Microsensors, MEMS and Smart Devices - Gardner Varadhan and Awadelkarim Part 16 doc

SMART SENSORS 433 Device antenna Reflectors IDT System antenna Figure 15.14 (a) Schematic representation of a wireless passive IDT-SAW sensor and (b) discrimination of different liquids through acoustic attenuation and phase parameters. The measurements were made at a peak frequency of 108.7 MHz. From Varadan and Gardner (1999) IDT sensor. 4 The device may be regarded as an 'electronic tongue' because the sweetness (sugar level) and saltiness (ion concentration) will affect the electrical conductivity and dielectric permittivity of the liquid, whereas the mouth feel and, possibly, the freshness will relate to the viscoelasticity of the liquid. The use of a second Love wave can help to improve the discrimination process by removing (screening out) the effects of electrical coupling from the measurement system. The advantage of this type of liquid sensor is that it does not require a biological coating but relies upon a physical principle. This greatly enhances the lifetime of the sensor, although its specificity is clearly reduced. However, we think that the low cost and rapid response makes the development of a passive wireless smart tongue extremely attractive. 4 For details of the principles of IDT-SAW microsensors, see Chapter 10 434 SMART SENSORS AND MEMS 15.3 MEMS DEVICES Silicon micromachining techniques are used to fabricate various micromechanical struc- tures and many of these form micromechanical actuators. Throughout this book, we have described a large number of different types of microactuators and some of these are, perhaps, rather loosely referred to as MEMS devices. Table 15.5 lists some of the different types of microactuators and MEMS described in the worked examples (WE) in this book. A more comprehensive list of micromechanical structures is provided in the following Table (15.6) and has been taken from Frank (1996). These include microvalves, microp- umps, microgears, and so on. MEMS devices can be considered as smart devices because they integrate sensors with actuators (i.e. a smart microsystem), but the degree of integration can vary significantly. There are many different types of MEMS devices being made today, but two of the most exciting ones are used in optical and chemical instrumentation. The former is sometimes referred to as microoptoelectromechanical system (MOEMS), and this is being driven by the optical telecommunications and biotechnology industries. One example shown in Figure 15.15 is a picture of an adaptive mirror IC. This array device permits the electronic correction of an optical system in an adaptive manner. Clearly, the ability to Table 15.5 Some examples of microactuators and MEMS Type Microactuator Microactuator Microactuator MEMS MEMS MEMS Description Linear motion actuator Rotor on a centre-pin bearing Rotor on a flange bearing Centre-pin bearing side-drive micromotor Gap comb-drive resonant actuator Overhanging microgripper Reference WE 6.2 WE 6.3 WE 6.4 WE 6.6 WE 6.7 WE 6.10 Table 15.6 Various micromechanical and MEMS devices made from silicon. (This table is based on Frank (1996)) Cryogenic microconnectors Fibre-optic couplers Film stress measurement Fluidic components IC heat-sinks Ink-jet nozzles Laser beam deflectors Laser resonators Light modulators Membranes Microaccelerometers Microairplanes Microaligners Microbalances Microfuses Microgears Micromoulds Micromotors Micropositioners Microinterconnects Microchannels Microrobots Micromachines Micromanipulators Micromechanical memory Microgyrometers Microchromatographs Microinterferometers Microspectrometers Micro-SEM Microprobes Micropumps Microswitches Microvacuum tubes Nerve regenerators Photolithographic masks Pressure switches Pressure regulators RMS converters Thermal printheads Thermopiles Torsion mirrors Vibrating microstructures MEMS DEVICES 435 Figure 15.15 Smart MOEMS: an adaptive mirror IC for electronic correction of optical system (Courtesy of Delft University) modify electronically the characteristics of a mirror offers the opportunity of automated focusing, beam direction and so forth. A second example of a smart optical MEMS chip is a visible light microspectrom- eter based on a monolithic dielectric slab waveguide with monolithic integrated focusing echelle grating produced by a micromoulding process, as in Figure 15.16. The spectrum is generated by coupling the light through a silica fibre into a three-layer waveguide. The light is split by a self focussing reflection grating (d = 0.2 urn, g = 2 um) with 625 lines per mm, and the different wavelengths are read out by a 256-element photodiode array. The device is small and has a wavelength reproducibility better than 0.1 nm. Measuring wavelength content is important in many sensing applications, from security labeling through to biological assays. In the chemical analytical instrumentation area, Stanford reported the first silicon micro- machined gas chromatography system as early as 1974. Figure 15.17(a) shows the basic layout of a gas chromatography system (Madou 1997) in which a gaseous sample is made to travel along a long capillary column. The various components in the sample have different retention times within the column and so appear at the far end as indi- vidual compounds. The chemical components are detected using, for example, a flame ionisation detector and the peak is identified according to known retention times. Both the capillary column and detector cavity has been mieromachined. Figure 15.17(b) shows a minigas system, manufactured by MTI, which is a chromatograph that is essentially portable and so suitable for remote fieldwork. 436 SMART SENSORS AND MEMS Figure 15.16 Smart MOEMS: (a) principle and (b) photograph of a microspectrometer 1C. (Cour- tesy of MicroParts, Germany) Efforts have also been made to make another miniature analytical chemical instru- ment, namely, the mass spectrometer. This instrument ionises the molecules after they have passed through an orifice and then separates the masses out in a mass filter by deflecting their motion with a quadrupole electrostatic or any other type of lens. Finally, the abundance of mass ions are measured using an ion detector (Figure 15.18(a) from Friedhoff et al. (1999)). Figure 15.18(b) shows a plot of the mass content of the headspace above a bacteria hazardous to human health, Escherichia coli, in two of its growth phases (Esteves de Matos et al. 2000). The measurements were taken with a conventional Agilent Technologies quadrupole mass spectrometer 4440. The difference between these mass spectrograms shows up when a pattern analysis technique, such as principal components MEMS DEVICES 437 Micromachined components Interlocked spiral-shaped capillary column Miniaturized components (a) Figure 15.17 Microgas analysis: (a) basic arrangement of a gas chromatograph (Madou 1997) and (b) a portable gas chromatograph manufactured by MTI (Fremont, CA) analysis or a neural network, is applied to the data. Thus, the growth phase of the bacteria can be identified and hence its response to antibiotics can be predicted. In fact, the Agilent 4440 system may be regarded as an 'intelligent instrument' because it couples the mass spectrometer together with sophisticated PARC software. Conventional quadrupole mass spectrometers (e.g. Agilent Technologies Inc.) sell for 20 000 euros, are heavy (>50 kg), consume considerable power (> 1 kW), and so there is a 438 SMART SENSORS AND MEMS Figure 15.18 (a) Schematic layout of a mass spectrometer. The ions need to travel through a vacuum. Redrawn from Friedhoff et al. (1999) (b) Plot of the abundance of ions of differing mass in the headspace of a bacteria, E. coli, grown in a nutrient aqueous solution. The top plot shows the spectrum in the first growth phase of the bacteria (lag) and the second plot (inverted for the sake of clarity) shows the spectrum in the later stationary phase. From Esteves de Matos et al. (2000) considerable demand for smaller, lighter, and portable units. Miniature mass spectrometers can be made using conventional precision machining techniques, but there have been recent efforts to make one using silicon micromachining techniques. Figure 15.19(a) shows an arrangement of a quadrupole lens employing silicon micromachined parts (Friedhoff et al. 1999). An optical microscopic view of the cross section is shown in Figure 15.19(b) (Syms et al. 1996). Making a complete silicon version is not straightforward because the device requires a good uniform field and vacuum. However, a unit that analyses a small mass range (up to 40 daltons) has been demonstrated successfully by Friedhoff et al. (1999) (Figure 15.19(c)), and further advances are expected shortly on ion sources and detectors. Biotechnology companies, such as Sequenome (USA), are developing MassArray™ chips to screen genetic defects called single nucleotide MEMS DEVICES 439 Figure 15.19 (a) Schematic of a micromachined quadrupole lens assembly for a MEMS mass spectrometer; (b) optical photograph of a cross section of the assembly quadrupole lens (Syms et al. 1996); (c) mass spectrum of a mixture of helium, argon, and air obtained with a 3 cm lens with 500 ^im diameter electrodes driven at 6 MHz. (Friedhoff et al. 1999) 440 SMART SENSORS AND MEMS polymorphisms (SNPs). These require a mass spectrometer capable of detecting masses in the range of 5000 to 9000 daltons, and work is underway in DARPA (USA). The advantages of carrying out chemical analysis and (bio)chemical reactions are primarily • Small, portable, low-power units • Small amounts of chemical reagents are required (microlitres or less) • Rapid screening of multiple experiments (assay microarray chips) The last reason is, probably, the most compelling, namely, that it is possible to perform hundreds, thousands, and, possibly, millions of (bio)chemical reactions on a single wafer, and thus the cost can be low for a large number of trials. This is essential in the rapidly emerging fields of genomics, proteomics, and pharmacogenomics. For example, Figure 15.20(a) shows the basic arrangement of a DNA probe. Here, light is shone on the elements and the observation of the activation of fluorescent markers permits the identification of the presence of certain genes. Figure 15.20(b) shows the biochip made using a 0.8 (im CMOS process with 128 DNA probes. The chip can be used in clinical diagnostics, for example, in screening for cancer. There are now a plethora of emerging technologies and products in the Bio-MEMS area, and most of these do not involve silicon but use glass and polymers. These must tackle the problems of sample preparation, sample movement, and readout when potential is required to detect hundreds of thousands of genetic defects or SNPs. For example, disposable microtitration plates with a capillary fill of 96 reaction wells are now being manufactured by MicroParts in Germany (Figure 15.21), and when coupled with off- chip optical readout, they can be used to screen drugs (pharmacogenomics). There is also a rapid development in the area of a Lab-on-a-chip - a microfluidic chip to perform microbiology - that is to grow bacteria in microcells, challenge antibiotics, and analyse on-chip and various other Bio-MEMS chips such as Nanogen™, which is an APEX chip system, Spectrochip™, which is a microarray assay chip, and e-Sensor Systems (Motorola), which is a clinical microsensor. Such diagnostic Bio-MEMS chips permit rapid bacterial identification and antibody susceptibility. In a related method, the use of combinatorial analysis makes it possible to try out, liter- ally, millions of different chemical reactions on a single chip, and so help in synthesizing new compounds in a type of rapid prototyping. Therefore, the rapid screening of biochemical reactions, identification of genetic defects, and testing of new drugs with a biochip are exciting prospects for smart MEMS devices. Such developments could revolutionise the field of medicine because it may be possible to individualise drug therapy; in other words, the rapid screening of people in clinics (and, perhaps, eventually at home) will result in the tailoring of both the choice of drug and its quantity delivered to them. It is essential when you consider the natural diversity in the response of individuals to both pathogenic agents and drug therapies. A medical advance like this should improve the targeting of drugs and, thereby, reduce the time for patient recovery. In addition, there are other medical applications of MEMS in surgery to enhance manipulators and catheters for better minimum access intervention (i.e. keyhole surgery) in smart drug delivery systems to control dosage, and in human implants to replace and augment body parts (e.g. cochlea, heart valves, pacemakers). MEMS DEVICES 441 Figure 15.20 Rapid screening of biological material: (a) optical principle, and (b) CMOS array biochip for DNA analysis. From Caillat et al. (1999) 442 SMART SENSORS AND MEMS Figure 15.21 Microtiterplate for clinical microbiological applications: Lilliput chip from MicroParts (Germany) 15.4 CONCLUDING REMARKS This chapter has described some of the recent advances in silicon micromachining tech- niques that seek to integrate the processor electronics with the sensor and actuators. These so-called smart sensors and MEMS are becoming increasingly sophisticated as the power of the processing unit increases from that provided by a simple eight-bit micro- controller (e.g. a 6805) through CISC technology (68 000-based) to RISC and digital signal-processing (DSP) technologies. It is now possible to perform on-chip a lot of intel- ligent features such as self-testing, fault-diagnostics, and adaptive control. Furthermore, there has been an increased use of artificial intelligence on sensors with the incorporation of artificial neural networks, expert systems, and fuzzy logic. Fuzzy controller ICs are already used today, for example, in buildings to operate lifts - we are just not aware of them! Looking to the future, it seems that there are two areas that will develop further. The first is that we will be able to make increasingly sophisticated micromachines that parallel the human senses, such as the microgrippers, micronoses, microtongues etc. presented in this book. The second area is the way in which we communicate with these microdevices and especially micromachines, such as microrobots, microcars, and microplanes (Fujimasa 1996). The human-machine interface will probably become a limiting step and there will be a need to communicate remotely, perhaps via speech, with these intelligent micromachines. Perhaps, we will even see the day in which we implant these microdevices in our own body to augment our own limited senses, that is, we could have infrared (or night) vision, detect poisonous and toxic biological agents, and feel magnetic or electric fields. We could also have speech translators implanted in our ears, and so understand foreign languages, and, perhaps, attach devices that monitor our health and automatically warn us of imminent problems. It [...]... z a f P n m c d da h k M G Yacto- (10–24) Zepto- (10 –21 ) Atto- (10–18) Femto- (10–15) Pico- (10–12) Nano- (109) Micro- (10–6) Milli- (10 –3 ) Centi- (10 –2 ) Deca- (10 – 1 ) Deca- (10+1) Hecto- (10+2) Kilo- (10+3) Mega- (10+6) Giga-(10+9) APPENDIX B: LIST OF SYMBOLS AND PREFIXES Symbol Meaning T P E Z Y Tera- (10+12) Peta- (10+15) Exa- (10+l8) Zetta- (10+21) Yotta- (10+24) 453 This page intentionally... semiconductor devices, International patent WO98/32009, July 1998 Udrea, F et al (2001) "Design and simulations of a new class of SOI CMOS micro-hotplate gas sensors," Sensors and Actuators B, 78, 180–190 van der Horn, G and Huijsng, J H (1998) Integrated Smart Sensors: Design and Calibration, Kluwer Academic Publishers, Dordrecht Varadan, V K and Gardner, J W (1999) Smart tongues and smart noses, Proc... Actuators B, 27, 261–266 Gardner, J W and Bartlett, P N (1994) "A brief history of electronic noses," Sensors and Actuators B, 1 8-1 9, 211–220 Gardner, J W and Bartlett, P N (1999) Electronic Noses: Principles and Application, Oxford University Press, Oxford, UK, p 245 Gardner, J W and Hines, E L (1996) Pattern analysis techniques, in E Kress-Rogers ed Handbook of Biosensors: Medicine, Food and the Environment,... quality of life and is not misused REFERENCES Baltes, H and Brand, O (2000) CMOS-based microsensors, Proceedings of Eurosensors XIV, Copenhagen, Denmark, August 27–30th, ISBN 87–8993 5-5 0-0 Barney, G C (1985) Intelligent Instrumentation, Prentice-Hall, Englewood Cliffs, New Jersey, p 532 Bissell, C C (1994) Control Engineering, Chapman & Hall, London, p 266 Breckenbridge, R A and Husson, C (1978) Smart sensors... n-region Refractive index Donor concentration in n-type semiconductor material Acceptor concentration in n-type semiconductor material Refractive index Acceptor and donor densities Indicates type of silicon doping-electron acceptors Highly-doped p-region Donor concentration in p-type semiconductor material Acceptor concentration in p-type semiconductor material Pressure APPENDIX B: LIST OF SYMBOLS AND. .. (Advisor J W Gardner) , University of Warwick, Coventry, UK Syms, R R A., Tate, T J., Ahmad, M M and Taylor, S (1996) "Fabrication of a microengineered quadrupole electrostatic lens," Electronic Lett., 32, 209 4-2 095 444 SMART SENSORS AND MEMS Tsuchiya, K and Davies, S T (1998) "Fabrication of TiNi shape memory alloy microactuators by ion beam sputter deposition," Nanotechnology, 9, 6 7-7 1 Udrea, F and Gardner, ... functions or parts For instance, a simple microprocessor can be used as a microcontroller Measurand MEMS MEMS device Microactuator Microdevice Microcontroller 456 APPENDIX C: LIST OF SOME IMPORTANT TERMS Term Microelectronics Microprocessor Microsensor Microtechnology Multifunctional material Nanotechnology Processor Sensor Smart actuator Smart electronics Smart material Smart sensor Smart structure... (micron level) electronic parts A miniature processor that has been fabricated using microtechnology, e.g an 8-bit, 1 6- bit or 32-bit processing chip A sensor that has at least one physical dimension at the submillimetre level The science and history of the mechanical and industrial arts used to make extremely small (micron level) parts More commonly thought of as the methodologies and processes required... sensing using non-optical microelectromechanical systems," J Vac Sci Technol., 17, 230 0-2 307 Fujimasa, I (1996) Micromachines: A New Era in Mechanical Engineering, Oxford University Press, Oxford, UK, p 156 Gandhi, M V and Thompson, B S (1992) Smart Materials and Structures, Chapman Hall, London, p 309 Gardner, J W (1995) "Intelligent gas sensing using an integrated sensor pair," Sensors and Actuators... spacecraft: the impact and trends, Proc of the AIAA/NASA Conf on Smart Sensors, Hampton, USA, pp 1-5 Brignell, J E and White, N (1994) Intelligent Sensor Systems, IOP Publishing, Bristol, UK, p 256 Caillat, P et al (1999) Biochips on CMOS: an active matrix address array for DNA analysis Chapman, P W (1996) Smart sensors, Research Triangle Park, NC, 162 p Culshaw, B (1996) Smart Structures and Materials, Artech . temperature Yacto- (10 –24 ) Zepto- (10 –21 ) Atto- (10 –18 ) Femto- (10 –15 ) Pico- (10 –12 ) Nano- (10 9 ) Micro- (10 –6 ) Milli- (10 –3 ) Centi- (10 –2 ) Deca- (10 –1 ) Deca- (10+ 1 ) Hecto- (10 +2 ) Kilo- . a Lab-on-a-chip - a microfluidic chip to perform microbiology - that is to grow bacteria in microcells, challenge antibiotics, and analyse on-chip and various other Bio -MEMS chips. and MEMS Type Microactuator Microactuator Microactuator MEMS MEMS MEMS Description Linear motion actuator Rotor on a centre-pin bearing Rotor on a flange bearing Centre-pin bearing side-drive