EC465 MEMS MODULE V

Bonding techniques for MEMS

Bonding is one of the essential technologies for MEMS andMicrosystems packaging. It is
required in the following three principal areas:

1. Joining one substrate to another substrate such as wafer towafer or wafer to other
supporting wafers made of glass, quartz, sapphire, ceramic and metals.
2. Securing micro components to support substrates such asmounting silicon dies to their
constraint bases made ofglass or ceramic
3. Attaching wires and electrical leads to and from transducers in micro devices

Types (1) and (2) relate to the “surface bonding”,whereas Type (3) is referred to as “wire
bonding”. Two principal requirements for achieving good surfacebonding for micro components
are (1) intimate surface contact, and (2) temperature. "Intimate surface contact" isrequired for
quality of the bonding and "temperature" supplies the required energy for the bonding. Intimate
surface contact requires adequatecontact pressure as well as "clean" contact surfaces.

Surface bonding
There are four (4) techniquesavailable for surface bonding in MEMS and microsystems:

(1) Adhesives

(2) Eutectic soldering

(3) Anodic bonding

(4) Silicon fusion bonding (SFB)

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

1
 EC465 MEMS MODULE V

Adhesives

This bonding technique has been in existence for decades.Many die attaches in integrated
circuits are produced by thismethod. It is the least expensive way to bond two separatesurfaces.
A generic arrangement of this bonding technique isillustrated in figure:

Fig: Bonding by adhesion

The chamber in which bonding takes place supplies therequired heating to the substrate
for reaching the desiredbonding temperature. This temperature usually is close to butbelow the
glass transition temperature, Tg, of the thin adhesivethat is dispensed on the top of the substrate
by a micro dispenser. The part to be bonded is then placed on the top of theadhesive film.
Mechanical pressure is normally applied toensure the quality of bonding. The chamber must be
keptclean, free from dust or other solid contaminant. Often thebonding operation is carried out in
vacuum.

Epoxy resinand silicone rubbersare two commonly used adhesives. Epoxy resins
provide flexibility for the bonded dies as well as good sealing. Good bonding rely on surface
treatments and curing processcontrol. They are also vulnerable to thermal environment. The
bond should be kept below Tg, which is normally 150 -1700C.Soft silicone rubbersare used for
bondingparts require “ flexibility .”It is vulnerable to chemicals and air.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

2
 EC465 MEMS MODULE V

Eutectic soldering

Eutectic bonding involves the diffusion of atoms of eutecticalloys into the atomic
structures of the materials to be bondedtogether, and thus forms solid bonding of these materials.
A commonly used material to form a eutectic alloywith silicon is thin film made of gold (Au) or
alloys thatinvolve gold.

Eutectic bonding takes place when the assembly of the twobonding surfaces, e.g. silicon
substrate, with a eutectic alloycomponent such as gold is heated to the temperature above
theeutectic temperature. A eutectic temperature is the lowestfusion temperature of an alloy with
melting point lower thanthat of any other combination of the same components. Atsuch time, the
atoms of the interface material, e.g,. Au, start todiffuse rapidly into the contacting substrates of
silicon. Sufficient migration of these atoms into the bonding substratesurface will result in the
formation of a eutectic alloy. EgAu-Si. As the temperature continues to rise above the
eutectictemperature, more eutectic alloy is formed. This process willcontinue until the atoms,
e.g,.the gold atoms, in the eutecticalloy at the interface material near complete
depletion.Thenewly created eutectic alloy at the interface serves as a solidbond as well as
hermetic sealing for many MEMS andmicrosystems applications.

A thin film alloy made of 80% Au-20% Snis sandwiched betweenthe silicon piezoresistor
and the silicon substrate as in figure. The gold-tin alloy film was 0.025 mm thick. Aweight that
was equivalent to 1 MPa was placed on the top ofthe upper silicon substrate to ensure intimate
contact of thebonding surfaces and the thin gold-tin film during the bondingprocess. The
assembly was placed in a clean evacuatedenclosed chamber in which heat was supplied. The
chamberwas gradually heated to 280°C and held at that temperaturefor one hour. The chamber
was then cooled to room temperature in three hours.As for bonding by adhesives, eutectic
bonding requiresproper process control.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

3
 EC465 MEMS MODULE V

Fig: Eutetic bonding chamber

Anodic bonding
This bonding technique is most commonly used in bondingwafers of different materials.
It is also called electrostaticbonding, or field-assisted thermal bonding.Anodic bonding is
popular in microsystems packaging due tothe relatively simple set-up with inexpensive
equipmentrequired. Anodic bonding providesreliable hermetic sealing that is important for
applications inmicro valves and channels in microfluidics networks, as well asfor micro pressure
sensor dies. Another major advantage ofthis bonding technique is that bonding can take place at
low tomoderately high temperatures in the range 180 to 5000 Cwhich results in low risk of
residual stress and strain in thematerials after the bonding.

Anodic bonding has been used to bond wafers made of thefollowing materials:

Glass-to-glass

Glass-to-silicon

Glass-to-silicon compounds, e.g. GaAs

Glass-to-metals

Silicon-to-silicon

However, the most common application of anodic bonding isto bond wafers made of Pyrex glass
or quartz to silicon wafers.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

4
 EC465 MEMS MODULE V

Fig: Anodic bonding of Silicon to glass

Fig: Formation of SiO2 layer in anodic bonding

The set up for anodic bonding of a sodium rich wafer e.g, Pyrex 7740 glass (rich in
sodium) to silicon wafer is shown in figure. A modest weight is normally placed on the top of the
pyrex glass wafer to ensure good contacting pressure. An electric field with a 200-1000 DC
voltage is applied to the system. The sandwiched delectricpyrex glass and silicon wafers between
the two electrodes form an effective parallel plate capacitor. The bonding between them is
accomplished by the formation of an extremely thin layer of SiO2 interface as the result of the
applied electric field. Under the influence of electric field, the sodium ions (Na+) in the glass are
attracted towards the negatively charged cathode, which leaves behind an Na+ depletion zone
with negatively charged O2 ions. These O−
2 ions can be chemically bonded to the contacting Si
+

ions with the supplied heating of the system, and form a very thin layer of SiO2of approximately
20nm at the interface. Anodic bonding normally takes 10 to 20 minutes to bond 100mm diameter
silicon and glass wafers in an evacuated chamber at 4500C with 1000 V DC. Successful anodic

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

5
 EC465 MEMS MODULE V

bonding requires the maintaining of constant temperture, applied voltage and current density
conditions during the bonding process. The surfaces of the wafers must be flat and smooth.

Silicon fusion bonding (SFB)

SFB is an effective and reliable technique for bonding two silicon wafers or substrates
without the use of intermediate adhesives. It is relatively simple and inexpensive bonding
process. SFB has been used to bond

• Silicon-to-silicon
• Silicon with oxide-t o -silicon
• Silicon with oxide-t o silicon with oxide
• GaAs-to-silicon
• Quartz-to-silicon
• Silicon-to-glass

SFB is accomplished by the induced chemical forces at the interface and bonding is spontaneous
at oxidising ambient, to be followed by annealing at high temperature.

SFB by hydration

This is a common process that is used in SFB. The procedureinvolved in this bonding process is
first to introduce oxygen(O)-hydrogen (H) bonds, called the silanol bond, at the interface through
a hydration process. Dehydration occurs whenthe wafers with silanol bonds at their surfaces are
put intointimate contact, resulting in the formation of water (H2O)vapour that escapes to the
environment upon subsequentannealing at high temperature. What are left behind are
theexcessive oxygen atoms that form a siloxane network at theinterface to form strong bonding
of the two wafer surfaces.

The chemical reaction for this process is

(Si with OH) + (OH with Si) → H2O + (Si-O-Si)

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

6
 EC465 MEMS MODULE V

Fig: SFB by hydration

The two oxidised siliconwafers are soaked in the (H2O2-H2SO4) solvent in the hydration
process. An O-H bond (silanol bond) will be developed onthe oxidised layer surface after the
wafers are removed fromthe acid solution and rinsed with deionised water. The twowafers are
then placed one on top of the other to form a bond. Heating the stacked wafers will drive the O2
and H2 molecules in the contacted O-H bond to form a Si-O-Si bond between the two wafers.

SFB by hydrophilic

Another popular SFB technique involves the use of a hydrophilic process. The wafers are soaked
in boiling nitric acid tocreate the hydrophobic surfaces for bonding. Bonding of thetwo wafers
takes place even at room temperature. However,an annealing process at high temperature is
normally appliedto enhance the bonding. The bonding mechanism in hydrophilic is not the same
as that in hydration. Hydrogen-inducedvan der Waals forces are believed to be the principal
bondingforces for the wafers in the hydrophilic process.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

7
 EC465 MEMS MODULE V

Silicon-on-Insulator (SOI)

SOI is a process that is used in microelectronics to avoid leakage of charges in p-n

junctions. It involves bonding the silicon with an amorphous material such as SiO2. Silicon is a
semiconductor which conducts electricity at high electric potentials or at elevated temperature.
For example, siliconbecomes increasingly electrical conductive at a temperature above 1250C.
This transformation of silicon from a semiconductor to a conductor limitssilicon sense elements
to being effective in elevated temperature applications. Theprocess of silicon-on-insulator offers
a viable solution to this problem.

The process uses two silicon substrates, one with one of its surfaces heavily doped with
boron atoms to produce a layer of p-silicon, and the other with a thin silicon oxide film on one of
its faces, as shown in Figure (a). The two substrates are then mounted one on the topof the other
as shown in Figure (b). The process of silicon fusion bonding joins thetwo substrates together.
The bonded substrates are then exposed to etching to etch theexposed surfaces of the bonded
substrates. The heavily p-doped region can act as an etch stop. Consequently, one may obtain
either ap-silicon layer on the SiO2 insulator, or sandwiched silicon substrates withSiO2insulator
in between as in Figure (c). This technique, which involves bonding of substrates followed by
etching the bonded substrates is termed as bonding-and – etchback technique

Fig: Silicon-on-insulator process

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

8
 EC465 MEMS MODULE V

Wire bonding

Wire bonding provides electrical connection to or from the core elements. Common wire
materials are gold and aluminum. Other wire materials include copper, silver, and palladium.
Wiresizes are in the order of 20 to 80 µm in diameter. Three wire bonding techniques
arecommonly used by the industry:

(1) thermocompression

(2) wedge-wedge ultrasonic wire bonding

(3) thermosonic

Thermocompression Wire Bonding

The principle of thermocompressionwire bonding is to press heated metal balls onto metal pads.
In Figure (a), the metal wire is fed through a capillary bonding tool withthe tip of the wire above
the metal pad. A torch heats the wire tip to about 400°C, atwhich temperature the wire tip tums
into the shape of a ball. At this point, the tool which has a hemispherical mold shape at the end,
is lowered to press the ball onto thepad (Fig. (b)). After typically 40 ms of pressing, the tool is
retracted and the wireis bonded to the pad, as illustrated in Figure (c). The solid bonding of the
wireball to the flat metal pad is accomplished by two physical actions: the plastic deformation of
the ball onto the pad by the tool and the atomic interdiffusion of the twobonded materials.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

9
 EC465 MEMS MODULE V

Wedge-wedge Ultrasonic Bonding

Unlike thermocompressionbonding,ultrasonic wire bonding is a low-temperature process. The

bonding tool is in theshape of a wedge, through which the wire is fed. The energy supplied for
the bonding is an ultrasonic wave that is generated by a transducer that vibrates the tool ata
frequency from 20 to 60 kHz. The bonding tool travels in parallel to the bondingpad as
illustrated in Figure (a). After moving overtop the desired bonding pad, the wedge tool is
lowered to the pad's surface. A compressive force is appliedand a burst of ultrasonic energy is
released to break down the contact surfaces andachieve the desired surface bonding (Fig. (b)).
Typical time required for bonding is about 20ms

Thermosonic Bonding

This technique combines the previous two wire bonding techniques. Ultrasonic energy is used
with thermocompression at moderate temperatures of 100 to 150°C. A capillary tool is used to
provide ball joints.For MEMS application, wire bonding is often combined with ball-wedge
jointsor wedge-wedge joints as illustrated in figure below

Fig: Wire bonding of microsystem

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

10
 EC465 MEMS MODULE V

Sealing

Sealing is a key requirement in MEMS and microsystems packaging.Hermetic sealing is

essential in devices or systems such as: microfluidic, optoMEMS, bioMEMS, pressure sensors,
etc.Mechanical sealing of the interfaces of mating components by epoxy resin iscommon in
microchannels in fluidic systems. Thismethod of sealing is usually adequate for systems that do
not operate at elevatedtemperature and/or with toxic working media, and they are not expected to
last for along period of time.

Sealing by Microshells

Microshells are produced to protect the delicate sensing or actuating elements in

microdevices.Asurface micromachining technique is used to produce microshells. The procedure
involves depositing a sacrificial layer over the die to be protected, as shown inFigure (a). A shell
material is then deposited over the sacrificial layer. An etchingprocess then follows to remove
the sacrificial layer. Consequently, a gap space between the die and the microshell is created as
shown in Figure (b). These gapscan be as small as 100 nm

Fig: Sealing by microshell

Reactive Sealing Technique

This technique relies on specific chemical reactions to produce the necessary sealing of the
mating components. As illustrated in figure the encapsulant cover for the die is initially placed
on the top of the die/constraint base with small gaps (Fig (a)). The unit is subject to a chemical
reaction such as the thermal oxidation process. The growth of SiO2 from both ends of the silicon

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

11
 EC465 MEMS MODULE V

encapsulant and the constraint base can provide a reliable and effective seal for the encapsulated
die.

Fig: Reactive sealing

Assembly of microsystems

Assembly of microsystems with components whose sizes range from 1 µm to a few millimeters
presents major challenges to engineers in terms of reliability and cost. Following are the reasons
for the high cost associated with microassembly.

1. There is lack of standard procedures and rules for such assemblies. Products are
assembled according to the specific procedures chosen on the basis of either individual
customer requirements or the personal experience of the design engineer.
2. There is lack of effective tools for micro assembly. Tools such as microgrippers
manipulators, and robots are still being developed. Microassemblies also require reliable
visual and alignment equipment such as stereo electron microscopes and electron-beam,
UV-stimulated beam, or ion beam imaging systems specially designed for microsystem
assembly.
3. There is a lack of established methodology in setting proper tolerances of parts in
insertion and assembly.
4. Microcomponents to be assembled are mostly made by physical-chemical processes that
have strong material dependence. Traditional assembly techniques are not suitable for
microdevices because of the minute size of the components and the close tolerances in
the order of submicrometer. Moreover, chemical and electrostatic forces dominate in

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

12
 EC465 MEMS MODULE V

microassembly whereas gravity and physics are primary considerations in macro

assembly.

The lack of effective tools and assembly strategies has resulted in lengthy time required for
microsystem assembly.

Despite the many differences between microassembly and traditional assembly in

methodologies and tools, major steps are quite similar between the two scales in automated
assembly. These steps include:

1. Part feeding: Some of the part feeding techniques such as the common tape-bonded
feeders used in microelectronics can be adopted in microsystem assembly.
2. Part grasping: Micro grippers, manipulators, and robots are desirable tools for this task.
However, these tools cannot function properly for handling minute parts without
intelligent end effectors. An intelligent end effector requires the integration of gripping,
positioning, sensing, and orientation for accurate alignment in microscale of mating parts.
3. Part mating: At the micrometer scale, electrostatic and chemical forces dominate the
interaction between grippers and parts, as well between the mating parts. Special design
of grippers that can discharge these forces for possible stiction is necessary for easy
release of parts from the gripper and for mating of small components.
4. Part bonding and fastening: Most of the bonding techniques involve microfabrication
processes. Automated assembly is possible with batch fabrication of the bonding parts.
Other methods of joining parts, including pulsed laser deposition, welding, soldering,
"snap-fits" based on surface chemistry, and thin-film chemistry can be used for fastening
microparts.
5. Sensing and verification: Three-dimensional machine vision systems such as stereo
microscopy are effective for visual identification of parts and for part alignment. Other
types of microsensors, e.g., tactile and thermal sensors, are also required in assembly and
inspection. Near- and far-field infrared (IR) sensors can be used for microthermal
feedback for monitoring welds or solder joints. Ideally these sensors should be integrated
with the grippers and/or other process tooling.One common problem with most of these
Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM
13
 EC465 MEMS MODULE V

sensors is the short depth of field near the wavelength of the light source in optical
sensing systems. Another problem is the requirement for the sensing head to access the
parts that need to be sensed.

As in traditional automated assembly, both serial and parallel assembly processes have
been used in microassembly. Self-assembly is an attractive option for free-form fabrication
of certain component groups of a microsystem. Parallel assembly is attractive for mass
production of microsystems.

Overview of MEMS areas

RFMEMS

RF MEMS refers to the design and fabrication of dedicated MEMS for RF (integrated)
circuits. There are various types of RF MEMS components, such as RF
MEMS tunable inductors, RF MEMS switches, switched capacitors and varactors. RF MEMS
components are fragile and require wafer level packaging or single chip packaging which allow
for hermetic cavity sealing. They are biased electrostatically using a bipolar NRZ drive voltage,
in order to avoid dielectric charging and to increase the lifetime of the device.

Capacitive RF MEMS switches

RF MEMS switches are the specific micromechanical switches that are designed to
operate at RF to mm wave frequencies. MEMS switches provide high isolation when open, low
insertion loss when closed, and can be operated at low power consumption. Because of
electromechanical isolation, the RF circuit doesn’t leak or couple significantly to the actuation
circuit.. Radio Frequency Micro Electro Mechanical Switches (RF MEMS) classification
depends on the type of actuation, deflection axis, contact type, circuit configuration, and
structure configuration. The most used RF MEMS mechanical structures are the cantilever beam
and the air bridge structures.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

14
 EC465 MEMS MODULE V

Fig: Capacitive RF MEMS switch

An air bridge base capacitive RF MEMS is shown here. Gold (Au) is used as a beam
material, and Silicon Nitride (Si3N4) with dielectric constant 8.5 is used as a dielectric material.
Silicon nitride thin film dielectrics are used in capacitive radio frequency micro-
electromechanical systems (MEMS) switches since they provide a low insertion loss, good
isolation, and low return loss. A capacitor is built up between the fixed electrode and movable
electrode

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

15
 EC465 MEMS MODULE V

In air bridge type MEMS switch, beam is fixed at both the ends and voltage is applied in
the middle of the beam to note down the displacement of the beam towards substrate. The
displacement is maximum when we go on increasing the applied voltage. The actuation voltage
or applied voltage or pull in voltage is the maximum voltage at which the electrostatic force
becomes superior over mechanical restoring force, causes pulling down of MEMS device
towards the ground plane. Initially the applied input voltage is 1mv, there is no deformation in
the switch under this condition (input is equal to output). But in the second case a 5v is added to
1mv,then some electrostatic force is created between the electrodes resulting in the deformation
of cantilever and it touches the ground (output is zero).

RF MEMS Capacitor

The capacitors usually used in RF MEMS are variable type capacitor. The capacitance is given
𝜀𝐴
by 𝐶 = .This capacitance can be tuned by changing either area , distance between the plates,
𝑑

or the dielectric constant. Therefore RF MEMS tunable capacitors are categorized according to
their tuning parameters:

• MEMS gap-tuning capacitors;

• MEMS area-tuning capacitors;
• MEMS dielectric tunable capacitors.

The gap-tuning capacitors can be made with two parallel electrodes. The lower electrode is fixed
while the upper is connected to a spring and is movable. The gap between the electrodes changes
with change in applied voltage. This principle can be implemented with three plates, the middle
one being movable and the other two being fixed. The bridge switch can be used also as a
variable capacitor.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

16
 EC465 MEMS MODULE V

Fig: Gap tuning capacitors

Comb like interdigitable tunable capacitors

The most common area tuning capacitor is the interdigital comb structure capacitor.One finger is
fixed while the other is moving. The tuning range is determined by the finger length.

Fig: Comb like interdigitable tunable capacitors

The interdigital structure can also be used as dielectric tunable capacitor. Both combs can
be fixed. The dielectric material (STO), grows in size as the temperature increases. Hence its
dielectric constant changes.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

17
 EC465 MEMS MODULE V

Fig : Dielectric Tunable capacitors

Fig: Cross section of RF MEMS capacitor

RF MEMS capacitor networks presented here are based on two-plate capacitors with dielectric
materials in between. Membranes included in the design are held by four suspensions. The total
capacitance is obtained from two different sections. The first section is with solid dielectric
media and photoresist layer, and the second one is with the dielectric media and additional air
gap. The dielectric medium consists of SiO2, Si3N4 and an air gap. The combination of SiO2 and
Si3N4 is used to compensate unwanted mechanical strain effect. In order to have an air gap, a
photoresist layer, used as a sacrificial layer, is deposited and is later removed at the end of the
process.

BioMEMS

The term “BioMEMS” has been a popular terminology in the MEMS industry in recent
years.

BioMEMS include the following three major areas:

• Biosensors for identification and measurement of biological substances,

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

18
 EC465 MEMS MODULE V

• Bioinstruments and surgical tools, and

• Bioanalytical systems for testing and diagnoses.
These sensors are extensively used in medical diagnosis, environmental protection, drug
discovery and delivery, etc.
Biomedical Sensors are used to measure biological substances as well as for medical
diagnosis purposes. Biomedical Sensors These sensors can analyze biological samples in
quick and accurate ways. These miniaturized biomedical sensors have many advantages over
the traditional instruments. They require typically a minute amount of samples and can per
form analyses much faster with virtually no dead volume

Working principle:
A small sample of blood is introduced to a sensor with a polyvinyl alcohol solution. Two
electrodes are present in the sensor: a platinum film electrode and a thin Ag/AgCl film (the
reference electrode. The glucose in patient’s blood sample reacts with the O2 in the polyvinyl
alcohol solution and produces H2O2.
Glucose +O2 → gluconolactone + 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 work on the principle of the interaction of the analytes that need to be detected
with biologically derived biomolecules, such as enzymes of certain forms, antibodies, and other
forms of protein. These biomolecules, when attached to the sensing elements, can alter the output
signals of the sensors when they interact with the analyte

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

19
 EC465 MEMS MODULE V

Proper selection of biomolecules for sensing elements (chemical, optical etc) can be used for
the detection of specific analyte

MOEMS

Micro Opto Electro Mechanical systems (MOEMS) are a special class of Micro-electro-
mechanical systems (MEMS) which involves sensing or manipulating optical signals on a very
small size scale using integrated mechanical and electrical systems. MOEMS includes a wide
variety of devices including optical switch, optical cross-connect, tunable VCSEL,
microbolometers amongst others. These devices are usually fabricated using standard
micromachining technologies using materials like silicon, silicon dioxide, silicon nitride and
gallium arsenide. The digital micromirror device, or DMD, is the micro-opto-electromechanical
system (MOEMS) that is the core of the trademarked DLP projection technology from Texas
Instruments.

Micro-optical switches

The direction in which the light beam is reflected can be changed by rotating the mirror
to different angles, allowing the input light to be connected to any output port. In a MEMS
optical switch, a micro-mirror is used to reflect a light beam. This type of optical switch has been
realized for the first time through the fusion of various techniques such as micro-machining
techniques for fabricating the mirror, optical design techniques for achieving low-loss optical
connections, and control techniques for positioning the mirror accurately. Since this device can

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

20
 EC465 MEMS MODULE V

switch large numbers of optical signals simultaneously, it can be used as a trunk switch for
handling large amounts of traffic, and as a switch in large urban communication networks.

Fig: Principle of MOEMS

MEMS Tunable VCSEL

Vertical Cavity Surface Emitting Lasers (VCSELs) are semiconductor-based devices that
emit light perpendicular to the chip surface. VCSELs were originally developed as low-cost,
low-power alternatives to edge-emitting diodes, mainly for high-volume datacom
applications.MEMS-tunable VCSELs utilize microelectromechanical mirror systems (MEMS) to
vary the cavity length of the laser, thereby tuning the output wavelength. The limited tuning
range and output power of these devices have precluded them from being used in OCT (Optical
coherence tomography) applications.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

21
 EC465 MEMS MODULE V

In order for a MEMS-tunable VCSEL to be successful for applications in OCT, it needs to meet
certain standards:

• Rapid Sweep Speed

• Broad Tuning Range

• Long Coherence Length

• High Laser Output Power

Fig: 3D view of MEMS Tunable VCSEL

Fig : MEMS Tunable VCSEL

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

22
 EC465 MEMS MODULE V

The tunable VCSEL is constructed by bonding a wide-gain, indium phosphide (InP)-based

quantum-well active region to a gallium arsenide (GaAs)-based oxidized mirror. An
electrostatically actuated dielectric mirror suspended over the top of this structure and separated
by an air gap moves to generate 1310 nm tunable emission as the device is pumped by a 980 nm
laser source. The emission is then coupled to a broadband semiconductor optical amplifier,
which not only increases output power but also shapes the spectral output of the tunable VCSEL
before it is input to an OCT system.

NEMS

Nanoelectromechanical systems (NEMS) are a class of devices integrating electrical and

mechanical functionality on the nanoscale. NEMS form the next logical miniaturization step
from so-called MEMS. NEMS typically integrate transistor-like nanoelectronics with mechanical
actuators, pumps, or motors, and may thereby form physical, biological, and chemical sensors.
The name derives from typical device dimensions in the nanometer range, leading to low mass,
high mechanical resonance frequencies, large quantum mechanical effects and a high surface-to-
volume ratio useful for surface-based sensing mechanisms. A key application of NEMS is atomic
force microscope tips. The increased sensitivity achieved by NEMS leads to smaller and more
efficient sensors to detect stresses, vibrations, forces at the atomic level, and chemical signals.
Many of the commonly used materials for NEMS technology have been carbon based,
specifically diamond, carbon nanotubes and graphene. This is mainly because of the useful
properties of carbon based materials which directly meet the needs of NEMS.

Nanoelectromechanical relay

A nanoelectromechanical (NEM) relay is an electrically actuated switch that is built on

the nanometer scale using semiconductor fabrication techniques. They are designed to operate in
replacement, or in conjunction, with traditional semiconductor logic. While the mechanical
nature of NEM relays makes them switch much slower than solid-state relays, they have many
advantageous properties, such as zero current leakage and low power consumption. A typical

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

23
 EC465 MEMS MODULE V

NEM Relay requires a potential on the order of the tens of volts in order to "pull in" and
have contact resistances on the order of gigaohms.

Fig: Schematic of a three terminal electromechanical relay

A NEM relay can be fabricated in two, three, or four terminal configurations. A three
terminal relay is composed of a source (input), drain (output), and a gate (actuation terminal).
Attached to the source is a cantilevered beam that can be bent into contact with the drain in order
to make an electrical connection. When a significant voltage differential is applied between the
beam and gate, and the electrostatic force overcomes the elastic force of the beam enough to
bend it into contact with the drain, the device "pulls in" and forms an electrical connection. In the
off position, the source and drain are separated by an air gap. This physical separation allows
NEM relays to have zero current leakage, and very sharp on/off transitions.

The nonlinear nature of the electric field, and adhesion between the beam and drain cause
the device to "pull out" and lose connection at a lower voltage than the voltage at which it pulls
in. This hysteresis effect means there is a voltage between the pull in voltage, and the pull out
voltage that will not change the state of the relay, no matter what its initial state is. This property
is very useful in applications where information needs to be stored in the circuit, such as in static
random-access memory.

Nanoelectro mechanichal switched capacitor

Consider a nanoelectromechanical (NEM) device consisting of a source, a drain and a

gate electrode. A vertically aligned multiwalled carbon nanotube (MWCNT) was grown on the
source electrode and was coated with a dielectric layer (SiNx ) and a metal layer (Cr) to form a
Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM
24
 EC465 MEMS MODULE V

CNT– insulator– metal (CIM) capacitor. The MWCNT grown on the drain electrode is the
mechanically active element of the cell; electrostatic forces cause it to bend and make contact
with the CIM structure on the source.There is no nanotube on the gate electrode.

Fig: Switched capacitor

In operation, the source is electrically connected to ground, the drain is connected to the
bit line, which has a constant positive voltage applied to it, and the gate is connected to the word
line. When this word line is positively biased (to a value higher than the bit line voltage), the
nanotube on the drain experiences a repulsive electrostatic force from the gate electrode and an
attractive electrostatic force from the source electrode, causing it to deflect until it makes contact
with the metal electrode on the capacitor .On contact, a transient current flows to charge the CIM
capacitor. This charge is used to represent a bit of stored information. When the gate bias is
removed, the electrostatic force giving rise to the deflection of the drain MWCNT is also
removed, and the nanotube springs back to the non-contact ‘OFF’ position.

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

25
 EC465 MEMS MODULE V

PREVIOUS YEAR KTU QUESTIONS

1. What is meant by BioMEMS. Discuss the challenges involved in BioMEMS. List three
applications of BioMEMS
2. Explain anodic bonding and Silicon Fusion Bonding
3. Explain with figures two applications which use NEMS technology
4. Explain with figures two RF MEMS applications
5. Explain the bonding techniques with figures a) Silicon-on-Insulator b) Wire bonding
6. Explain any one application of MOEMS with figures
7. Explain anodic bonding

PREVIOUS UNIVERSITY QUESTIONS

1. List a few RF MEMScomponents and give their applications

2. Write notes on optical MEMS
3. Explain the working of any two RF MEMS components and give their applications

Sifna N S, Assistant Professor, MESCE, KUTTIPPURAM

26