NANO-ELECTROMECHANICAL SYSTEM

ABSTRACT

Nanoelectromechanical systems are evolving, with new scientific studies and technical
applications emerging. Mechanical devices are shrinking in thickness and width to reduce
mass, increase resonant frequency, and lower the force constants of these systems.
Advances in the field include improvements in fabrication processes and new methods for
actuating and detecting motion at the nanoscale. Lithographic approaches are capable of
creating freestanding objects in silicon and other materials, with thickness and lateral
dimensions down to about 20 nanometers. Similar processes can make channels or pores of
comparable dimensions, approaching the molecular scale. This allows access to a new
experimental regime and suggests new applications in sensing and molecular interactions.

A host of novel applications and new physics could be unleashed as Micro-Electro-

Mechanical Systems (MEMS) shrink towards the nano scale. The time is ripe for a
concerted exploration of Nano-Electro-Mechanical-Systems (NEMS) - i.e. machines,
sensors, computers and electronics that are on the nano-scale. Many years of research by
university, government, and industrial groups have been devoted to developing cutting-edge
NEMS technologies for enabling revolutionary NEMS devices. NEMS technology allows
the complex electromechanical systems to be manufactured using batch fabrication
techniques, increasing the reliability of the sensors and actuators to greater than that of
integrated circuits. Thus, it provides a way to integrate mechanical, fluidic, optical, and
electronic functionality on very small devices, ranging from 1 nano meter to 100 nano
meters. NEMS devices can be so small that hundreds of them can fit in the same space as
one single micro-device that performs the same function and are lighter, more reliable and
are produced at a fraction of the cost of the conventional methods. Many device designs
have been proposed, some have been developed, and fewer have reached commercialization

1
 Chapter 1:

Introduction
Nano-Electro-Mechanical Systems (NEMS) is the integration of mechanical
elements, sensors, actuators, and electronics on a common silicon substrate through nano
fabrication technology. While the electronics are fabricated using integrated circuit (IC)
process sequences (e.g., CMOS, Bipolar, or BICMOS processes), the nano-mechanical
components are fabricated using compatible "micromachining" processes that selectively
etch away parts of the silicon wafer or add new structural layers to form the mechanical and
electromechanical devices.

Nano-electronic integrated circuits can be thought of as the "brains" of a system and

NEMS augments this decision-making capability with "eyes" and "arms", to allow nano
systems to sense and control the environment. Sensors gather information from the
environment through measuring mechanical, thermal, biological, chemical, optical, and
magnetic phenomena. The electronics then process the information derived from the sensors
and through some decision making capability direct the actuators to respond by moving,
positioning, regulating, pumping, and filtering, thereby controlling the environment for
some desired outcome or purpose.

NEMS promises to revolutionize nearly every product category by bringing together

silicon-based nano-electronics with micromachining technology, making possible the
realization of complete systems-on-a-chip. NEMS is an enabling technology allowing the
development of smart products, augmenting the computational ability of nano-electronics
with the perception and control capabilities of nano sensors and nano actuators and
expanding the space of possible designs and applications. Despite such optimistic statistics,
investment in NEMS design and production is insufficient. Most NEMS devices are
modeled using analytical tools that result in a relatively inaccurate prediction of
performance behavior. As a result, NEMS design is usually trial and error, requiring several
iterations before a device satisfies its performance requirements.

NEMS have a host of intriguing attributes. They offer access to fundamental

frequencies in the microwave range; Q’s, i.e. mechanical quality factors, in the tens of
thousands (and quite possibly much higher); active masses in the femtogram range; force
sensitivities at the attonewton level; mass sensitivity at the level of individual molecules,
heat capacities far below a “yoctocalorie” [2] — this list goes on. These attributes spark the
imagination, and a flood of ideas for new experiments and applications ensues. Of course,
in time this initial enthusiasm gives way to deeper

Figure 1: 3 terminal electromechanical device

Figure 2. SiC NEMS. This first family of submicron doubly clamped SiC beams
exhibit fundamental resonant frequencies from 2 to 134 MHz.

reflection, and a multitude of new questions and concerns emerge. Prominent among
these is what will be our ultimate ability for optimally controlling and engineering these
miniature systems. Clearly, the characteristic parameters of NEMS are extreme by all
current measures. This paper is in three main sections: in the first, with this latter point in
mind, I will attempt to convey a balanced introduction to the attributes of NEMS. In the
second section of the paper I offer my thoughts and projections regarding the most crucial
aspects of NEMS engineering. How shall transducers and actuators be realized at the
nanoscale? How shall surface properties be controlled? How can reproducible fabrication be
attained? The final section of the paper concerns ultimate limits. As we move forward in the
development of NEMS it will become increasingly apparent what aspects will be
susceptible to improvement through systematic engineering, and what hard, immutable
limits are imposed by their fundamental physics. However, certain issues are already clear
at the outset. I will try to summarize those that seem most apparent at this juncture.

1.1: Micro electro-mechanical system

MEMS are an abbreviation for Micro Electro Mechanical Systems. This is

a rapidly emerging technology combining electrical, electronic, mechanical, optical,
material, chemical, and fluids engineering disciplines. As the smallest commercially
produced "machines", MEMS devices are similar to traditional sensors and actuators
although much, much smaller. E.g. complete systems are typically a few millimeters across,
with individual features / devices of the order of 1-100 micrometers across.

Figure 1.1 Electromechanical system

MEMS devices are manufactured either using processes based on Integrated Circuit
fabrication techniques and materials, or using new emerging fabrication technologies such
as micro injection molding. A typical MEMS fabrication technology may have a 5 step
process. Due to the limitations of this "traditional IC" manufacturing process MEMS
devices are substantially planar, having very low aspect ratios (typically 5 -10 micro meters
thick). It is important to note that there are several evolving fabrication techniques that
allow higher aspect ratios such as deep x-ray lithography, electro deposition, and micro
injection molding.

MEMS devices are typically fabricated onto a substrate (chip) that may also contain
the electronics required to interact with the MEMS device. Due to the small size and mass
of the devices, MEMS components can be actuated electro statically (piezoelectric and
bimetallic effects can also be used). The position of MEMS components can also be sensed
capacitively. Hence the MEMS electronics include electrostatic drive power supplies,
capacitance charge comparators, and signal conditioning circuitry.

Figure 1.2: Linear comb drive

A common MEMS actuator is the "linear comb drive" (shown above) which
consists of rows of interlocking teeth; half of the teeth are attached to a fixed "beam", the
other half attach to a movable beam assembly. Both assemblies are electrically insulated. By
applying the same polarity voltage to both parts the resultant electrostatic force repels the
movable beam away from the fixed. Conversely, by applying opposite polarity the parts are
attracted. In this manner the comb drive can be moved "in" or "out" and either DC or AC
voltages can be applied. The magnitude of electrostatic force is multiplied by the voltage or
more commonly the surface area and number of teeth. Commercial comb drives have
several thousand teeth, each tooth approximately 10 micro meters long. Drive voltages are
CMOS levels.
 The linear push / pull motion of a comb drive can be converted into rotational motion by
coupling the drive to push rod and pinion on a wheel. In this manner the comb drive can
rotate the wheel in the same way a steam engine functions.

Chapter 2:

NEMS ATTRIBUTES

2.1 : NEMS as multiterminal electromechanical devices:

It is a generic multiterminal electromechanical device, where electromechanical transducers

provide input mechanical stimuli to the system, and read out its mechanical response. At
additional control terminals, electrical signals—either quasistatic or time varying—can be
applied, and subsequently be converted by the control transducers into forces to perturb the
properties of the mechanical element. NEMS devices fit the above-described general
portrayal. We can further classify the existing NEMS as resonant and quasistatic. In this
review, our focus will primarily be on the so-called resonant devices as most initial NEMS
applications involve these. The input transducers in resonant NEMS convert electrical
energy into mechanical energy by exciting a resonant mode of the mechanical element. The
mechanical response, namely the displacement of the element, is transduced back into
electrical signals. In the resonant mode of operation, external perturbations can be regarded
as the control signals, since they modify the vibrational characteristics, such as the
resonance frequency v0 /2p or the Q of the vibrating element. We shall discuss
electromechanical transduction mechanisms in NEMS and give examples of measurements
of external perturbations in Sec. IV.
 Figure 2.1

2.2 : Frequency

In Fig. 3, the frequencies experimentally attained for the fundamental flexural modes of
thin beams, for dimensions spanning the domain from MEMS to deep within NEMS.5,9
Continuum mechanics approxi mations appear to hold, i.e., the expression v0 /2p
=s1.05dÎE/rst/l 2 d determines the flexural resonance frequencies of thin doubly clamped
NEMS beams.13 Here, w 3t3l are the dimensions, E is the Young’s modulus, and r

Figure 2.2
FIG. 2.2. sColor onlined. Frequency versus effective geometry for doubly clamped beams
made from single-crystal SiC, Si, and GaAs ssee Refs. 5 and 9d.

2.3: Available dynamic range

The linear dynamic range sDRd is a widely used concept in amplifier characterization,
expressing the window of input power in which the amplifier behaves linearly. The bottom
of the DR is determined by the noise power generated within the amplifier sreferred back to
its inputd, and the top by the s input d power level at which 1 dB compression occurs.
Similarly, we shall attempt to define a DR for a NEMS resonator as the ratio of its
maximum vibration amplitude sat the onset of nonlinearity d to its rms displacement noise
floor within the operation bandwidth Df. In general, a NEMS resonator is always followed
by a transducer-amplifier cascade. In most narrowband operation, Df is determined by this
transducer-amplifier cascade. One might sometimes desire to use the entire resonant
response in which case the measurement bandwidth becomes the natural snoised bandwidth
of the resonator Df The block diagram for a generic NEMS resonator operated in its
fundamental mode and coupled to a noisy transducer-amplifier cascade is shown in Fig. 5.
The mechanical response for the fundamental mode of the resonator can be approximated
by that of a one-dimensional damped harmonic oscillator with the following parameters: an
effective mass Meff; an effective force constant sstiffnessd keff =Meffv0 2 ; and a quality
factor Q. The transfer function Gsvd for the resonator then becomes

(1)

To retain the generality of the discussion, we shall assume that the resonator is driven by a
noiseless drive force at v=v0, a dominant intrinsic noise and by the backaction force of the
transducer–amplifier. The drive has a force spectral density kxdl 2 dsv−v0d/uGsv0du2 .
kxdl is the rms vibration amplitude to which the resonator is driven and kxdl is assumed to
be within the linear regime of the resonator. The power spectral density SF sid svd swith
 units N2 /Hzd of the intrinsic force noise term is determined by the dominant physical noise
mechanism. We shall have more to say about intrinsic noise sources in NEMS in Sec. III B.
SF sbd svd quantifies the force noise resulting from the backaction of the transducer–
amplifier cascade. The backaction force24 represents the reverse coupling and drives the
mechanical element as electrical impulses are generated in the transducer– amplifier
cascade. Apparently, the displacement noise floor, hence, the DR of the NEMS–transducer–
amplifier cascade of Fig. 5 will be determined by the larger of the two distinct classes of
mechanisms—intrinsic noise processes fundamental to the nanomechanical resonator
itself25,26 or extrinsic processes that originate from the transducer–amplifier circuitry.27
We now obtain general expressions for the displacement noise floor due to each process.
The noise in the drive force generates the rms displacement noise,

(2)

Similarly, the voltage noise generated within the transducer– amplifier cascade with swhited
spectral power density SVsvd appears as a displacement noise at the input of the cascade
with an rms magnitude

(3)

Here, R=dv/dx is the cascade responsivity (with units V/m)

2.4 : Generating Movement in Rigid Nanosystems:

For very small structures, both the induction of motion and the detection of motion are
challenges. All of the moving devices discussed in the previous paragraph were actuated by
applied electric fields and observed by optical interference or angular deflection of a laser
beam. This is perhaps the most direct method for actuation and observation of small-scale
motion, and it can be used for a variety of experimental devices. Both static displacements
and resonant motion can be readily actuated in this way. Other methods can be used to
induce and observe nanoscale motion of NEMS. Lorentz forces have been used to drive
small conducting beams (14), with alternating current passing through a conducting wire in
a strong transverse magnetic field to drive motion. The induced electromotive force, or
voltage, can be detected as a measure of the motion. This method requires a fully
conducting path and works well, for example, with a beam clamped at both ends.
Piezoelectric elements and bilayers of differential thermal expansion, mounted on the
moveable elements, have been used to actuate MEMS devices. For resonant systems,
secondary actuation also works well for nanoscale systems, by using piezoelectric or other
actuation to oscillate the supports of resonant NEMS devices. Electron tunneling is a very
sensitive method that can detect subnanometer motion by the exponential dependence of the
electron tunneling current with the separation between tunneling electrodes (19). For
widespread commercial applications, miniaturization and integration of the entire system
will be desirable. A new method has been recently demonstrated, which uses a scanning
tunneling microscope (STM) as an actuator combined with a scanning electron microscope
to detect the motion (20). This approach provides capabilities that are important for
exploring oscillatory mode structure at a scale much finer than that obtainable by laser
Doppler techniques. Electron micrographs of a STM tip and a silicon nitride cantilever are
shown in Fig. 4, left. An ac voltage applied to the piezoelectric axial drive of the tip imparts
a localized drive to the mechanical system. By observing interaction of the focused electron
beam with the moving oscillator, a measure of the motion can be obtained. The STM could
also be used to image surface nanostructure and correlate surface structure with mechanical
response and losses.
 Figure 2.4

Chapter 3:
How to make NEMS

NEMS must overcome a final important hurdle before nanoscale machines, sensors and
electronics emerge from ' industrial production lines. Put simply, when we combine state-
of-the-art processes from two disparate fields - nanolithography and MEMS
micromachining - we increase the chances that something will go awry during
manufacturing. Fortunately, sustained and careful work is beginning to solve these
problems and is revealing the way to build robust, reliable NEMS. Given the remarkable
success of microelectronics, it seems clear that such current troubles will ultimately become
only of historical significance. But there is a special class of difficulties unique to NEMS
that cannot be so easily dismissed. NEMS can respond to masses approaching the level of
single atoms or molecules. However, this sensitivity is a double-edged sword.
 Figure: 3.1

On the one hand it offers major advances in mass spectrometry; but it can also make device
reproducibility troublesome, even elusive. For example, at Caltech we have found that it
places extremely siringent requirements on the cleanliness and precision of nanofabrication
techniques

3.1: Fabrication:

There are three basic building blocks in NEMS technology, which are the ability
to deposit thin films of material on a substrate, to apply a patterned mask on top of the films
by photolithographic imaging, and to etch the films selectively to the mask. A NEMS
process is usually a structured sequence of these operations to form actual devices and
includes:

 Deposition processes
 Lithography
 Etching processes
 3.1.1 : Deposition Processes

One of the basic building blocks in NEMS processing is the ability to deposit
thin films of material. The thin film can have a thickness anywhere between a few
nanometers to about 100 nanometer. Chemical methods are often used in NEMS deposition
technology and major among them are:

1) Chemical Vapour Deposition (CVD)

2) Epitaxy

3.1.1.1 : Chemical Vapour Deposition (CVD)

In this process, the substrate is placed inside a reactor to which a number of gases
are supplied. The fundamental principle of the process is that a chemical reaction takes
place between the source gases. The product of that reaction is a solid material with
condenses on all surfaces inside the reactor. The two most important CVD technologies in
NEMS are the Low Pressure CVD (LPCVD) and Plasma Enhanced CVD (PECVD). The
LPCVD process produces layers with excellent uniformity of thickness and material
characteristics. The main problems with the process are the high deposition temperature
(higher than 600° C) and the relatively slow deposition rate. The PECVD process can
operate at lower temperatures (down to 300° C) thanks to the extra energy supplied to the
gas molecules by the plasma in the reactor. However, the quality of the films tends to be
inferior to processes running at higher temperatures. Secondly, most PECVD deposition
systems can only deposit the material on one side of the wafers on 1 to 4 wafers at a time.
LPCVD systems deposit films on both sides of at least 25 wafers at a time. A schematic
diagram of a typical LPCVD reactor is shown in the figure 1
 Figure 3.1.1.1: Typical hot-wall LPCVD reactor

CVD processes are ideal to use when you want a thin film with good step coverage.
A variety of materials can be deposited with this technology. The quality of the material
varies from process to process, however a good rule of thumb is that higher process
temperature yields a material with higher quality and less defects.

3.1.1.2 : Epitaxy

This technology is quite similar to what happens in CVD processes, however, if the
substrate is an ordered semiconductor crystal (i.e. silicon, gallium arsenide), it is possible
with this process to continue building on the substrate with the same crystallographic
orientation with the substrate acting as a seed for the deposition. If an
amorphous/polycrystalline substrate surface is used, the film will also be amorphous or
polycrystalline.
 There are several technologies for creating the conditions inside a reactor needed to
support epitaxial growth, of which the most important is Vapour Phase Epitaxy (VPE). In
this process, a number of gases are introduced in an induction heated reactor where only the
substrate is heated. The temperature of the substrate typically must be at least 50% of the
melting point of the material to be deposited. An advantage of epitaxy is the high growth
rate of material, which allows the formation of films with considerable thickness (>100µm).
Epitaxy is a widely used technology for producing silicon on insulator (SOI) substrates. The
technology is primarily used for deposition of silicon. A schematic diagram of a typical
vapour phase epitaxial reactor is shown in figure 2.

Figure 3.1.1.2: Typical cold-wall vapour phase epitaxial reactor

This has been and continues to be an emerging process technology in NEMS. Some
processes require high temperature exposure of the substrate, whereas others do not require
significant heating of the substrate. Some processes can even be used to perform selective
deposition, depending on the surface of the substrate.
3.1.2 : Lithography

Lithography in the NEMS context is typically the transfer of a pattern to a

photosensitive material by selective exposure to a radiation source such as light. A
photosensitive material is a material that experiences a change in its physical properties
when exposed to a radiation source. If we selectively expose a photosensitive material to
radiation (e.g. by masking some of the radiation) the pattern of the radiation on the material
is transferred to the material exposed, as the properties of the exposed and unexposed
regions differ.

Figure 3.1.2: Transfer of a pattern to a photosensitive material.

In lithography for micromachining, the photosensitive material used is typically a

photo resist (also called resist, other photosensitive polymers are also used). If the resist is
placed in a developer solution after selective exposure to a light source, it will etch away
one of the two regions (exposed or unexposed). If the exposed material is etched away by
the developer and the unexposed region is resilient, the material is considered to be a
positive resist (shown in figure 4a). If the exposed material is resilient to the developer and
the unexposed region is etched away, it is considered to be a negative resist.
 Figure 3.1.2: a) Pattern definition in positive resist, b) Pattern definition in negative resist

Lithography is the principal mechanism for pattern definition in

micromachining. Photosensitive compounds are primarily organic, and do not encompass
the spectrum of materials properties of interest to nano-machinists. However, as the
technique is capable of producing fine features in an economic fashion, a photosensitive
layer is often used as a temporary mask when etching an underlying layer, so that the
pattern may be transferred to the underlying layer. Photo resist may also be used as a
template for patterning material deposited after lithography.

The resist is subsequently etched away, and the material deposited on the resist is "lifted
off". The deposition template (lift-off) approach for transferring a pattern from resist to
another layer is less common than using the resist pattern as an etch mask. The reason for
this is that resist is incompatible with most NEMS deposition processes, usually because it
cannot withstand high temperatures and may act as a source of contamination.
 CHAPTER 4:
Case study

4.1: NEMS Devices for Accelerometers Compatible with Thin

SOI Technology

A lot of MEMS are fabricated with stand-alone approaches even if an ASIC is generally
provided in order to ensure driving and signal processing. Vertical integration tends to
realize the integration at the substrate level, the interconnection level, the above-IC level or
the module level. Horizontal integration allows to integrate the MEMS and the IC on the
same surface. Monolithic integration of MEMS and microelectronics on thick SOI wafers
has already been reported, the MEMS being manufactured before or after the CMOS.

Thin SOI technology is capable to push forward the possibilities by offering the ability to
provide horizontal integration directly at the front-end level. Thin SOI technology means
thin silicon top layer, for example 160nm thickness if we consider the 0.13 ptm CMOS SOI
technological node. The idea is to use this thin silicon layer, traditionally dedicated to the
CMOS process, for the MEMS fabrication (Fig. 1). The first consequence is that at least one
dimension of the MEMS structure is shrinked to the nanoscale. As a consequence the
mechanical device must be considered as a NEMS (Nano-Electro-Mechanical Systems).

Figure 4.1

NEMS are very promising for ultimate sensing, autonomous sensors, battery-operated
systems and can allow new architectures. Nowadays, NEMS are investigated in order to
face the challenges of nano-world and have a better understanding of the fundamental issues
of mechanics in the nano-scale [1]. In particular, thin SOI technology is a very good
candidate since the same thin layer of single-crystal silicon can be used for both mechanical
structures and advanced ICs.

Modelling and Design:

5OG and 10G accelerometers have been designed in two configurations: In-Plane (IP)
accelerometers with sensitive axis included in the plane of the wafer, and Out-of-Plane
(OOP) accelerometers with sensitive axis perpendicular to the wafer (Fig.2). An
acceleration causes the motion of a proof-mass and the corresponding displacement is
measured by means of a variation of capacitance between two sets of electrodes. The
electrodes are either interdigitated combs (IP) or two parallel plates (OOP).

Figure 4.1.1

If we consider 0.13µm CMOS SOI technology, the first constrain to take into account is the
160nm thickness of the silicon top layer that will be used for the NEMS and MEMS
structures. In order to avoid parasitic deformations and undesirable vibration modes, the
mass size has been limited to typical dimensions of 75xl50µm and 100x100µm, respectively
for IP and OOP accelerometers. As a consequence the mass is about 4ng. In addition, for IP
accelerometers, low cross-sensitivity implies to increase the ratio thickness/width of the
mechanical springs, leading to 50nm wide mechanical springs.
 Figure: 4.1.2

Based on this model, 50G accelerometers have been designed in both IP and OOP
configurations. IP50G accelerometers are constituted of a 75µmxl50µm proof mass,
electrostatic combs with teeth 80nm wide and gaps 130nm to 400nm wide. With 130nm
gap, the mechanical spring is 50nm wide and 4ptm long. With 400nm gap, the mechanical
springs are 50nm wide and 11µm long. OOP accelerometers are constituted of a
100µmx100µm proof mass with an electrostatic gap 400nm wide. The mechanical springs
are 10µm wide and 80µm long.

Figure 4.1.3
Fig.5 and 6 give expected mechanical and electrical performances respectively for IP-50G
and OOP-50G accelerometers.

Expected performances of IP 50G accelerometers (130nm- 400nm gap)

Expected performances of OOP 5OG accelerometers (400nm gap)

IP accelerometers exhibit pull-in voltage from a few 100mV to 1.5V respectively for 130nm
and 400nm gap width. FS capacitance change is 3.5fF down to 0.4fF respectively for
130nm and 400nm gap width. OOP accelerometers exhibit driving voltage about a few
10mV to 100mv and FS capacitance change around a few 1OfF. The sensitivities are
respectively in the ranges of 10-100aF/g and 500-10OOaF/g for IP and OOP
accelerometers. These figures are far from conventional MEMS but can be considered as
acceptable considering the fact that ultra low noise CMOS ICs are capable to resolve better
than laF/sqrt(Hz) of capacitance variation. Moreover NEMS structures developed with "In-
IC" integration approach will benefit of reduced parasitic capacitances allowing to expect to
resolve 10-3aF/sqrt(Hz).
 THIN SOI FABRICATION

The fabrication of NEMS devices and accelerometers (Fig.7) is based on Thin SOI 130nm
technology and 200mm facilities. The Si top layer is 160nm thick and the BOX oxide is
400nm thick.

Simplified flow-chart for fabrication of thin SOI NEMS

Scaling considerations have been addressed in order to realize on the same wafer the nano-
structures (springs, electrostatic combs) and the large structures (proof mass, electrical pads
...). To reach this goal, hybrid lithography combining advanced UHR e-beam and 248nm
DUV lithographies has been used. In addition, FH-vapor technique has been improved to
enable the release and protection against sticking despite the 500nm diameter of the release
holes and the 400nm thickness of the sacrificial oxide. Finally, AlSi metal was used for
electrical pads.
 Chapter 5:
Recent advances in Nano-Electromechanical System and nano-
fluidic pumping

SURFACE TENSION AS A DRIVING FORCE FOR A

NANOELECTROMECHANICAL MOTOR:

Researchers at the University of California have developed a nano-motor, which operates

by transferring atoms between two molten metal droplets in a carbon nanotube (Regan et
al., 2005). The relaxation oscillator consists of two liquid metal droplets along a surface
within a 200 nm enclosure. Figure 1 illustrates the general schematic of the device.
Electrical current passes through the nanotube, which moves the metal atoms from the
larger droplet along the nanotube to the smaller droplet. As a result, the small droplet
expands and eventually touches the larger droplet. This contact generates a rapid release of
energy at a contact relaxation point, when the larger droplet absorbs atoms that were
previously transported through the nanotube. The smaller droplet has higher internal
pressure than the larger droplet, due to higher interfacial curvature, so this higher pressure
drives fluid from the smaller to the larger droplet at the point of physical contact between
the droplets. The recovery process after the droplet contact point is an atomic transport
process, no hydrodynamic. Atoms are transported individually between the droplets through
surface diffusion. The collective surface energy is accumulated gradually until the contact
point, when energy is released suddenly over a brief period.

Figure: 5.1
Regan et al. (2005) implemented a nano-electromechanical relaxation oscillator with droplet
radii of about 30 and 90 nm, respectively. The large droplet’s radius grows to about 91 nm
and 5 fJ of energy is released after the droplet contact. The time scale for coalescence of the
two droplets is about 200 ps, which implies that a device using this nanoenergy source
could operate at frequencies approaching the gigahertz range. The device implemented by
Regan et al generates a peak power of about 20 mW and a peak force of 50 nN. This power
output from oscillating droplets could be used to exert mechanical forces on the
neighbouring objects. Furthermore, numerous oscillators could be constructed in parallel, so
shrinking and expanding droplets would cause the substrate nanotube to flex at a particular
frequency. The resulting deflections could provide high-frequency operation of a
nanoelectromechanical actuator. The process of gradual buildup of surface tension,
followed by rapid release of energy at the contact point is found in many natural
phenomena, such as earthquakes, heartbeats and so forth. Although the peak power output
of 20 mW is tiny, the power density achieved during relaxation is very large. According to
Regan et al. it is about 100 million times more power per unit volume than a 225
horsepower, 3.3 l V-6 car engine. This large power density occurs because the tendency of
atoms or molecules to resist separation (characterized by the surface tension) becomes more
significant at small scales. The nano-device generates power output of mechanical motion
from electricity supplied to the carbon nanotube, so it can be classified as a nano-motor,
rather than a nano-engine. Nano-motors have many promising applications, such as optical
circuits to re-direct light in optical switching technology. Some researchers envision a future
biomedical possibility of using nano-motors to power nano-machines that move within a
human body to find tumours and repair damaged cells.

THERMOCAPILLARY PUMPING AS A DRIVING MECHANISM

FOR A MICRO HEAT ENGINE

The previous nano and micro energy sources are promising technologies for powering
future MEMS and NEMS. This section describes a novel alternative, which provides certain
advantages over past methods. Rather than a fuel source such as propane or gasoline, this
section considers TCP as a driving mechanism to move a discrete droplet within a closed-
ended microchannel and generate electricity. In Figure 4, the micro-droplet would act like a
piston to expand and compress the adjoining gas chambers, thereby flexing a piezoelectric
membrane at the channel ends to generate electricity. TCP is a non-mechanical method of
fluid pumping that takes advantage of temperature-dependent changes of fluid surface
tension. It is driven by internal pressure differences, resulting from surface tension
variations at the ends of a discrete droplet, due to an applied temperature gradient. This
section will focus on a promising embodiment of a novel micro energy source, called a
thermocapillary MHE. A prototype of a MHE was manufactured with microchannels, fluid
access holes, heaters and heater leads at the University of Manitoba, Canada (Glockner and
Naterer.

Figure: 5.2

Consider a droplet enclosed within a closed microchannel (see Figure 4). Heat is transferred
through a thermal bridge across a substrate from a stationary cyclic source, which turns
on/off to control thermocapillary motion of the enclosed micro-droplet. During heat input,
thermocapillary forces induce fluid motion from left to right. Heat transfer to one end of the
droplet leads to temperature variations within the liquid. Thermocapillary pumping arises
when such variations, as well as differences between contact angles at both ends of the
droplet, contribute to an effective pressure difference across the liquid that drives fluid
motion within the microchannel. When the droplet moves from left to right, the air pressure
increases in the right section of the microchannel and decreases in the left section. During
the cooling period when the heat source is turned off, these relative air pressures drive fluid
back towards the initial equilibrium position of the droplet. The cyclic heat source can be
readily connected to several microchannels in parallel to augment power output through a
larger network of microchannels. In Figure 4, the capillary pressure difference across the
droplet can be expressed in the following way:

where Pc,L is the capillary pressure on the left side of the droplet, Pc,R is the capillary
pressure on the right side of the droplet, y is the contact angle between the liquid and solid
wall and H is the channel height. In Equation (1), G is a constant (specific to the problem
geometry). For example, G ¼ 4 for a circular microtube and G ¼ 2 for slit-like and
rectangular microchannels. In multiphase systems with droplets, the droplet is hydrophilic if
the droplet is the wetting fluid, in which case y590: On the other hand, the droplet is
considered hydrophobic if the surrounding fluid is the wetting fluid, for which y>90. The
surface tension at the heated meniscus varies with temperature, T, in the following way:

where A and B are constants (i.e. A=75.83 dyn cm1 and B=0.1477 dyn cm1 K1 for water).
A desirable feature of the working fluid is a high change of surface tension with
temperature, which would generate the largest pressure difference for a given heat input.
For example, water and H2O2 have high coefficients of B (most desirable). But surface
tension must overcome friction within the droplet. When divided by dynamic viscosity,
other fluids (such as toluene) become more effective as a working fluid. The working fluid
should generate a high tensioninduced pressure difference, in excess of a fixed frictional
resistance within the liquid.

RESULTS OF THERMAL CYCLING AND THERMOCAPILLARY PUMPING A finite volume method (FVM)
based on a standard SIMPLE procedure (Patankar, 1980) was used to perform simulations of a free
convection cell to estimate approximate parameters for a cycled MHE. Using the parameters in
Table I, a Rayleigh number of Ra=103 was obtained for
a 3 mm square cell containing water with a temperature difference of DT=2.38C. The
resulting centreline u and v velocity fields (without the presence of a rotating MHE) are
plotted in Figures 8(a) and (b), respectively. For the current problem parameters, the
resulting temperature difference across the MHE was too small to generate sufficient
capillary pressure differences across the droplet. As a result, other alternatives such as a
longer droplet were studied to increase the temperature difference across the droplet.
Additional results indicated that a temperature difference of DT=1.28C across the droplet
will displace the droplet by about 0.31 mm and generate an external pressure difference of
approximately 242 Pa across the droplet. Another alternative is to increase the temperature
difference across the cavity. Maintaining a droplet length of 1.5 mm and increasing the
temperature difference across the cavity to 92.88C yielded a natural convection cell with a
Rayleigh number of 40 000 and a temperature difference across the droplet of about 468C.
This would displace the droplet by approximately 9 mm and generate an external pressure
difference across the droplet of approximately 7 kPa. The cycle time for this configuration
would be about 1 s.
 Chapter 5:
Advantages

 NEMS is a rapidly growing technology for the fabrication of miniature devices

using processes similar to those used in the integrated circuit industry.
 NEMS technology provides a way to integrate mechanical, fluidic, optical, and
electronic functionality on very small devices, ranging from 0.1 nanons to one
millimeter.
 NEMS devices have several important advantages over conventional counterparts.

Cost effectiveness

 Like integrated circuits, they can be fabricated in large numbers, so that cost
of production can be reduced substantially.
 They can be directly incorporated into integrated circuits; so that far
more complicated systems can be made than with other technologies.
 NEMS is an extremely diverse technology that potentially could significantly
impact every category of products. Already, NEMS is used for everything ranging
from neural probes to active suspension systems for automobiles

High Precision

 NEMS-based switches must be extremely reliable to meet the standards and

requirements of optical telecommunications networks – they must remain in
precise position over millions of operations, and they must be designed to meet
stringent environmental specifications involving temperature and vibration.
 However, there is a high degree of confidence that mechanical NEMS devices
can meet these requirements, as similar devices based on the same manufacturing
processes have proven to be exceedingly robust in the automotive, military and
aerospace industries.
 Small size

 NEMS based devices are extremely small in size because of the large scale
integration of the nano electronics and the mechanical systems which
include sensors and actuators.
 NEMS devices can be so small that hundreds of them can fit in the same space
as one single macro-device that performs the same function

Application

Thermal Actuator

Figure 5.1: Thermal Actuator

Thermal actuator is one of the most important NEMS devices, which is able to
deliver a large force with large displacement, thus they have found various applications in
NANO-ELECTROMECHANICAL SYSTEM

electro-optical-communication, micro-assembly and micro-tools. Currently Si-based

materials have been predominantly used to fabricate thermal actuators due to its mature
process and stress-free materials.

Thermal actuators based on metal materials generally have a number of advantages

over Si-based ones due to their large thermal expansion coefficients, thus they can deliver
large displacements and forces and consumes less power, and therefore they are much more
efficient than Si-based ones.

We have developed a single-mask NEMS process based on Si-substrate and

electroplated Ni active materials. Various thermal actuators and their enabled microsystems
have been fabricated and electrically tested.

Biotechnology

NEMS technology is enabling new discoveries in science and engineering such as

the Polymerase Chain Reaction (PCR) nano systems for DNA amplification and
identification, nano machined Scanning Tunneling Nano-scopes (STMs), biochips for
detection of hazardous chemical and biological agents, and nano systems for high-
throughput drug screening and selection.

Accelerometers

NEMS accelerometers are quickly replacing conventional accelerometers for crash

air-bag deployment systems in automobiles. The conventional approach uses several bulky
accelerometers made of discrete components mounted in the front of the car with separate
electronics near the air-bag; this approach costs over $50 per automobile.

31
NANO-ELECTROMECHANICAL SYSTEM

Figure 5.2: Accelerometers

NEMS technology has made it possible to integrate the accelerometer and electronics onto
a single silicon chip at a cost between $5 and $10. These NEMS accelerometers are much
smaller, more functional, lighter, more reliable, and are produced for a fraction of the cost
of the conventional macro scale accelerometer elements

NEMS in Optical Networks

An important new application for NEMS devices is in fiber optic networks. At the
nanons level, NEMS-based switches route light from one fiber to another. Such an approach
enables a truly photonic (completely light-based) network of voice and data traffic, since
switching no longer requires conversion of light signals into digital electronic signals and
then back to optical.This is important because switching using optical-electrical-optical
(OEO) conversion can often cause substantial bottlenecks, preventing the realization of
truly broadband networks. But NEMS and nano machined devices can be used as more than
switches in the optical network. Additional applications include active sources, tunable
filters, variable optical attenuators, and gain equalization and dispersion compensation
devices.

32
 Figure 7: Optical networks

The result is an end-to-end photonic network which is more reliable and cost-effective, and
which has minimal performance drop-off. However the development of an all-optical
network has been complex and challenging due to the integration of optics, mechanics and
electronics

Future outlook

NEMS offer unprecedented and intriguing opportunities for sensing and

fundamental measurements. Both novel applications and fascinating physics will
undoubtedly emerge from this new field, including single-spin magnetic resonance and
phonon counting using mechanical devices.

But there remains a gap between today's NEMS devices that are sculpted from bulk
materials and those that will ultimately be built atom by atom. In the future, complex
molecular-scale mechanical devices will be mass-produced by placing millions of atoms
with exquisite precision or by some form of controlled self-assembly. This will be true
nanotechnology. Nature has already mastered such remarkable feats of atomic assembly,
forming molecular motors and machinery that can transport biochemical within cells or
move entire cells.

Nano Electro Mechanical Devices (NEMS) involve the relative motion of one
interface past a second. The properties of this interface, including its electrical, mechanical
and tribological characteristics, ultimately depend on the arrangement of the atoms.
Recently, we have shown how the alignment of two atomic lattices has dramatic effects on
the friction and dynamics of the objects in contact. Through atomic force microscopy
manipulation, we have shown the carbon nano-tubes show the full range of dynamics
including sliding and rolling. On graphite, the atomic lattices can come into registry, and the
interlocking atoms cause the nano-tube to roll. The atomic lattices also dictate the electronic
states at the interface. We have measured the electrical properties of atomic lattices in
contact and show a change in the contact resistance of over one decade as the lattices move
in and out of registry. The further implications of the mechanical and electrical properties of
contacting lattices in NEMS devices will be explored, including applications in actuators,
encoders and oscillators.

The search for the limits of mechanical motion is a central theme. At low
temperature, quantum friction starts to limit the Q-factor and vibrating NEM-devices are
limited by zero-point motion. This quantum limitation poses an ultimate limit to sensitivity
of NEM-devices. In addition, other quantum phenomena are expected to be present.
Quantum optics-like experiments with phonons, phonon lasers or quantum-tunneling
experiments with massive objects (strained suspended nano-tubes placed between two gate
electrodes) are just a few examples. As the size of NEM-devices shrinks down, electron-
phonon coupling translates into an increasingly strong interplay between electrical and
mechanical degrees of freedom. Device operation results in charge distributions that are
inhomogeneous on the nanometer scale, giving rise to Coulomb forces that are strong
enough to change device geometry. The classical theory of elasticity breaks down and the
regime of quantum elasticity has been entered.

Current projects involve Coulomb blockade and noise properties (quantum

transport) of single-wall nano-tubes, mixing experiments to detect the guitar-like modes of
SWNTs and the fabrication of a SET in the vicinity of a suspended SWNT to detect its
motion. Singly-clamped semi conducting nano-wires are used as switches with the goal to
fabricate nano-mechanical shuttles.
 Conclusion

Nano-systems have the enabling capability and potential similar to those of nano-
processors in the 1970s and software in the 1980s.Since NEMS is a nascent and synergistic
technology, many new applications will emerge, expanding the markets beyond that which
is currently identified or known. As breakthrough technology allowing unparalleled synergy
between hitherto unrelated fields of endeavor such as biology and nano-electronics, NEMS
is forecasted to have growth similar to its parent IC technology. For a great many
applications, NEMS is sure to be the technology of the future.

The automotive industry, motivated by the need for more efficient safety systems
and the desire for enhanced performance, is the largest consumer of NEMS-based
technology. In addition to accelerometers and gyroscopes, nano-sized tire pressure systems
are now standard issues in new vehicles, putting NEMS pressure sensors in high demand.
Such nano -sized pressure sensors can be used by physicians and surgeons in a telemetry
system to measure blood pressure at a stet, allowing early detection of hypertension and
restenosis. Alternatively, the detection of bio molecules can benefit most from NEMS-based
biosensors. Medical applications include the detection of DNA sequences and metabolites.
MEMS biosensors can also monitor several chemicals simultaneously, making them perfect
for detecting toxins in the environment.
 References

1. Mohamed Gad-el-Hak, ed., The MEMS Handbook, CRC Press 2001, ISBN 0-8493-
0077-0
2. P. Rai-Choudhury, ed., Handbook of Microlithography, Micromachining, and
Microfabrication, Vol 1 and Vol 2, SPIE Press and IEE Press 1997, ISBN 0-8529-
6906-6 (Vol 1) and 0-8529-6911-2 (Vol 2)
3. Julian W. Gardner, and Vijay K. Varadan, and Osama O. Awadelkarim,
Microsensors, MEMS and Smart Devices, Wiley 2001, ISBN 0-4718-6109-X
4. Nadim Maluf, An Introduction to Micro-electro-mechanical Systems Engineering,
Artech House 1999, ISBN 0-8900-6581-0
5. A.N. Cleland and M.L. Roukes, "Fabrication of high frequency nanometer scale
mechanical resonators from bulk Si substrates", Appl. Phys. Lett., 69, 2653
(1996).
6. M.L. Roukes, “Yoctocalorimetry: Phonon Counting in Nanostructures”, Physica
B: Condensed Matter 263-264, 1 (1999).
7. Y.T. Yang, K.L. Ekinci, X.M.H. Huang, L.M. Schiavone, C. Zorman, M.
Mehregany, and M.L. Roukes, “Monocrystalline Silicon Carbide NEMS”, to
be published.
8. See, e.g., K. Visscher, M.J. Schnitzer, and S.M. Block, “Kinesin motors studied
with an optical force clamp”, Biophysical Journ. 74, A49 (1998).
9. J.C. Meiners and S.R. Quake, “A direct measurement of the
hydrodynamic interaction between two particles”, Phys. Rev. Lett. 82,
2211 (1999).
10. M.L. Roukes and M.C. Cross; “BioNEMS: nanomechanical devices in solution.”,
to be published.
11. D.R. Baselt, G.U. Lee, K.M. Hansen, L.A. Chrisey, and R.J. Colton, “A high-
sensitivity micromachined biosensor”, Proc. IEEE 85, 672-680 (1997).

http://www.physicsweb.org

http://www.nemsnet.org

http://www.nemsrus.com