Lectures on MEMS and MICROSYSTEMS DESIGN and MANUFACTURE

Chapter 2 Working Principles of MEMS and Microsystems

Hsu 2008

This Chapter will present the working principles of various micro sensors and actuators in microsystems. Minute sensors are expected to detect a variety of signals associated with: Accelerations (velocity and forces), Biological and biomedical Chemical, Forces (e.g., microaccelerometers and gyroscopes) Optical, Pressure, Thermal (temperatures), etc. Input samples may be: motion of a solid, pressurized liquids or gases, biological and chemical substances. Due to the minute sizes, microactuators work on radically different principles than the conventional electromagnetic means, such as solenoids and ac/dc motors. Instead, electrostatic, thermal, piezoelectric and shape-memory alloys are extensively used in microactuations.

Working Principles for Microsensors

Power Supply

Input Signal

Micro Sensing Element

Transduction Unit

Output Signal

Acoustic Wave Sensors

Acoustic wave sensor does not related to the sensing of acoustic waves transmitted in solids or other media, as the name implies. Primary application of these sensors is to act like band filters in mobile telephones and base stations. Other applications include: Sensing of torques and tire pressures Sensing biological and chemical substances Sensing vapors, humidity and temperature Monitor fluid flow in microfluidics 2 sets of Interdigital Transducers (IDT) are created on a piezoelectric layer attached to a tiny substrate as shown Energize by an AC source to the Input IDT will close and open the gaps of the finger electrodes, and thus surface deformation/ stresses transmitting through the piezoelectric material The surface deformation/stresses will cause the change of finger electrodes in the Output IDT Any change of material properties (chemical attacks) or geometry due to torques will alter the I/O between the Input IDT and Output IDT. The sensing of contact environment or pressure can thus be accomplished

BioMEMS
The term BioMEMS has been a popular terminology in the MEMS industry in recent years due to the many break-through in this technology, which many believe to be a viable lead to mitigate the sky-rocketing costs in healthcare costs in many industrialized countries. BioMEMS include the following three major areas: (1) Biosensors for identification and measurement of biological substances, (2) Bioinstruments and surgical tools, and (3) Bioanalytical systems for testing and diagnoses.

Major Technical Issues in BioMEMS Products:

(1) Functionality for the intended biomedical operations. (2) Adaptive to existing instruments and equipment. (3) Compatibility with biological systems of the patients. (4) Controllability, mobility, and easy navigation for operations such as those required in laparoscope's surgery. (5) Functions of MEMS structures with high aspect ratio (defined as the ratio of the dimensions in the depth of the structure to the dimensions of the surface) Note: Almost all bioMEMS products are subjected to the approval for marketing by the FDA (Food and Drug Administration) of the US government.

Biomedical Sensors and Biosensors

These sensors are extensively used in medical diagnosis, environmental protection, drug discovery and delivery, etc.

Biomedcial Sensors
For the measurements of biological substances in the sample and also for medical diagnosis purposes. Input signal: Biological sample (e.g., blood samples or body fluids typically in minute amount in L or nL) Microsensing element: a chemical that reacts with the sample. Transduction unit: the product of whatever the chemical reactions between the sample and the chemical in the sensing element will convert itself into electrical signal (e.g. in milli volts, mV). Output signal: The converted electrical signal usually in mV.

Example of a biomedical sensor: A sensor for measuring the glucose concentration of a patient.
Pt electrode H+ H+ Polyvinyl alcohol solution H+ H+ H+

Blood sample

i
Working principle:

Ag/AgCl Reference electrode

The glucose in patients blood sample reacts with the O2 in the polyvinyl alcohol solution and produces H2O2. The H2 in H2O2 migrates toward Pt film in a electrolysis process, and builds up layers at that electrode. The difference of potential between the two electrodes due to the build-up of H2 in the Pt electrode relates to the amount of glucose in the blood sample.

Biosensors
These sensors work on the principle of interactions between the biomolecules in the sample and the analyte (usually in solution) in the sensor. Signal transduction is carried out by the sensing element as shown below:
ANALYTE

Biomolecule Layer

B B

Biomolecule Supply

Output Signals

B B B

Sensor

Chemical Optical Thermal Resonant Electrochemical ISFET (Ion Sensitive Field Effect Transducer)

Chemical Sensors
Work on simple principles of chemical reactions between the sample, e.g. ,O2 and the sensing materials, e.g., a metal. Signal transduction is the changing of the physical properties of the sensing materials after specific type of chemical reactions. There are four (4) common types of chemical sensors: (1) Chemiresistor sensors. (2) Chemicapacitor sensors.
Chemically Sensitive Polyimide Input current or voltage Input Voltage Metal Insert Metal Electrodes Output: Change of Resistance Output: Capacitance Change

Measurand Gas

Chemical Sensors-Contd

(3) Chemimechanical sensors: Work on certain materials (e.g. polymers) that change shapes when they are exposed to chemicals. Measuring the change of the shape of the sensing materials determines the presence of the chemical. (4) Metal oxide gas sensors: Sensing materials: certain semiconducting materials, e.g., SnO2 change their electrical resistance when exposed to certain chemicals.
Measurand Gas

Electric Contact

SnO2 SiO2

Silicon Substrate

Chemical Sensors-Contd
Available metal oxide gas sensors:
Semiconducting Metals BaTiO3/CuO SnO2 SnO2 SnO2 SnO2 ZnO WO3 Fe2O3 Ga2O3 MoO3 In2O3 Catalyst Additives La2O3, CaCO3 Pt + Sb Pt Sb2O3 CuO V, Mo Pt Ti-doped + Au Au None None CO2 CO Alcohols H2, O2, H2S H2S Halogenated hydrocarbons NH3 CO CO NO2, CO O3 Gas to be Detected

Optical Sensors
These sensors are used to detect the intensity of lights. It works on the principle of energy conversion between the photons in the incident light beams and the electrons in the sensing materials. The following four (4) types of optical sensors are available:
Photon Energy

Semiconductor A
is more transparent to photon energy in incident light
R
Semiconductor A Junction Semiconductor B (a) Photovoltaic junction

Photon Energy

R
(b) Photoconductive device

Bias Voltage Reverse _ Bias + Voltage

Photon Energy

Photon Energy R
p-Material

p n
Leads

Vout

n-Material

(c) Photodiodes

Optical Sensors-Contd
Photon Energy

Collector

p n p

Photon Energy

Emitter

Collector

p n p

Emitter

Base
(d) Phototransistors

Base

Silicon (Si) and Gallium arsenide (GaAs) are common sensing materials. GaAs has higher electron mobility than Si- thus higher quantum efficiency. Other materials, e.g. Lithium (Li), Sodium (Na), Potassium (K) and Rubidium (Rb) are used for this purpose.

Pressure Sensors Micro pressure sensors are used to monitor and measure minute gas pressure in environments or engineering systems, e.g. automobile intake pressure to the engine. They are among the first MEMS devices ever developed and produced for real world applications. Micro pressure sensors work on the principle of mechanical bending of thin silicon diaphragm by the contact air or gas pressure.
Measurand Fluid Inlet Silicon Die with Diaphragm Cavity Constraint Base Measurand Fluid Inlet (a) Back side pressurized (b) Front side pressurized Cavity

Pressure Sensors-Contd The strains associated with the deformation of the diaphragm are measured by tiny piezoresistors placed in strategic locations on the diaphragm. These tiny piezoresistors are made R , R , R , R = Piezoresistors from doped silicon. They work on the similar principle as foil strain R Metal Pad Metal Pad gages with much smaller sizes (in R R R m), but have much higher sensitivities and resolutions.
1 2 3 4 4 3 1 2

Top view of silicon die

Wire bond Metal film

Piezoresistors Dielectric layer

+
Vin

R3 (+ve) a Vo R2(-ve) b

R1(+ve)

Silicone gel
Silicon Diaphragm Pyrex Glass Constraining Base or Metal Header

R4(-ve)

Die Attach

Metal Casing

Wheatstone bridge for signal transduction

Interconnect

Passage for Pressurized Medium

R1 R3 Vo = Vin R +R R2 + R3 4 1
R1,R3 = resistance induced by longitudinal and transverse stresses R2,R4 = reference resistors

Pressure Sensors-Contd Other ways of transducing the deformation of the diaphragm to electronic output signals are available, e.g.,
Metallic Electrode Silicon Cover Metallic Electrode Silicon Die Cavity Constraint Base Measurand Fluid Inlet

Signal output: capacitance changes

(for higher temperature applications)

C = r o

A = Overlap area D = Gap between plate electrodes

r = Relative permittivity = 1.0 with air o = Permittivity in vacuum = 8.85 pF/m

A d

By resonant vibration
(for higher resolutions) Signal output: Shift of resonance frequencies by change of stresses in lower plate electrode by applied pressure loading

Diffused p-type electrode Silicon diaphragm

1200 m sq.x 100 m thick

x 600 m long x 6 m thick)

Vibrating beam: (n-type Si wafer,40 m wide

Silicon die

(400 m thick)

Pressurized medium

Constraint base

Two Common Types of Micro Pressure Sensors Sensors using piezoresistors: Small in size Linear I/O relation Temperature sensitive

Sensors using capacitances: Tends to be bulky Suited for elevated temperature application Nolinear I/O relations Lower cost Nonlinear I/O with plate pressure sensors using electrodes
Electric circuit bridge for converting capacitance changes to voltage output:

Vo =
Change of Capacitance, pF

C Vin 2(2C + C )

C Vo C

Variable capacitor Vin

14 12 10 8 6 4 2 0 0 0.5 1 1.5 2 2.5 Gap, micrometer

Pressure Sensors-Contd

Major problems in pressure sensors are in the system packaging and protection of the diaphragm from the contacting pressurized media, which are often corrosive, erosive, and at high temperatures.

Thermal Sensors
Thermal sensors are used to monitor, or measure temperature in an environment or of an engineering systems. Common thermal sensors involve thermocouples and thermopiles. Thermal sensors work on the principle of the electromotive forces (emf) generated by heating the junction made by dissimilar materials (beads):
Metal Wire A Heat Metal Wire A Cold Junction

Heat
Hot Junction

i i
Bead

i
V

V Voltage Output

i
Metal Wire B

Metal Wire B

Voltage Output

(a) A thermocouple

(b) A dual junction thermocouple

The generated voltage (V) by a temperature rise at the bead (T) is:

V = T
where = Seebeck coefficient

Thermal Sensors-Contd
The Seebeck coefficients for various thermocouples are:
Type Wire Materials Seebeck Coefficient (V/oC) 58.70 at 0oC 50.37 at 0oC 39.48 at 0oC 10.19 at 600oC 38.74 at 0oC 11.35 at 600oC Range (oC) Range (mV)

E J K R T S

Chromel/Constantan Iron/Constantan Chromel/Alumel Platinum (10%)-Rh/Pt Copper/Constantan Pt (13%)-Rh/Pt

-270 to 1000 -210 to 1200 -270 to 1372 -50 to 1768 -270 to 400 -50 to 1768

-9.84 to 76.36 -8.10 to 69.54 -6.55 to 54.87 -0.24 to 18.70 -6.26 to 20.87 -0.23 to 21.11

Common thermocouples are of K and T types

Thermal Sensors-Contd
Thermopiles are made of connecting a series of thermocouples in parallel:
Thermocouples Hot Junction Region, Th

Cold Junction Region, Tc V

The induced voltage (V) by the temperature change at the hot junction (T) is:

V = N T
with N = number of thermocouple pairs in the thermopile.

Thermal Sensors-Contd
A micro thermal sensor:
3.6 mm 32 Thermocouples 16 m wide Cold Junction Region

32 polysilicon-gold thermocouples dimension of thermopile is: 3.6 mm x 3.6 mm x 20 m thick Typical output is 100 mV

3.6 mm

Hot Junction Region

Diaphragm: 1.6 mm dia x 1.3 m thick

Response time is 50 ms

Top view

Hot Junction Region Thermocouples

20 m

Diaphragm

Silicon Rim Support

Elevation

Working Principles for Microactuators

Power Supply

Output Action

Micro Actuating Element

Transduction Unit

Power supply:

Electrical current or voltage

Transduction unit: To covert the appropriate form of power supply into the desired form of actions of the actuating element Actuating element: A material or component that moves with power supply Output action: Usually in a prescribed motion

Actuation Using Thermal Forces

Solids deform when they are subjected to a temperature change (T) A solid rod with a length L will extend its length by L = T, in which
= coefficient of thermal expansion (CTE) a material property.

When two materials with distinct CTE bond together and is subjected to a
temperature change, the compound material will change its geometry as illustrated below with a compound beam:
Heat

1 > 2 1 2

These compound beams are commonly used

in MEMS products.

as microswitches and relays

Actuation Using Shape Memory Alloys (SMA)

SMA are the materials that have a memory of their original geometry (shape) at a typically elevated temperature of production. These alloys are deformed into different geometry at typically room temperature. The deformed SMA structures will return to their original shapes when they are heated to the elevated temperature at their productions. Ti-Ni is a common SMA. A microswitch actuated with SMA:
Shape Memory Alloy Strip e.g. TiNi or Nitinolor Resistance Heating Strip

Silicon Cantilever Beam Constraint Base

Actuation Using Piezoelectric Crystals A certain crystals, e.g., quartz exhibit an interesting behavior when subjected to a mechanical deformation or an electric voltage. This behavior may be illustrated as follows:
Mechanical Forces
Induced Mechanical Deformation Applied Voltage, V

Mechanical force induced electric voltage

Electric voltage induced mechanical deformation

This peculiar behavior makes piezoelectric crystals an ideal candidate for microactuation as illustrated in the following case:

Actuation Using Piezoelectric Crystals-Contd

A micro relay or microelectrical switch

Electrodes

Piezoelectric Silicon Cantilever Beam Constraint Base

Actuation Using Electrostatic Forces

Electrostatic Force between Two Particles The Coulombs Law:

(with charge q)

,r ce an is t ion F D t ac F tr At ion ls pu Re

A
(with charge q)

The attraction or repulsive force:

1 qq ' F = 4 r 2

where = permittivity of the medium between the two particles = 8.85 x 10-12 C2/N-m2 or 8.85 pF/m in vacuum (= o) r = Distance between the particles (m)

Actuation Using Electrostatic Forces-Contd

Electrostatic Force Normal to Two Electrically Charged Plates:

Length, L

V
Gap, d
dt Wi h, W

The induced capacitance, C is: The induced normal force, Fd is:

C = r o

A WL = r o d d

1 r o WL 2 Fd = V 2 2 d
in which r = relative permittivity of the dielectric material between the two plates (see Table 2.2 for values of r for common dielectric materials).

Actuation Using Electrostatic Forces-Contd

Electrostatic Force Parallel to Two Misaligned Electrically Charged Plates:

Fd L d

Fw
FL

Force in the Width direction:

Fw =

1 r o L 2 V 2 d

Force in the Length direction:

1 r o W 2 FL = V 2 d

Applications of Microactuations

Microgrippers An essential component in microrobots in assembly microassemblies and surgery Two gripping methods:
Gripping Arms Closing the gap,d

Electrodes

The normal plate electrodes V - Not practical b/c requiring more space.

V
Aligning the electrodes,L

The sliding plate electrodes V - Popular method. Can have many sets to make Comb drive actuators

A Typical Microgripper with Comb drive Actuators:

400 m 100 m
Drive Arm

Arrangement of electrodes:
V
E m n Ar nsio xte

10 m

Closure Arm

Required Voltage, v

Drastic reduction in required actuation voltage with increase of number of pairs of electrodes:

160 140 120 100 80 60 40 20 0 0 20 40 60 80 100 120 Number of Electrode Pairs

Applications of Microactuations
Miniature Microphones
A niche market in mobile telecommunications and intelligent hearing aides

Acoustic Wave Input

(air pressure wave) dB MPa Electrical signal output: Diaphragm: 1m thick C Acoustic holes Pressure equalization hole

dB = unit of noise level:

Air gap (2 m)

Backplate (2 m)

P dB = 20 log10 P o
where P = Air pressure (Pa) Po = Reference air pressure at threshold sound level

Most microphones are designed for 20-80 dB in the frequency range of 150-1000 Hz A major challenge in MEMS microphone design and manufacture is the packaging and integration of MEMS and CMOS integrated circuits for signal conditioning and processing

Applications of Microactuations
Micromotors Unlike traditional motors, the driving forces for micro motors is primarily the parallel electrostatic forces between pairs of misaligned electrically charged plates (electrodes), as will be demonstrated in the following two cases: Linear stepping motors: Two sets of electrodes in the form of plates separated by dielectric material (e.g. quartz film). One electrode set is fixed and the other may slide over with little friction. The two sets have slightly different pitch between electrodes
W/3 Moving set electrodes: Fixed set electrodes: A W W W A Pitch: w+w/3 B Dielectric material B C D Step Movements C D

Applications of Micro Actuations-Contd

W/3 Moving set electrodes: Fixed set electrodes: A W W W A Pitch: w+w/3 B Dielectric material B C D Step Movements C D

Energize the set A-A will generate a force pulling A over A due to initial misalignment. Once A and A are aligned, the pair B and B become misaligned. Energize the misaligned B-B will generate electrostatic force pulling B over B. It is now with C and C being misaligned. Energize C and C will produce another step movement of the moving set over the stationary set. Repeat the same procedure will cause continuous movements of the moving sets The step size of the motion = w/3, or the size of preset mismatch of the pitch between the two electrode sets.

Applications of Micro Actuations-Contd

Rotary stepping motors: Involve two sets of electrodes- one set for the rotor and the other for the stator. Dielectric material between rotor and stator is air. There is preset mismatch of pitches of the electrodes in the two sets.

Applications of Microactuations-Contd
Working principle of this rotary motor is similar to that in linear motors.
A micro motor produced by Karlsruhe Nuclear Research Center, Germany:

Rotor Stator

Gear for transmitting torque

Microvalves A special microvalve designed by Jerman in 1990. Circular in geometry, with diaphragm of 2.5 mm in diameter x 10 m thick. The valve is actuated by thermal force generated by heating rings. Heating ring is made of aluminum films 5 m thick. The valve has a capacity of 300 cm3/min at a fluid pressure of 100 psig. Power consumption is 1.5 W.
Electric Resistance Heating Rings Flexible Silicon Diaphragm INLET FLOW

Silicon Base

Constraint Base

FLOW OUTLET

Centerline

Micropumps
Electrostatically actuated micropump: An electrostatic actuated pump in 1992. The pump is of square geometry with 4 mm x 4mm x 25 m thick. The gap between the diaphragm and the electrode is 4 m. Pumping rate is 70 L/min at 25 Hz.
Deformable Silicon Diaphragm Electrode Pumping Chamber

Inlet Check Valve

Outlet Check Valve

Constraint Base

Low Pressure Fluid Inlet

High Pressure Fluid Outlet

Piezoelectrically actuated pump: An effective way to pump fluid through capillary tubes. Tube wall is flexible. Outside tube wall is coated with piezoelectric crystal film, e.g. ZnO with aluminum interdigital transducers (IDTs). Radio-frequency voltage is applied to the IDTs, resulting in mechanical squeezing in section of the tube (similar to the squeezing of toothpaste) Smooth flow with uniform velocity profile across the tube cross section.

Piezoelectric coating with transducer

Flow

V
Flexible Tube Wall

Micro Heat Pipes

Heat pipes = Closed systems that transport heat from heat source @ higher temperature to
heat sink @ lower temperature. They are often referred to as Heat pumps. Micro heat pipes provide promising solution to effective heat dissipation in micro and molecular electronics circuits as will be presented in Chapter 12. A pipe with triangular or trapezoidal x-section (dp 100 m) is in contacts with heat source, e.g., IC and a heat sink, e.g., ink r ce ambient cool air with cooling air by a fan. Heat S Heat Sou nser The pipe contains liquid, e.g., Ethanol Conde Section tic Liquid vaporizes near the heat source Adiaba The vapor flows towards heat sink due to rator Evapo temperature difference The vapor condenses in the motion due to drop in temperature Cross-Sections Vapor turns into liquid near the heat sink Heat The condensed liquid moves in the sharp LIQUID Sink corners towards the heat sink due to the capillary effect Heat VAPOR Source The liquid vaporizes upon arriving at the heat sink Elevation The heat transport cycle repeats itself as long as temperature differences between the heat source and sink maintain.

Microaccelerometers Accelerometers are used to measure dynamic forces associated with moving objects. These forces are related to the velocity and acceleration of the moving objects. Traditionally an accelerometer is used to measure such forces. A typical accelerometer consists of a proof mass supported by a spring and a dashpot for damping of the vibrating proof mass:
The accelerometer is attached to the vibrating solid body

Spring k Mass M

Vibrating Solid Body

Dashpot with damping C

Microaccelerometers-Contd The instantaneous displacement of the mass y(t) induced by the attached moving solid body is measured and recorded with respect to time, t. The associated velocity, V(t) and the acceleration (t) may be obtained by the following derivatives:
Vibrating Solid Body The accelerometer is attached to the vibrating solid body

Spring k Mass M

Dashpot with damping C

dy (t ) V (t ) = dt

and

dy (t ) d 2 y (t ) (t ) = = dt dt 2

The associated dynamic force of induced by the moving solid is thus obtained by using the Newtons law, i.e. F(t) = M (t), in which M = the mass of the moving solid. In miniaturizing the accelerometers to the micro-scale, there is no room for the coil spring and the dashpot for damping on the vibrating mass. Alternative substitutes for the coil spring, dashpot, and even the proof mass need to be found.

Microaccelerometers-Contd There are two types micro accelerometers available. (1) The cantilever beam accelerometer:
Silicon Cantilever Beam Mass, M Piezoresistor

Casing

ConstraintBase Base Constraint

Vibrating Base

In this design: Cantilever beam = coil spring; Surrounding viscous fluid = dashpot for damping of the proof mass The movement of the proof mass is carried out by the attached piezoresistor.

Microaccelerometers-Contd (2) Balanced force micro accelerometer: This is the concept used in the air-bag deployment sensor in automobiles In this design: Plate beam = proof mass; Two end tethers = springs Surrounding air = dashpot

Be

am

Stationary electrodes Moving electrode

Ac ce ler

Mo ve me

nt

ati o

The movement of the proof mass is carried out by measuring the change of capacitances between the pairs of electrodes.

Microgyroscopes
Gyroscope is a form of accelerometer that measures angular rotation rates. The change of the rotational speed () of a solid can induce Coriolis force (Fc). For a moving solid with a linear velocity v, any rotation at the rate can be related to the induced Coriolis force Fc by the following expression:
z
x y y x y x z

r r r Fc = 2mv x

V V
x

Fc

where m = the mass of the moving solid The sense and direction of the vectorial quantities in the above expression are illustrated to the right Gyroscopes can thus be used as compass in self-correction navigation systems for ships and aircraft, space crafts, and Segway human transport.

Fc
z

(a)
z

(b) Fc V
x z y x y

x y

z y

V Fc
z

(c)

(d)

Microgyroscope Structure Microgyroscopes are attached to the moving solid structure. Like microaccelerometers, microgyroscopes use comb-drive actuators and beam or tether springs to actuate the motion of the proof mass, as well as to sense and measure the rate of rotation of the gyroscope.
y Gyro Frame x-Position
y-Spring for Force Measurements

Resonator for Linear Motion Generation

Proof Mass

x-Spring x

y-Position
y

The induced Colioris forces are used to self-regulate the navigation of the moving structures.

End of Chapter 2