachining have since reduced the size and improved the accuracy of piezoresistive pressure sensors. A cross-section and plan view of a typical anisotropically etched silicon piezore- sistive pressure sensor is shown in Figure 6.15. The diaphragm is etched as described above and the resistors are located along the edge of the diaphragm, one on each side. The resistors are all orientated in the same direction, and therefore, two are in parallel with the maximum strain (R l ) and two are perpendicular (R t ). The change in resistance of each resistor is calculated from (5.10). The piezoresistive coefficients associated with these resistors will depend upon the orientation of the wafer and dia - phragm, the type and amount of doping, and the temperature. Given a (100) wafer, 132 Pressure Sensors Implanted Etched silicon diaphragm Glass silicon constraint Drilled or etched pressure port R t R l Figure 6.15 Cross-section and plan view of a typical bulk micromachined piezoresistive pressure sensor. the edges of the diaphragm will be in the (110) directions. The piezoresistive coeffi - cients of p- and n-type silicon are presented graphically by Kanda [17]. Assuming p-type doping, which produces the largest and most linear piezoresistive effect, π l and π t are equal and opposite at +/−69 m 2 /N, respectively. From (5.10) it can be seen that the resistor orientation shown in Figure 6.15 will produce equal and oppo - site changes in the resistance of the two pairs of resistors. Placing the two pairs of resistors on opposite sides of a full bridge circuit will therefore maximize the sensi - tivity of the sensor to strains arising from pressure induced deflection of the dia - phragm. The stress can be calculated from (6.35) and for a full bridge the fractional bridge output is given by (6.37). This is the most common resistor arrangement and has been modeled analytically extensively [18–20]. ()() ()() ∆ ∆∆ ∆∆ V V RR RR RR RR lt lt       = − ++2 (6.37) Piezoresistive pressure sensors in the form described above have been commer - cially available for many years. Manifold absolute pressure sensors are an estab - lished application of these devices in the automotive industry. An example of such a device has been developed by Motorola and has been described in detail in [21]. Other, more recent automotive applications based upon piezoresistive sensing include diesel injection pressure [22] and exhaust gas recirculation systems [23]. Circular diaphragms are less common and have been analyzed by Matsuoka et al. [24]. Variations on the theme involve changes to the diaphragm structure (including bossed and ribbed diaphragms), temperature compensation techniques, and the use of alternative fabrication processes. Modifications to the basic diaphragm structure have been investigated in order to improve the linearity and sensitivity of the sensors. Bossed diaphragms have been fabricated using anisotropic etching processes that incorporate the rigid center seen on traditional diaphragms [9, 25]. This approach enables a resistor layout shown in Figure 6.16, which enables equal and opposite strains to be experienced by the inner and outer resistor pairs. This arrangement improves the nonlinearity of the dia - phragm in both directions, making it suitable for differential applications [26]. Another design uses a double boss at the diaphragm center [27] while researchers at Honeywell have used FE techniques to design a ribbed and bossed diaphragm [28]. The Honeywell device takes a standard diaphragm anisotropically etched from the 6.5 MEMS Technology Pressure Sensors 133 Boss Resistors Diaphragm Figure 6.16 Resistor placement on a bossed diaphragm. back and patterns the bosses and ribs on the front of the diaphragm. The resistors were positioned in the standard layout (Figure 6.15) and were located on the top surface of the rib, which served to magnify the stress by removing the resistor further from the neutral axis. The bosses were stiffened regions along each side of the dia - phragm leaving the center unstiffened like a standard diaphragm. Meandering resistors have also been applied to basic and bossed diaphragms [29]. The meander incorporates different levels of doping in each direction, which maximizes the strain sensitivity of the resistor. The meander pattern increases the length of the resistor, and this approach improves sensitivity compared with stan - dard resistors. The temperature cross-sensitivity is an obvious drawback of silicon piezoresis - tors. The change in resistance due to temperature will often exceed that arising from the change in the measurand. Several techniques are therefore employed to compen - sate for temperature. The first technique arises from the use of a full bridge with the resistors arranged as shown in Figure 6.15. In such an arrangement the change in temperature is a common mode effect acting on all resistors simultaneously, and therefore, the temperature effects should cancel out. Due to manufacturing toler - ances, however, the temperature coefficients of each resistor will invariably be slightly different. The change in resistance due to temperature and its resulting effect on the output of the bridge can be expressed in the following equations [30]: () () ()RT R T T=++01 2 αβ (6.38) () () () () () [] ()() [] () () () () [] ∆VT V RR RR TT RR RR A 012 12 2 12 12 2 12 12 2 00 00 00 00 = + ×− +− − αα ββ ()() [] ×− +−αα ββ 12 12 2 TT (6.39) The incorporation of a temperature sensor onto the sensor chip can enable temperature compensation via a look-up table or algorithm. Such an approach, however, requires extensive temperature and pressure calibration, which is a time consuming and expensive operation. An alternative technique is to include a dummy bridge on the sensor chip in addition to the pressure sensitive bridge. The dummy resistors should be positioned at least 100 µm away from the edge of the diaphragm to ensure they do not experience any pressure-induced stresses [31]. This compensa - tion technique has been applied with the dummy resistors arranged in either a full bridge [29] or a half-bridge [32]. The temperature limits of the implanted piezoresis - tive approach are approximately 120°C due to the limitations of the p-n junction. This temperature limit can be extended by using doped polysilicon resistors depos - ited on the top surface of the diaphragm. Polysilicon resistors are, however, less sen - sitive to applied stress (see Chapter 5). Over the years, developments in materials and fabrication processes have also had an effect on piezoresistive pressure sensors. Silicon fusion bonding, for example, has enabled a reduction in chip size by enabling a diaphragm wafer to be bonded to the back of an anisotropically etched cavity as shown in Figure 6.17 [33]. The use of SOI wafers has improved performance in several ways. The buried oxide can act as an etch stop, facilitating fabrication [34] and precisely controlling the diaphragm 134 Pressure Sensors thickness, or as an electrical insulator, enabling higher temperature operation [35–37]. Ultimate high-temperature operation of piezoresistive pressure sensors has been developed using micromachined silicon carbide [38]. The diaphragms are etched by a photoelectrochemical process in a diluted HF etchant. A prototype device has been demonstrated operating at 600°C [39] and in a dynamic sensing application on a gas turbine engine [40]. Finally, silicon nitride diaphragms have been realized by bulk wet anisotropic etching. The nitride membrane is formed by wet etching through the silicon entirely from the back of the wafer. The wet etch stops upon reaching the nitride, and the piezoresistors are protected due to the high-dose boron implant used to define them [41]. Nitride membranes are stronger than their silicon counterpart but may suffer from in-built stresses due to the deposition process. The need to reduce the size of devices, and therefore the cost of production, has led to the use of surface micromachining to fabricate the mechanical sensing ele- ment and resistors [42]. In addition to reduced size, surface micromachining is more compatible with IC fabrication technology. It is a flexible fabrication approach ena - bling the diaphragm to be fabricated from a range of deposited materials such as polysilicon [43] and silicon nitride [44]. In both cases an underlying sacrificial layer is removed. For the polysilicon diaphragm the sacrificial material is silicon dioxide and a wet etch is used to remove it. The nitride membrane uses a polysilicon sacrifi - cial material. In both cases the lateral dimensions of the membrane are defined by previous patterning of the oxide, or doping of the polysilicon, respectively. Both devices use polysilicon resistors to sense diaphragm deflections. Both are absolute pressure sensors since a CVD process is used to deposit nitride to seal sacrificial etch holes. The vacuum used in the CVD process is therefore trapped in the sealed volume under the diaphragm. A cross-section of each device is shown in Figure 6.18. Other examples of surface-micromachined piezoresistive pressure sen - sors include a cardiovascular pressure sensor for measurement of blood pressure inside coronary arteries [45]. This is based on a square polysilicon diaphragm with edge lengths of 103 µm with a vacuum-sealed cavity underneath. One polysilicon resistor is used to detect the deflection of the diaphragm, and a second dummy resis - tor is used for temperature compensation. As discussed in the earlier analysis, the boundary conditions of the diaphragm will play an important role in the behavior of the diaphragm. With surface 6.5 MEMS Technology Pressure Sensors 135 Resistors Diaphragm Pressure Figure 6.17 Fusion bonded piezoresistive pressure sensor. micromachining there are more variations in the nature of the clamping at the edge of the diaphragm. Depending on the profile of the sacrificial layer, the dia - phragm could be flat along its entire length [Figure 6.19(a)] or have a step at the edge from where the diaphragm material was deposited over the sacrificial layer [Figure 6.19(b)]. Flat membranes have been found to be preferable since the stepped structure exhibits inferior drift characteristics [46]. The extra flexibility offered by surface micromachining has also enabled more complex pressure-sensing structures to be realized. An example of this is a duel beam pressure sensor, which couples the diaphragm deflection to a cantilever beam. A polysilicon piezoresistive strain gauge is located on the top surface of the cantilever, as shown in Figure 6.20 [47]. The cantilever, and its attachment to the underside of the diaphragm, acts as a mechanical lever, amplifying the strain experienced by the piezoresistor compared to straightforward mounting on the diaphragm. For 136 Pressure Sensors Figure 6.18 Surface-micromachined pressure sensors with (a) nitride and (b) polysilicon diaphragms. ( a ) (b) Figure 6.19 Diaphragm edge conditions: (a) flat diaphragm, and (b) stepped diaphragm. Piezoresistor Vacuum cavity Dummy beam Diaphragm Cantilever Metal contact Figure 6.20 Dual beam pressure sensor configuration. temperature compensation, a second beam with piezoresistor is positioned along - side but not coupled to the diaphragm. The diaphragm is a polysilicon layer that coats the entire chip surface (except bond pads), thereby physically, electrically, and thermally isolating the strain gauges and beams from the pressurized media. 6.5.3 Capacitive Pressure Sensors Capacitive pressure sensors are typically based upon a parallel plate arrangement whereby one electrode is fixed and the other flexible. As the flexible electrode deflects under applied pressure, the gap between electrodes decreases and the capacitance increases. The principles of capacitive sensing have been described in Chapter 5. Capacitive pressure sensors were first developed in the late 1970s and early 1980s [18, 48]. An early device, shown in Figure 6.21, consists of an ani - sotropically etched silicon diaphragm with the fixed electrode being provided by a metallized Pyrex 7740 glass die [49]. The glass and silicon die were joined using anodic bonding at die level. This device demonstrated the main attractions of capacitive sensing, these being high sensitivity to pressure, low power consumption, and low temperature cross-sensitivity. The combination of materials and bonding mechanisms demonstrated remain a common choice for capacitive sensors [50, 51]. All silicon devices fabricated by silicon fusion bonding [52, 53] and glass frit bond- ing [54] have also been reported along with many surface-micromachined devices, which are discussed below. An example of an all-silicon fusion bonded device is a vacuum sensor developed by NASA [55]. This sensor uses a circular diaphragm and demonstrates a sensitivity of ∼1 pF mbar –1 . Quartz has also been used to realize micromachined capacitive sensors [56]. This technology uses fused quartz compo- nents laser-welded together, and the fixed electrode is another diaphragm that is free to deflect but does not experience any pressure (see Figure 6.22). This means it is free to deflect under acceleration and will therefore move in the same manner as 6.5 MEMS Technology Pressure Sensors 137 Figure 6.21 Early silicon/Pyrex capacitive pressure sensor. Pressure Acceleration Figure 6.22 Acceleration compensated quartz capacitive pressure sensor. the pressure-sensitive diaphragm. This technique greatly reduces the cross- sensitivity to accelerations. The main drawbacks associated with the capacitive approach are the inherently nonlinear output of the sensor and the complexity of electronics (compared with the resistive bridge). Assuming parallel deflection in the flexible diaphragm, the change in capacitance is inversely proportional to the gap height. In addition to this, a basic diaphragm such as that shown in Figure 6.21 will bend as it deflects. The diaphragm will therefore no longer be parallel to the fixed electrode and this introduces a fur - ther nonlinearity in the sensor output. The use of bossed diaphragms will mitigate this effect to some degree [57, 58]. Another linearizing approach is to pattern the electrodes such that the sensing capacitance is measured from a particular part of the diaphragm. Maximum deflection occurs at the diaphragm center but this is also the location of maximum nonlinearity. By sensing the capacitance at an annulus removed a short distance from the diaphragm center, non-linearity is reduced but at the expense of sensitivity [59, 60]. Another approach, again at the expense of sensi - tivity, is to clamp the center of the diaphragm such that the pressure-sensitive struc - ture becomes a ring shape. The sensitivity of such a structure is reported to be half that of an equivalent flat plate diaphragm, but nonlinearity falls to 0.7% FS [61]. The final approach commonly employed to improve linearity is to operate the sensor in touch mode, where the diaphragm touching the fixed electrode. The center of the diaphragm is bought into contact by a sufficient pressure, and as pressure increases an increasing area of the diaphragm touches the fixed electrode [62–64]. The output of such a sensor is more linear than that of a typical sensor operated in noncontact mode, as shown by the graph in Figure 6.23. One potential drawback of touch-mode devices is hysteresis arising from friction between the surfaces as they move together and apart, as well as the risk of stiction. The increased circuit complexity associated with capacitive devices and the influ- ence of parasitic capacitances on sensor performance has lead to the development of capacitive interface chips and further research into integrated sensor and circuit solu - tions. Capacitive interface chips have been designed by a number of manufacturers (including Microsensors Capacitive Readout IC MS3110, Analogue Microelectron - ics CAV414, Xemics XE2004, and Smartec’s Universal Transducer Interface chip 138 Pressure Sensors Capacitance Pressure Noncontact region Touch mode region Figure 6.23 Typical capacitance versus pressure relationship for noncontact and touch-mode pressure sensors. (UTI)). However, in order to reduce the effects of parasitic capacitance and achieve higher performance devices, the pressure sensor should ideally be integrated with electronics. This has been achieved by combining a bulk-etched device similar to that shown in Figure 6.21 with basic CMOS circuitry [65, 66], but the more common solution is to employ surface micromachining. Standard sacrificial surface micromachining processes have been combined with CMOS capacitance measure - ment circuitry in a number of devices [67–69]. A common theme with these sensors is the use of an array of sensing diaphragms to increase the measured capacitance sig - nal. In some instances, diaphragms with different pressure sensitivities have been incorporated onto the same die in order to broaden the range of operation [70, 71]. A common application of capacitive pressure sensor arrays with integrated electron - ics is intravascular blood pressure measurement [72] and intracranial pressure [73]. This last device was coated in a silicon elastomer, NUSIL, for reasons of biocompati - bility. A discussion of biocompatible coatings is included in Chapter 4. Similar devices to the surface-micromachined pressure sensors have also been realized using SOI wafers [74]. These devices use the buried oxide as the sacrificial layer, and the hole to allow the undercutting etch is located at the center of the dia - phragm. The hole is sealed afterwards by silicon nitride deposition, which results in a ring shaped diaphragm as described previously. The buried oxide also isolates the diaphragm from the surrounding silicon, thereby reducing parasitic capacitances. A cross-section of the device is shown in Figure 6.24. Another more recent development is the integration of planar coils on the capacitive pressure sensor chip. The capacitor and coil form a resonant LC circuit the frequency of which varies with applied pressure. By integrating the coil on the sensor chip itself, it can also be used to inductively couple power into the sensor chip from an external coil. After energizing the sensor circuit, the external coil is used as an antenna to detect the resonant frequency. This approach is attractive for wireless sensing and can be used in applications where wire links are not suitable (e.g., harsh environments). Several devices have been reported in the literature from different research groups including two integrated devices using electroplated coils [75, 76] and a prototype microsystem on a ceramic substrate with a printed gold coil [64]. 6.5.4 Resonant Pressure Sensors Resonant pressure sensors typically use a resonating mechanical structure as a strain gauge to sense the deflection of the pressure-sensitive diaphragm. Resonant sensing has been discussed in Chapter 5. The resonant approach is more technically chal - lenging for a number of reasons discussed below, but it does offer performance specifications beyond that achievable with piezoresistive and capacitive techniques. 6.5 MEMS Technology Pressure Sensors 139 Diaphragm Nitride seal Metal contact Sacrifical oxide Figure 6.24 Cross-section through SOI capacitive pressure sensor. [...]... Solid-State Sensors and Actuators, Digest of Technical Papers, Vol 1, 2001, pp 482 – 485 6.7 Conclusions 147 [37] Kroetz, G H., M H Eickhoff, and H Moeller, “Silicon Compatible Materials for Harsh Environment Sensors, ” Sensors and Actuators, Vol A74, 1999, pp 182 – 189 [ 38] Okojie, R S., A A Ned, and A D Kurtz, “Operation of a (6H)-SiC Pressure Sensor at 500C,” Sensors and Actuators, Vol A66, 19 98, pp... Corrugated Silicon Diaphragms,” Sensors and Actuators, Vol A23, 1990, pp 988 –992 [13] Timoshenko, S P., and S Woinowski-Krieger, Theory of Plates and Shells, New York: McGraw-Hill, 1 983 146 Pressure Sensors [14] Stankevic, V., and C Šimkevicius, “Use of a Shock Tube in Investigations of Micromachined Piezoresistive Pressure Sensors, ” Sensors and Actuators, Vol 86 , 2000, pp 58 65 [15] Entran EPIH series... Electron Devices, ED-29, 1 982 , pp 64–70 [ 18] Clark, S K., and K D Wise, “Pressure Sensitivity in Anisotropically Etched ThinDiaphragm Pressure Sensors, ” IEEE Trans Electron Devices, ED-26, 1979, pp 188 7– 189 6 [19] Elgamel, H E., “Closed-Form Expressions for the Relationships Between Stress, Diaphragm Deflection, and Resistance Change with Pressure in Silicon Piezoresistive Pressure Sensors, ” Sensors and Actuators,... Sanchez, J C., “Semi-Conductor Strain Gauge Pressure Sensors, ” Instruments and Control Systems, November 1963, pp 117–120 [8] Peake, E R., A R Zias, and J V Egan, “Solid-State Digital Pressure Devices,” IEEE Trans on Electron Devices, ED-16, No 19, October 1969, pp 87 0 87 6 [9] Mallon, J R., et al., “Low-Pressure Sensors Employing Bossed Diaphragms and Precision Etch-Stopping,” Sensors and Actuators,... Single-Crystalline Piezoresistors,” Sensors and Actuators, Vol A54, 1996, pp 488 –492 [42] Lin, L., and W Yun, “Design, Optimization and Fabrication of Surface Micromachined Pressure Sensors, ” Mechatronics, Vol 8, 19 98, pp 505–519 [43] Guckel, H., “Silicon Microsensors: Construction, Design and Performance,” Microlectronic Engineering, Vol 15, 1991, pp 387 –3 98 [44] Sugiyama, S., et al., “Micro-Diaphragm... Sensor,” Sensors and Actuators, Vol A37– 38, 1993, pp 82 85 [80 ] Harada, K., et al., “Various Applications of Resonant Pressure Sensor Chip Based on 3-D Micromachining,” Sensors and Actuators, Vol A73, 1999, pp 261–266 [81 ] Ikeda, K., et al., “Silicon Pressure Sensor Integrates Resonant Strain Gauge on Diaphragm,” Sensors and Actuators, Vol A21–23, 1990, pp 146–150 [82 ] Ikeda, K., et al., “Three-Dimensional... New Age for Silicon Sensors and Actuators,” Sensors and Actuators, Vol A46, 1995, pp 8 16 [35] Diem, B., et al., “‘SIMOX’: A Technology for High-Temperature Silicon Sensors, ” Sensors and Actuators, Vol A23, 1990, pp 1003–1006 [36] von-Berg, J., et al., “A Piezoresistive Low-Pressure Sensor Fabricated Using Silicon-onInsulator (SOI) for Harsh Environments,” TRANSDUCERS ’01, EUROSENSORS XV, 11th International... (Transducers 85 ), June 11–14, 1 985 , Philadelphia, PA, pp 186 – 188 [55] Catling, D C., “High-Sensitivity Silicon Capacitive Sensors for Measuring MediumVacuum Gas Pressures,” Sensors and Actuators, Vol A64, 19 98, pp 157–164 [56] Mulkins, D F., and Pogany, K., “Quartz Capacitive Pressure Sensor: Product Applications and Technical Description,” Sensors Expo West Proceedings, 1 989 , pp 206A/1–206A/12 [57] Zhang,... resonator possesses a Q-factor of more than 50,000 [82 ] Other similar devices have been fabricated using a variety of techniques including silicon fusion bonding [83 ], surface-micromachined resonators on bulk etches diaphragms [84 ], and more recently using SOI wafer technology [85 ] and entirely surfaced-micromachined sensors [86 ] Surface micromachining offers the opportunity for using comb-drive structures... T., and H.-S., Lee, “An Integrated Air-Gap-Capacitor Pressure Sensor and Digital Readout with Sub-100 Attofarad Resolution,” IEEE J MEMS, Vol 1, No 3, September 1992, pp 121–129 [67] Scheiter, T., et al., “Full Integration of a Pressure Sensor into a Standard BiCMOS Process,” Sensors and Actuators, Vol A67, 19 98, pp 211–214 [ 68] Paschen, U., et al., “A Novel Tactile Sensor System for Heavy-Load Applications . Digest 1 985 Int. Conf. on Solid State Sensors and Actuators (Transducers 85 ), June 11–14, 1 985 , Philadelphia, PA, pp. 186 – 188 . [55] Catling, D. C., “High-Sensitivity Silicon Capacitive Sensors. Micromachined Reso - nant Pressure Sensor,” Proc. 8th Int. Conf. on Solid State Sensors and Actuators (Transduc - ers ’95) and Eurosensors IX, Stockholm, Sweden, June 25–29, 1995, pp. 586 – 589 . [85 ] Beeby. Silicon Reso - nant Sensor,” Sensors and Actuators, Vol. A37– 38, 1993, pp. 82 85 . [80 ] Harada, K., et al., “Various Applications of Resonant Pressure Sensor Chip Based on 3-D Micromachining,” Sensors