M.Tech Thesis

“NANOCRYSTALLINE POROUS SILICON BASED
INTEGRATED MEMS PRESSURE AND

TEMPERATURE SENSOR”
By
Mr. Ishita Sinha
Roll No –
Reg. No -
Under the guidance of

Prof. H. Saha
Thesis submitted in partial fulfillment

Of the requirements for the degree of
Master of Technology in Nanoscience and Technology
May 2007
IC Design and Fabrication Centre

Department of Electronics & Telecommunication Engineering
Jadavpur University
Kolkata – 700032, India.
1
ACKNOWLEDGEMENT
I would like to take this opportunity to express my gratitude and heartiest
thanks to Prof. H. Saha, IC Design and Fabrication Center, Department of Electronics
& Telecommunication Engineering, Jadavpur University, for his inspiration,
encouragement and guidance in the fulfillment of my project work.
I convey my sincere thanks to the Head of the department of Nano Science &
Technology Prof.S.Mukherjee for kindly assigning my M.Tech project work in the I.C
Centre. I am also thankful to the faculty members, technical and non technical staffs of
the same department of their help when ever I needed it.
I am also grateful to Dr. U.Ganguly and Mr. Ashok Mondal for their immense
help during the experiment. My sincere thanks also goes to Mrs.Chirasree Pramanick
for her constant help throughout the progress of my work.
I am grateful to Prof.S.Basu, Ex Professor, I.I.T Kharagpur and at present
Research Advisor I.C Design & Fabrication Centre, Jadavpur University, Kolkata, for
necessary suggestions and helpful scientific discussions to improve the quality of my
project work.
I would also like to pay my gratitude to Mrs. Joyita Das, Mr.Palash Basu,
Mr. Biplob Mondal, and my all my friends of IC Center, Department of electronics and
Telecommunication Engg, Jadavpur University, Kolkata.
Last but certainly not the least, I would also pay my heartiest thanks to my
parents, my husband , my respected teachers and my friends and colleagues for their
encouragement and support at different level of my work.
2
CONTENTS
Chapter : 1 Introduction
1.1 General Introduction
1.2 Objective of this project
1.3 Literature Review
1.4 Scope of this work
References
Chapter : 2 Overview of micromachining and micromachined pressure

and temperature sensor
2.1 Introduction
2.2 Overview of Micromechanical Structure
2.2.1 Micromachining
2.2.2 Anisotropic Etching of Silicon
2.2.3 Top view of the micromachined silicon membrane
References
Chapter : 3 Fabrication of the sensor

3.1 Introduction
3.2 Fabrication steps
3.2.1 Cleaning
3.2.2 Oxidation
3.2.3 Photolithography
3.2.5 Preparation of Porous Silicon
3
3.3 Flowchart For Fabrication of Silicon MEMS
Piezoresistive Pressure Sensor
3.4 Flowchart For Fabrication of porous Silicon Pressure Sensor
3.5 Flowchart for fabrication of nanocrystalline porous silicon based
pressure and temperature sensor
References
Chapter: 4 Design and simulation of the sensor
4.1 Design of nanocrystalline porous silicon based integrated pressure
and temperature sensor
4.2 Sensing Principles
4.2.1 Pressure sensing principles
4.2.2 Temperature sensing principles
4.2.3 Integrated temperature and pressure sensor
4.3 Equivalent model of the sensor
4.4 Design of Conventional Pressure sensor using lateral contacts
4.5 Results and Discussions
References
Chapter : 5 Summary and Conclusion
5.1 Summary
5.2 Conclusion
4
Chapte1: Introduction
1.1 General Introduction:
Micro Electro Mechanical systems (MEMS) are integrated micro devices or systems
combining electrical and mechanical components fabricated using Integrated Circuit
(IC) compatible batch-processing techniques and range in size from micrometers to
millimeters. These systems can sense, control and actuate on the micro scale and
function individually or in arrays to generate effects on the macro scale.
Using similar fabrication techniques as building microprocessors, we are now able to
build sensors and actuators on the same microscopic level with the processor chip.
Measured in microns, thermal sensors, pressure sensors, inertial sensors, flow and
viscosity sensors, resonators, levers, gears, transmission systems, micro-mirrors,
valves, pumps, motors, can be batch produced together on the same chip with the
processing unit. They indeed compose a “system on a chip”[1].
Monocrystalline silicon MEMS pressure sensor suffers from high temperature
sensitivity while polysilicon MEMS suffer from lower piezosensitivity .To overcome
both these limitations, the idea of porous silicon based MEMS pressure sensor has
been conceived[2]. The extremely large surface-to-volume ratio (>=500m2/cm3) of a
porous silicon nanostructure, the ease of its formation and control of its surface
morphology through variation of the formation parameters, its compatibility of silicon
IC technology and hence amenability to the development of smart systems-on-chip
sensors—all these properties have made it a very attractive sensing material[3,4].
5
Chapter 1: Introduction
1.2 Objective of the Work :
There are two objectives of the present work :
• Design and simulation of nanocrystalline porous silicon based MEMS
integrated pressure and temperature sensor by ANSYS 10.0
• Fabrication of nanocrystalline porous silicon based MEMS integrated
pressure and temperature sensor.
1.3 Literature Review :
Following the invention of bipolar transistor in 1947, a great deal of effort was put into
characterizing the properties of single crystal semiconductors .In 1954, smith reported
the piezoresistive effect of silicon ands germanium, which is a change of resistance
with applied stress[5].This discovery enabled production of Semiconductor - based
sensors.
Silicon strain gauge, metal diaphragm sensors were first introduced commercially in
1958[2]. Piezoresistive pressure sensors have piezoresistors mounted on or in a
diaphragm. For thin diaphragms and small deflections, the resistance change is linear
with applied pressure[7]. In these early sensors high-cost, low-volume biomedical and
aerospace applications were targeted[11]. The fabrication of early pressure sensors
started with metal diaphragm sensors with bonded silicon strain gauges. The strain
gauges were bonded by epoxies, phenolics, or eutectics [8]. These first designs had
6
low yield and poor stability due to such factors as thermal mismatch with the metal–
epoxy–silicon interface[9].
Metal diaphragms were quickly superseded by single crystal diaphragms with diffused
piezoresistors. These new types of sensor had many advantages related to the
properties of silicon and the availability of high-quality silicon substrates. Hysteresis
and creep associated with metal diaphragms were eliminated. At low temperatures
(<500° C), silicon is perfectly elastic and will not plastically deform [9], but instead will
fracture in a brittle manner. Silicon obeys Hooke’s law up to 1% strain, a tenfold
increase over common metal alloys . Also, the ultimate tensile strength of silicon can
be three times higher than that of stainless steel wire [10]. As a piezoresistive material,
silicon has gauge factors that are over an order of magnitude higher than those of
metal alloys [5].
Some of the first silicon diaphragms were created by mechanical milling spark
machining followed by wet chemical isotropic etching, to create a cup shape. These
diaphragms were bonded to silicon supports by a gold–silicon eutectic (Teutectic = 370 °
C). While this technique of fabrication had the advantages of increased sensitivity and
reduced size, cost was still high, and diaphragms were created one at a time, rather
than in batch mode[2].
7
In 1960 the invention of the planar batch-fabrication process tremendously improved
the reliability and cost of semiconductor devices. In addition, the planar process
allowed for the integration of multiple semiconductor devices onto a single piece of
silicon (i.e., monolithic integration)[2]. This invention heralded the beginning of the IC
industry. Although the early planar process produced relatively large devices (>mm), it
was a tremendously scaleable process that could micromachine an increasing
number of devices. Micromachined pressure sensors were available in 1963 [7] and
advances in fabrication technology have led to the bulk-micromachined sensors
available today.
By the late 1960s and early 1970s, three key technologies were being developed:
anisotropic chemical etching of silicon , ion implantation, and anodic bonding [13,2].
Ion implantation was used to place strain gauges in single-crystal silicon diaphragms.
Ion implantation is generally better than diffusion for doping because both the doping
concentration and doping uniformity are more tightly controlled. Anisotropic etching
improved the diaphragm fabrication process in a number of ways[6]:
(i) diaphragm sizes and locations were now well controlled by IC photolithography
techniques;
(ii) strain gauge placements were improved;
(iii) anisotropic etching was well suited to batch fabrication, allowing hundreds of
diaphragms to be created simultaneously; and
(iv) overall size was decreased further.
Anodic bonding, which uses voltages (500–1500 V) and heat (400–600 C), was used
to bond finished silicon diaphragm wafers to Pyrex glass supports[19]. Several types
8
of glass formulations designed to reduce thermal mismatch to silicon were used.
Anisotropic etching and anodic bonding are batch techniques, and hence hundreds (or
more) of pressure sensors could be manufactured simultaneously
on a single wafer. This amounted to a significant cost reduction.
By 1970s and 1980s, MEMS commercialization was started by several companies
(e.g., IC Transducers, Foxboro ICT, Transensory Devices, IC Sensors and
Novasensor) that produced parts for the automotive industry. The microsensor
companies began to move toward higher-volume, lower cost applications , specifically,
the automotive industry [16].
The mechanical properties of single-crystal silicon are excellent, as reported in a
landmark article . It has high strength, high stiffness, high mechanical repeatability,
high Q, and no mechanical hysteresis. Furthermore, single-crystal silicon is available
in large quantities with high purity and low defect densities[10].Piezoresistive gauge
factors in silicon are higher than in metal, but temperature coefficients of resistance
(TCRs) are high. Because of high TCRs, silicon microsensors often require
temperature compensation techniques. He discussed the development of many
micromechanical devices and has been instrumental in increasing the awareness of
the possibilities that MEMS has to offer[20] .
In 1984 Howe and Muller at the University of California, Berkeley (UCB) developed the
polysilicon surface micromachining process and used it to produce MEMS with
integrated circuits [15]. This technology has served as the basis for many MEMS
products[1].
9
The direct bonding method was first reported in 1985[23]. This method was first used
for making silicon-on-insulator (SOI) material, but was quickly applied to
micromachined devices [22]. Also, surface-micromachined devices have been
reported, which have silicon nitride [24] or polysilicon [25]
diaphragms. These sensors decrease required die size and may simplify integration
with electronics, but at the cost of reduced sensitivity and reproducibility of mechanical
properties[16,21].
In 1989 Researchers at UCB and MIT independently developed the first
electrostatically controlled micromotors that used rotating bearing surfaces [14].
Although no commercial product presently uses this micromotor technology, it served
as a valuable technology driver for the field of MEMS[1].
By 1990s, A tremendous increase in the number of devices, technologies, and
applications (too many to mention individually) has greatly expanded the sphere of
influence of MEMS—and it continues today[1].
So this period is often called the micromachining period [2], since diaphragm
dimensions are shrinking to hundreds of micrometres and minimum feature sizes are
shrinking to micrometres. Also, anisotropic etching and bonding technologies are
being improved[2].
In 1991 Microhinges developed at UCB by Pister extended the surface micromachined
polysilicon process so that large structures could be assembled out of the plane of the
substrate, finally giving MEMS significant access to the third dimension[1].
10
During the last decade the interest of researchers to silicon, which was considered
before as quite well-known material, has grown enormously. The triggering point was
the paper by Dr Leigh Canham (Defence Research and Evaluation Agency, UK) who
published the observation of bright red photoluminescence from the surface of
electrochemically etched Si wafer [12]. Porous silicon a substance which is produced
by a treatment of Silicon wafers in hydrofluoric acid solutions was known since the
fifties due to the works by Uhrlir, Turner , Memming and Schwandt [17]. The material
was considered as suitable for electronic applications (local insulation, gettering of
impurities, sacrificial layers, etc.) but never in relation with optical applications. Energy
gap of silicon (1.1 eV) corresponds to the infrared region and is indirect that makes
radiative recombination processes quite ineffective[3].
The observation of red bright photoluminescence from PS has produced a sort of
sensation (although the first publication on the visible light emission from porous
silicon was made in 1984 by Pickering et al. pose of building Si-based Light-Emitting
Devices (LEDs). In 1990, Canham [18] reported that if porous silicon is further etched
in HF for hours after preparation, it emits bright red light when illuminated with blue or
UV light. Efforts of scientific community undertaken during the years 1991 to 1996
brought many useful results about the aspects of PS formation and its physical and
chemical properties(19).
11
1.3 Scope of the Project Work:
The latest piezoresistive pressure sensors are up to ten times more sensitive than
old transducers and with response times as rapid as a millisecond. These sensors are
used for applications as automotive, hi-fi, aerospace , robotics , space and
medical equipment industries. More than a dozen applications for pressure
sensors have been identified and silicon thin-diaphragm piezoresistive sensors are
responsible for many of these system needs. These sensors are used in vehicles to
control the ignition and the composition of the petrol mixture, in audio systems
to compensate for loudspeaker resonance and in medical for dialysis, middle ear
diagnosis, and disposable blood pressure meters[25]. But these sensors also show
temperature sensitivity. To compensate for the temperature effect, a p–n-junction-
based temperature sensor is often integrated with the silicon pressure sensor for
commercial applications[24].
12
References
[1] Microelectromechanical systems (MEMS): fabrication, design and applications

Jack W Judy
[2] Micromachined pressure sensors: review and recent developments W P

Eatony and J H Smith, Smart Mater. Struct. 6 (1997) 530–539.
[3] Porous Silicon-Based Sensors: Prospects and Challenges H. Saha and C

Pramanik Materials and Manufacturing Processes, 21: 239–246, 2006
[4] Porous Silicon as Pressure Sensing Material U. Gangopadhyay

Journal of the Korean Physical Society, Vol. 47, November 2005, pp.
S450_S453
[5] Piezoresistance effect in germanium and silicon, C.S. Smith, Physical

Review,94(1), pp. 42-49 (1954).
[6] Anisotropic etching of silicon IEEE Trans. Bean K E 1978 Electron Devices
ED-25 1185–93
[7] Pressure sensitivity in anisotropically etched Thin diaphragm pressure

sensors Clark S K and Wise K D 1979 IEEE Trans. Electron Devices ED-26
1887–96 [23]
[8] Piezoresistivity of silicon quantum well wire C Pramanik, S Banerjee, H Saha

and C K Sarkar
[9] Micro-diaphragm pressure sensor S. Sugiyama, T. Suzuki, K. Kawahata,

Shimaoka M. Takigawa and I. Igarashi Toyota Central Research And
Development Labs., Inc.
[10] Silicon as a Mechanical Material Kurt e. Petersen, member, IEEE

proceedings of the IEEE. vol. 70, no. 5, may 1982
[11] Development of a Miniature Pressure Transducer for Biomedical Applications

IEEE transactions on electron devices, vol. ed-26, no. 12, december 1979
13
[12] Porous silicon : mechanisms of growth and applications V. Parkhutik

Solid-State Electronics 43 (1999) 1121±1141
[13] A Novel Convex Corner Compensation for Wet Anisotropic Etching on

(100) Silicon Wafer Huai-Yuan Chu and Weileun Fang
[14] Materials issues in microelectromechanical systems (MEMS) S.M. Spearing

Acta mater. 48 (2000) 179±196
[15] Surface Micromachining for Microelectromechanical Systems James m.

Bustillo, Roger T. Howe, fellow, IEEE, and Richard s. Muller, life fellow, IEEE
[16] High-temperature Pressure and Temperature Multi-function Sensors

Liu Xiaoviei, Wang Wei, Wang XiIian Liu Yuqiang. Liu Zhenmao
[17] Electrode design and planar uniformity of anodically etched large area
porous silicon S M Hossain, J Das, S Chakraborty, S K Dutta and H Saha
[18] Light-Emitting Porous Silicon: Material science, Properties, and Device

Applications P. M. Fauchet, Senior Member, IEEE , L. Tsybeskov, C. Peng,
S. P. Duttagupta, J. von Behren, Y. Kostoulas, J. M. V. Vandyshev, and
K. D. Hirschman
[19] A Microsystem Based on Porous Silicon-Glass Anodic Bonding for Gas

and Liquid optical Sensing Luca De Stefano , Krzysztof Malecki ,
Francesco G. Della Corte , Luigi Moretti , Ilaria Rea , Lucia Rotiroti and
Ivo Rendina Sensors 2006, 6, 680-687
[20] Micro-diaphragm pressure sensor Sugiyama S, Suzuki T, Kawahata K,

Shimaoka K, Takigawa M and Igarashi I 1986 Tech. Digest 1986 Int.
Electron Devices Meeting (IEDM ’86) pp 184–7
[21] Fabrication techniques for integrated sensor Guckel H and Burns D W 1986
microstructures Tech. Digest IEEE Int. pp 122–5
[22] Surface micromachined micro-diaphragm pressure sensors Sugiyama S,

Shimaoka K and Tabata O 1991 Digest Tech. Papers 1991 Int. Conf. on
Solid-State Sensors and Actuators (Transducers ’91) pp 188–91
14
[23] Silicon –on- Insulator(SOI) by Bonding and Etch-Back, J.B.Lasky, S.R.Stiffler,

F.R.White and J.R.Abernathy, Technical Digest, IEEE International Electron
Device meeting, IEDM 85, pp. 684-687 (1985).
[24] Silicon fusion Bonding for Pressure sensors, K.E.Peterson, P.W.Barth, J.

Poydock, J.Mallon and J.Bryzek, Technical Digest: IEEE Solid state Sensor
and Actuator Workshop, Hilton Head 88, p. 144 (1988).
[25] An integrated pressure and temperature sensor based on nanocrystalline

Porous silicon C Pramanik, H Saha and U Gangopadhyay, J. Micromech.
Microeng.16 (2006) 1340–1348
[26] A Silicon Piezoresistive Pressure Sensor Ranjit Singh, Low Lee Ngo,Ho Soon
Seng, Frederick Neo Chwee Mok
15
Chapter 2: Overview of Silicon Micromachining
2.1 Introduction
Anisotropic etching of silicon is used to realize the micromechanical structures.
The pressure sensors can be broadly classified as piezoresistive and capacitive. In
piezoresistive sensors, the resistivity and the geometrical dimensions of the sensor
changes on application of stress which in turn affects the resistance. In case of
capacitive sensors, on application of stress the deflection of the diaphragm affects the
distance between the metal plates of a parallel plate capacitor, which in turn changes
the capacitance of the sensor.
2.2 Overview of Micromechanical Structures
Development of silicon microsensors often required the fabrication of
micromechanical parts. These micromechanical parts were fabricated by etching
areas of the silicon substrate away selectively to leave behind the desired geometries.
This is micromachining, used to designate the mechanical purpose of the fabrication
process, which were used to form these micromechanical parts. Silicon
micromachining is of great importance for the development of inexpensive, batch
fabricated, high performance sensors, which can easily be interfaced with
microprocessors.
The following properties of silicon have made micro machining feasible [1]:
(a) Silicon can be readily oxidized by exposing it to steam or dry oxygen. It allows
silicon wafer to be masked selectively during chemical etching.
16
(b) Single crystal silicon is brittle and can be cleaved like diamond but it is harder
than most metals. It is resistant to mechanical stresses and the elastic limit of
silicon is greater than that of steel.
(c) A single crystal silicon can withstand repeated cycles of compressive and
tensile stresses.
(d) The crystal orientation of single crystal silicon wafer decides the rate of chemical
etching in certain etching solutions, which is important in creating various structures.
Bulk micromachining and surface micromachining are the two distinctly different
approaches of micromachining silicon for realizing microsensors and actuators.
Isotropic and anisotropic etching of silicon has been used for realizing micro
mechanical parts from bulk silicon wafer and forms the basis of “bulk
micromachining”. In another approach for micro machining called “surface micro
machining”, the silicon substrate is primarily used as a mechanical support upon which
the micro mechanical elements are fabricated. The bulk of the silicon wafer itself is not
etched in surface micro machining. There are no holes through the wafer and no
cavities on the backside. When bulk micro machining silicon, the backside of the wafer
is conventionally protected against an etchant with an oxide or nitride layer in which
windows are opened where the micro mechanical structures are to emerge. An
accurate alignment of the etch windows is essential to obtain the structures at a
proper position with respect to the photolithographic patterns at the front. In surface
17
micro machining a sacrificial layer is deposited on the silicon substrate, which may be
coated first with an isolation layer. Windows are opened in the sacrificial layer and the
micro structural thin film is deposited and etched. Selective etching of the sacrificial
layer leaves a freestanding micro mechanical structure[2].
There are two main methods of etching – wet etching and dry etching. Wet etching is
done with the use of chemicals. A batch of wafers is dipped into a highly concentrated
pool of acid and the exposed areas of the wafer are etched away. Dry etching refers to
any of the methods of etching that use gas instead of chemical etchants. Bulk
micromaching of silicon uses wet and dry etching techniques in conjunction with etch
masks and etch stops to sculpt micromechanical devices from silicon substrate. The
selective etching of silicon can be carried by using isotropic and anisotropic etchant.
The isotropic etchant under-etch large area in lateral direction than the area defined
by mask opening. On the other hand, anisotropic etchant, which are also known as
orientation dependent or crystallographic etchant, etch the silicon surface at different
rates in different directions in the crystal lattice. They can form well-defined shapes
with sharp edges and corners[3].
2.2.2 Anisotropic etching of silicon

One common MEMS (Micro-Electro Mechanical Systems) fabrication
technique is the anisotropic etching of crystalline silicon, where etch rate is a function
of orientation. The anisotropic etching of silicon is ubiquitous process in
micromachining. Complex microsystems can be generated using the anisotropic
properties of single crystal silicon in an orientation dependent dissolution reaction. V-
groove structures for example, useful for the passive alignment of optoelectronic
18
devices are easily fabricated using an anisotropic etchant like KOH or
tetramethylammonium hydroxide[4].
Modern exacting demands in this rapidly growing industry require fundamental
understanding of these processes in order to achieve a well defined anisotropic ratio
nada good surface finish. Mostly used technology for bulk structuring for microsensors
and actuators is the anisotropic etching with KOH. Specifications for the etched
structures (such as high etch rate ratios of <110> and <100> to <111> planes, short
etch times and minimum roughness) can be obtained by optimization of the etch
parameters. For sensor applications <100> oriented silicon is mostly used[4].
Anisotropic etchants of silicon such as EDP, KOH and hydrazine are orientation
dependent. This means that they etch the different crystal orientations with different
etch rates. Anisotropic etchants of silicon etch the <100> and <110> crystal planes
significantly faster than the <111> crystal planes. The etch rate for <110> surfaces lies
between those for <100> and <111> surfaces. Figure 1 demonstrates the basic
concepts of bulk micromachining by anisotropic etching of a <100> silicon substrate.
For example for a <100>-silicon substrate etching proceeds along the <100> planes
while it is practically stopped along <11> planes. Since the <111> planes make a
54.75Ο angle with the <100> planes, the slanted walls (Fig.1.1) result. Due to the
slanted <111> planes, the size of the etch mask opening determines the final etch
result. If the etch masks openings are rectangular or square and the sides are aligned
with the <110> direction, no undercutting of the etch mask feature takes place[3].
The width of the bottom surface Wb is given by [4],
19
Wb = Wo – 2 t cot(54.75Ο ) (1)
where Wo is the width of the etch mask window on the wafer back surface, and t is the
etched depth. If <110> oriented silicon is etched in KOH water etchant, essentially
straight walled grooves with sides of <111> planes can be formed. The anisotropic
etch rate in the <100> direction of monocrystalline silicon of <100> oriented wafers
was investigated focusing on the dependence of temperature and concentration of
potassium hydroxide. Advantages and disadvantages of the different etching
conditions are anisotropic direction selectivity, speed and surface roughness of the
pattern. In isotropic etching all orientations or planes etch at the same rate, hence a
square hole would get rounded corners. In anisotropic etching, because of the
differences in rates [5], some planes grow while others disappear. There are two main
classifications that describe how the initial mask shape will evolve into the final etched
shape. Firstly, etched shapes may be classified as either pegs or holes. Holes are
lower than the surface of the wafer and pegs are higher than the wafer, Holes enlarge
with time while pegs shrink.
After long times, holes are dominated by slow planes, while pegs become dominated
by fast planes. Secondly, within a shape (be it a peg or a hole) there can be two
corners: convex (peg like) and concave (hole like). While etching convex corners, fast
planes dominate: fast planes increase in length while slow planes decrease in length.
For concave corners, slow planes dominate[5].
20
Figure 1(a) : Anisotrophic Etching of Silicon
Figure 1(b) : Isotropic etching of silicon
21
Etchants can be characterized by using the following characteristics:
a) Direction dependency (isotropic or anisotropic)
b) Etch rate
c) Anisotropic etch rate ratio (only for anisotropic etchants, 1:1 to 400:1 for
<100>/<111> planes
d) Dopant dependence/selectivity
e) Temperature of etching (20ΟC to 100ΟC)
Direction dependency
The most important feature in classifying silicon etchants is their ability to have
different etch rates in different directions of the crystal lattice that is exposed to them.
Isotropic etchants etch in all directions with the same etch rate, resulting in rather
rounded shaped pits and also previous sharp edges and corners. The result of
anisotropic etchants, on the other hand, is different, looking perpendicular on each of
the crystal planes. This makes it possible to fabricate sharply formed structures or
narrow gaps, whose borders have to coincide with the crystal planes. Depending on
what kind of structure is desired, the proper etchant has to be chosen.
Etch rate
Basically the etch rate can vary with temperature, mix of ingredients, sometimes
optical circumstances (light intensity), or it can be stable over a wide range [9].
Anisotropic etch rate ratio
22
If an anisotropic etchant is chosen, the ratio of etch rates concerning the different
crystal planes, can vary in a wide range. Again, the desired result influences the
choice of etchant, since ratios from 1:1 to 400:1 are possible, if one compare the
<100> and the <111> planes.
Dopant dependence (selectivity)
Another very important attribute is the dopant dependency of etchants. Some etchants
are very selective on the material that they are exposed to, so that a doped layer or a
layer of different material can be used as etch stop or a direction of a much higher
etch rate. If this is not desired, it is advantageous to choose a non-selective
etchant[10].
Etching temperature
In general lower temperatures are better than higher ones, since temperature induced
stress concentrations are minimized when the processing temperature is low. In
addition, the occurrence of hazardous gases is lower at low temperatures.
The etching process carried out in this project is by using EDP which has following
properties :-
EDP has three properties, which make it indispensable for micromachining:
a) It is anisotropic.
b) It is highly selective.
c) It is dopant dependent (stops on highly boron doped layers).
23
Other than EDP , the following etchants can be used for micro machining :
KOH and water
The main advantage of KOH is [4-7], that it is orientation dependent with much higher
<110>: <111> etch rate ratio than EDP, therefore useful for groove etching on <110>
wafer.
HNA
The HNA system is highly variable in its etching rates and characteristics depending
on
a) Silicon dopant concentration
b) Mix ratios
c) Degree of etchant agitation
A major disadvantage is that SiO2 is etched somewhat for all mixtures, so that it can
only be used for short etching times, otherwise, Si3N4 can be used instead.
TMAH
Tetra-Methyl ammonium hydroxide (TMAH) is an anisotropic silicon etchant that is
gaining considerable use in silicon micromachining due to its excellent silicon etch
rate, etch selectivity to masking layers, degree of anisotropy, and relatively low toxicity.
24
2.2.3 TOP VIEW OF MICROMACHINED SILICON MEMBRANE
Figure – 2(a)
25
Figure – 2(b)
26
References
[1] Silicon as a mechanical material ”, K.E.Peterson, IEEE Proc. Vol. 70,1982.
[2] Comparison of Bulk- and Surface- Micromachined Pressure Sensors William

P. Eaton, James H. Smith, David J. Monk, Gary O’Brien, and Todd F. Miller
Micromachined Devices and Components, Proc SPIE, Vol 3514, p. 431
[3] Foundation of MEMS Chang Liu Prentice Hall Chapter-10,11
[4] Anisotropic etching of silicon, K.E.Bean, IEEE Trans. ED-25,10,p.1185,1978.
[5] A Novel Convex Corner Compensation for Wet Anisotropic Etching on (100)
Silicon Wafer Huai-Yuan Chu and Weileun
[6] Anisotropic etching of silicon, D.B.Lee, J. Appl.Phys. Vol. 40,No.1,p.4569,

1969.
[7] Surface Micromachining for Microelectromechanical Systems James M.

Bustillo, Roger T. Howe, fellow, IEEE, and Richard S. Muller,
life fellow, IEEE
[8] A new theory for the anisotropic etching of silicon and some underdeveloped
chemical micromachining concepts”, D.L.Kendall, Journal of Vacuum
Science Technology A, 8(4), pp. 3598-3604 (1990).
[9] Silicon Carbide as a new MEMS technology Pasqualina M. Sarro
[10] Methods for the Fabrication of Convex Corners in Anisotropic Etching of

(100) Silicon in Aqueous KOH”, H.L.Offerins, K.Kuhl nad H.Sandmaier,
Sensors and Actuators A,25-27, pp. 9-13 (1991).
27
Chapter 3 : Fabrication
Of The Sensor
3.1 Introduction :
3.1.1 Resistivity of the silicon wafer

We measured the resistivity with the 4-probe resistivity meter and measured
the average resistivity of the wafer. We placed the wafer at 5 different positions (a)
Lower part (b) Upper part (c) middle part (d) Right part & (e) Left part of the wafer and
measured the resistivity by passing a known value of current (µA) & noting down the
voltages at the different positions of the wafer.
Using the formula: Rs = f * (v)/(I).
where "f' is the Scaling factor=4.53, Rs is the Sheet resistance.
3.1.2 Type test of the silicon wafer

We measured the type of the silicon wafer by the method of Hot Probe test. In
this method, we warmed one terminal of the probe and placed the terminals on the
wafer. If the wafer is of P- type, then a negative voltage will be displayed in the meter
and if the wafer is of N -type then a positive voltage will be displayed in the meter. In
our case the reading in the meter was negative. Hence we came to conclusion that the
wafer is of P type.
28
Of The Sensor
3.2 Fabrication Steps:
3.2.1 Cleaning
Step1: Sample + Acetone. Then boil for 3min. And then treat in Ultrasonic cleaner for
3min. to remove dust particles,oil and greese.
Step2: Dip in Methanol to avoid Oxidation of the sample.
Step3: (1:1) conc. H2SO4 + H202. Wait till the completion of reaction. Then clean in DI-
water to remove the oxide of the sample.
Step4: Dip in 10% HF solution. Then clean in DI-water to remove the oxide of the
sample.
Step5: Treat the sample with Std. Cleaning-1 [H20 (5):H202(1):NH40H(1)] at a temp- of
70°C for 10min. And then pass through cold water to remove the residual organic
material.
Step6: Treat the sample with Std. Cleaning-2 [H2O (6): H2O2(1):HCI(1)} at 700C for
10min. Then pass through DI-water to remove the residual organic material and dry
with N2.
29
Chapter 3 : Fabrication Of The Sensor
3.2 .2 Oxidation
Oxidation is a fundamental process in all silicon device fabrication. Oxidation of silicon
wafers is used for: (i) passivation of the silicon surface (i.e., the formation of a
chemically and electronically stable surface), (ii) masking of diffusion and ion
implantation, (iii) dielectric films, and (iv) an interface layer between the substrate and
other materials, such as in chemical and biosensors. Silicon, when exposed to the air
at room temperature, will grow a native oxide (about 20 A thick). Thicker oxides can be
grown at elevated temperatures in dry or wet oxygen environments. At a given
temperature, the relationship between the thickness of oxide and time is parabolic.
The rate of growth is also affected by the partial pressure of the oxygen atmosphere
and the crystal orientation.
The ability to grow a chemically stable protective layer of silicon dioxide SiO 2 on a
silicon wafer makes silicon the most widely used semiconductor substrate. The silicon
dioxide layer is both an insulating layer on the silicon surface and a preferential
masking layer during the fabrication sequence. Semiconductors can be oxidized by
various methods. These include thermal oxidation, electrochemical oxidation and
plasma A silicon dioxide layer is grown in an atmosphere containing either oxygen (O2)
or water vapour(H20) at temperature in the range of 900°C to 1300°C .
The following chemical reactions describe the thermal oxidation of silicon in oxygen or
water vapour:
Si(solid) + 02(gas) →SiO2(solid)
30
Si(solid) + 2H20(gas) →SiO2(solid) + 2H2(gas)
The silicon - silicon dioxide interface interface moves into the silicon during the
oxidation process. This creates a fresh interface region, with surface contamination on
the original silicon ending up on the oxide surface. The system used for Oxidation
comprises of:
<a> Gas flow system consists of a nitrogen and oxygen cylinder at the back with the
controls at the front to control the flow (Iit/min) of the gases. Then we have the bubbler
with temperature controller.
<b> The Furnace consists of a quartz furnace separated from the body by quartz
wool. The furnace also consists of a PID controller with 3-zone digital temperature
controller along with other facilities like alarm setting.
31
The Working:
Step1: The clean silicon wafer is mounted on a quartz stack with two butters.
Step2: The temperature of the furnace is set at 1000°C at all the 3 zones.
Step3: The Nitrogen cylinder is opened at 2Kg/cm3 and the flow is adjusted at 2lit/min
for 5min.to purge the furnace.
For Dry Oxidation:
Step4: Then oxygen cylinder is opened at 2Kg/cm3 along with Nitrogen for 2min.
Step5: Now Nitrogen flow is stopped and oxygen at 1litlmin. is made to flow through
the furnace for 5min.ln this way dry oxidation is done.
Step6: Reduce the oxygen flow and switch on the bubbler.
For Wet Oxidation:
Step7: Now Oxygen is made to flow through the bubbler with temperature at 93°C and
the flow is increased to 11itlmin slowly and kept it for 15min.
Step8: Again dry oxidation is done for 5min.
Step9: The oxidized wafer is brought out from the furnace and the color is observed to
find the thickness. Also the wafer is observed under the microscope.
32
3.2.3 Photolithography
Lithography is done by standard techniques to expose the region for micromachining.
Figure – 3
33
Microcircuit fabrication requires the precise positioning of a number of appropriately
doped regions in a slide of semiconductor, followed by one or more interconnection
patterns. These regions include a variety of implants and windows in protective cover
layers through which connections can be made to the bonding pads. A sequence of
steps is required, together with a specific layout pattern, for each of these regions
lithographic processes are used to perform these operations and are carried out in
succession during circuit fabrication. The major steps in lithography are:
<1>Fabrication of masks (or pattern generation); and Lithography is the technique of
transferring the pattern on the mask to a layer of radiation sensitive material (resist)
which, in turn, is used to transfer the pattern to the films or substrates through etching
I processes. The radiation used may be optical, X-ray, electron beam (e-beam), or ion
beam. Each technique involves a specialized technology.
<2> Transfer of the pattern to the wafer. In Photolithography, a film of the photo resist
is first applied to the substrate. Radiation is shone through a transparent musk plate,
on which has been imprinted a copy of desired pattern in an opaque material. The
resulting image is focused on to the resist-coated substrate, producing areas of light
and shadow corresponding to the image on the mask plate. In those regions where
light was transmitted through the plate, the resist solubility is altered by a
photochemical reaction. Shadowed areas remain unaffected in solubility. This step is
termed exposure. Following exposure, the substrate is washed with a solvent that
34
preferentially removes the resist areas of higher solubility. This step is called
development. Depending on the type of the resist, the washed-away may be either the
illuminated or shadowed regions of the coating. A resist that loses solubility when
illuminated form a negative image of the plate and is called a negative resist. If
exposure increases resist solubility resist is washed away in the areas corresponding
to the transparent zones of the mask plate. The resist image is identical to the opaque
image on the plate, and the pattern is a photographic positive. Therefore, the resist is
called positive resist. After development, the substrate bearing the patterned resist, is
exposed to an echant. The enchant removes those portions of the substrate
unprotected by the resist while the covered areas remain unetched. Finally, the resist
coating is removed and discarded, leaving a duplicate of the mask plate pattern
etched into a substrate film.
35
The Working:
Step1: The cleaned wafer is put on the top of the spinner hold by vacuum.
Step2: The negative photo resist is spread on the center of the wafer and the spinner
is switched at a speed of 400r.p.m. for 25 sec. for uniform distribution.
Step3: The wafer is put inside the furnace for pre-bake at a temperature of 90°C for
20min.
Step4: The wafer is then put inside the furnace at 120°C for 20min.We used special
photo resist, which did need hard bake.
Step5: The wafer is then put under the microscope and the vacuum pump is put in the
holder of the mask-aligner.
Step6: The wafer is then loaded and the lid is closed and properly aligned by watching
under the microscope thereafter.
Step7: The U-V source is then put in proper place and the wafer is then exposed for 5
sec. to U-V ray.
Step8: The film over the wafer is then developed and rinsed in Acetone to remove
photo resist.
36
Take a solution of 18gm by weight pyrocatachol, 102cc by volume ethylene diamine,
and 48 cc by volume DI water in a beaker and heat it to a temperature of 110 °C.Then
insert the sample in the solution. The etching rate is observed as 80-90 micron/hour at
a constant temperature of 110.The required thickness of the mask is nearly 20 micron
for both pressure and temperature sensors.
Figure 4(a) : Top View of the micromachined silicon membrane
37
Of The Sensor
Figure 4(b) : Top View of the micromachined silicon membrane
38
3.2.5 Porous Silicon Formation
Porous silicon is a derivative of bulk crystalline silicon . Porous silicon consists of
numerous pores and inter-connected silicon crystallites resembling sponge-like / Fur-
tree-like structure , the pores ranging from < 2nm to a few microns.
Porous silicon is currently gaining interest in silicon microsystem technology for its
multifarious application in sensing and photonic devices .The extremely large surface
to volume ratio ( 500 m2 cm-3) of the porous silicon nanostructures, the ease of its
formation and control of the surface morphology through variation of the formation
parameters, its compatibility to silicon IC technology and hence amenability to the
development of smart systems-on chip sensors have made it a very attractive sensing
material. Porous silicon is a three phase mixture of silicon, oxide and voids. The
elastic properties of porous silicon which are drastically different from those of bulk
silicon are responsible for the peculiar characteristics of the porous silicon layers: its
fragility or mechanical instability[1].
There are different techniques for preparation of porous silicon. Two methods have
mainly been applied for the preparation of PS: (i) anodic etching in HF and (ii)
chemical (stain) etching. Our studies on PS have primarily been focused on PS
prepared by anodic etching. In some cases, stain etching has been used subsequent
39
to PS formation by anodic etching in order to increase the porosity of the PS layers.
Many of the physical properties e.g. electrical and optical properties of porous silicon
differ significantly from those of crystalline silicon and these are closely related to the
Porous silicon nanostructure which in turn depends on various parameters like
formation current density, HF concentration, c-Si surface morphology, doping type and
level of c-Si etc[2].
It is commonly prepared by anodic dissolution of Silicon in a solution containing HF
and ethanol as shown in figure - 5
Anodic reaction during pore formation:
During pore formation ,two hydrogen atoms evolve for each silicon atom dissolved.
40
Figure – 5
Porous Silicon is Characterized by a term called POROSITY
Porosity = Volumetric fraction of pores with respect to total volume of porous silicon.
Porosity ranges generally from 20% to 90%[3].
41
42
Figure 6 : FESEM of the nanoporous silicon surface
Figure 7 : Porous Silicon based pressure sensor

3.3 Flowchart For Fabrication of Silicon MEMS Piezoresistive
Pressure Sensor
43
Conventional process
CONVENTIONAL PROCESS

3.4 Flowchart for Fabrication of Porous Silicon Pressure Sensor
44
LATERAL CONTACT CONFIGURATION

3.5 Flowchart for Fabrication of Nanoporous Silicon mems
45
Integrated Pressure and Temperature Sensor
46
References :
[1] Porous Silicon-Based Sensors: Prospects and Challenges H. Saha and C.

Pramanik Materials and Manufacturing Processes, 21: 239–246, 2006
[2] Porous silicon -mechanisms of growth and applications V. Parkhutik

Solid-State Electronics 43 (1999) 1121±1141

Porous silicon C Pramanik, H Saha and U Gangopadhyay
47
Chapter 4 : Design And Simulation Of MEMS Pressure And Temperature Sensor
4.1 Design of nanocrystalline porous silicon based integrated

pressure and temperature sensor
Silicon-based piezoresistive pressure sensors have been widely used for various
applications ranging from Automobiles, robotics, space, process industries to
biomedical applications. But these sensors also show Temperature sensitivity. To
compensate for the temperature effect, a p–n-junction-based temperature sensor is
often integrated with the silicon pressure sensor for commercial applications [1]. There
are also recent reports on the development of an integrated pressure and temperature
sensor on silicon in a suspended microchannel for high-pressure flow studies [2].
Integrated pressure and temperature sensors based on silicon and polysilicon have
also been reported [3]. However, the existing limitation of such a sensor is low-
pressure sensitivity, which makes them incompetent for ultra-low-pressure
applications such as vacuum measurements in the range of a few millitorrs and in
biomedical applications such as respirators. Various methods to improve the
sensitivity paved the way to increased nonlinearity along with the requirement of
complex fabrication steps [4]. The increasing urge towards improving the performance
of pressure sensors without considerably increasing the cost and complexity of
fabrication has stimulated the use of nanocrsytalline silicon, which is also IC
compatible as an active material for pressure sensors due to its increased
piezoresistive coefficient [5].This fact has motivated the use of porous silicon, which is
a natural nanocrystalline material fabricated nonlithographically by electrochemical
etching of silicon, as the active pressure sensing material. Various nonlithographic
techniques have been reported to fabricate well-defined nanostructured porous silicon
48
[4]. The pressure sensing characteristics of the porous silicon–silicon structure have
been recently reported [6].In such reports, the structure of the porous silicon–silicon
pressure sensor is in the sandwich configuration. An added advantage of such a
structure is that it can be used for temperature sensing also since the reverse
saturation current of the porous silicon–silicon heterojunction is reported to vary with
temperature [7].
4.2 Sensing principles:
4.2.1 Pressure sensing principle: The functional principle of a piezoresistive
nanocrystalline porous silicon based pressure sensor is basically similar to that of a
bulk silicon piezoresistive pressure sensor.However, an enhancement in piezoresistive
coefficient is expected in nanocrystalline porous silicon due to the change in
conductivity of the porous silicon layer caused by the deformation in the quantized
valence subbands in the nanocrystallites on application of pressure [4,7]. This
deformation leads to a redistribution of the carriers within the different energy bands,
which is measured as a change in the resistance of the nanocrystalline membrane as
a function of applied pressure. The piezoresistive coefficient in porous silicon can be
tailored by changing the dimensions of the silicon nanocrystallites.
4.2.2 Temperature sensing principle: A porous silicon–silicon heterojunction has
been found to possess temperature sensitivity similar to that of a p–n junction and
can be used as a temperature sensor. It has been observed that in reverse biased
conditions, the porous silicon–silicon heterojunction is more sensitive to temperature
change than in the forward biased mode. This is because under the reverse bias,
injection of minority carrier electrons from crystalline silicon substrate into porous
49
silicon takes place. The injection increases considerably with increasing temperature
and subsequently the resistance of the porous silicon layer decreases significantly. In
the forward biased condition, the current through the heterojunction results primarily
from the electron injection from the aluminum electrode rather than hole flow from the
substrate to porous silicon. This electron injection is apparently tunneling current
through the potential barrier, which has weak temperature dependence[13].
4.2.3 Integrated pressure and temperature sensor: To fabricate an integrated
pressure and temperature sensor, it is important to maintain an optimum distance
between the two sensors such that both the sensors attain a temperature within a
small tolerance of say±0.3% and also avoid the cross-coupling effect. For our
structure of the pressure and temperature sensor, when the ambient temperature rises
to a particular value, the surface of the sensor chip gets heated and a temperature
gradient is developed along the lateral direction. This temperature gradient is primarily
due to the conduction heat losses laterally through silicon and the convection losses
from the edges. The boundary conditions in our case are that there is almost no
temperature gradient just at the center and the conducting heat flux towards the edges
is equal to the convective heat loss from the edges[13].
4.3 Equivalent Model of the Sensor: To analyze the cross-coupling effect, an
equivalent electrical model diagram of the integrated sensor system as shown in figure
8 has been proposed. It is observed from that the resistance between the two sensors
should be large compared to the sensor resistance to avoid any cross coupling.
50
.
Figure 8 : Equivalent Model of the Sensor
In the above figure, Rps and Rpst represent the resistance of the porous silicon layer
of the pressure and temperature sensors respectively, Rsi and Rsit represent the
resistance of the underlying silicon layer in the pressure and temperature sensors,
respectively, Rs represents the isolation resistance between the two[13].
From figure 8, it is observed that to avoid cross coupling and maintain a 0.3% variation
in temperature, the following relationship (equation (2a)) should be maintained:
Rs >> Rps + Rsi or Rs >> Rpst + Rsit, (2a)
and separation between the two sensors d should be such that
d < 0.7L, (2b)
where 2L is the length of the sensor chip.
Now, Rs=ñps(Lps-Wps)/2Wpstps+ñsi0.7L/Wpstps+ñpst(Lpst-Wpst)/2Wpsttpst (2c)
Rps = ñpstps/WpsWps, Rpst = ñpsttpst/WpstWpst,
Rsi = ñSitSi/WpsWps, RSit =ñSitsit/WpstWpst (2d)
51
where Lps, Lpst, Wps and Wpst are the length of the porous silicon layer and contact
width of the pressure and temperature sensors, respectively, tps, tpst, tsi and tsit
represent the thickness of the porous silicon layer and silicon layer in the pressure and
temperature sensors, respectively, ñps, ñps, ñsi and ñsit represent the resistivities of
the porous silicon layer and silicon layer in the pressure and temperature sensors,
respectively.
Porosity for the temperature sensor is also 55% to minimize the fabrication steps. The
temperature sensor is also micromachined to the similar thickness as pressure sensor
for a faster thermal response. The area of the temperature sensor is chosen to be 500
µm × 500 µm and the area of the temperature sensor is 600 µm × 600 µm. The total
thickness of the pressure and temperature sensor is 20 µm of which the thickness of
the porous silicon layer is 15 µm[13].
Figure 9 : Cross-sectional view of the sensor.
52
3.4 Design of Conventional Pressure sensor using lateral contacts :
A piezoresistive pressure sensor consists of a thin monocrystaline silicon
membrane supported by a thick silicon rim as shown in Figure 5. The diaphragm is
fabricated by etching away the bulk silicon on a defined region until a required
thickness is reached. Piezoresistors are made by diffusing or implanting into the
membrane typically close to the edges. The diaphragm acts like a mechanical
stress amplifier. The silicon is not only used as a substrate for the diffused resistors
but also as an elastic material [3].
Figure 10 : A Piezoresistive pressure sensor
When a pressure difference is applied across the device, the thin diaphragm will
bend downward or upward, indicating traction or compression on the
piezoresistors. The resistance change caused by this stress can be easily
measured.
53
Figure 6 shows the four piezoresistors connected in the Wheatstone bridge
configuration. Two resistors are oriented so that they can sense stress in the
direction of their current axes and two are placed to sense stress perpendicular
to their current flow. Therefore, resistance change of the first two piezoresistors
will always be opposite to that of the other two. This is achieved by placing two
piezoresistors parallel to opposite edges of the diaphragm and the other two
perpendicular to the other two edges. When the is diaphragm bent downwards,
causing the tensile stress on the diaphragm surface at the edges, the parallel
resistors are under lateral stress and show a decrease in resistance while the
perpendicular ones are under longitudinal stress and show an increase in
resistance. If the resistors are correctly positioned with respect to the stress field
over the diaphragm, the absolute value of the four resistors changes can be made
equal.
Figure 11 : Wheatstone Bridge Configuration
The sensor has been designed and simulated using MEMSPRO and ANSYS software
tools.
54
Dimension of Silicon membrane = 2500 μm × 2500 μm × 50 μm
Dimension of piezoresistors = 20 μm × 75 μm × 2 μm
Dimension of the metal contact pads = 20 μm × 20 μm × 0.5 μm
Applied pressure = 0.01MPa perpendicular to the surface.
Applied voltage on the contact pads of the piezoresistor = 5V
a = 2500 µm
u = 50 µm
x = 75 µm
t = 50 µm
Figure 12 : Design of the piezoresistive pressure sensor.
55
4.5 Results and discussions :
The design and simulation of the sensor is done by MEMSPRO and ANSYS 10.0
software. Initially the area of the pressure sensor was chosen as 900 µm × 900 µm
and the distance between the temperature and pressure sensor was taken as 100 µm
to avoid cross-coupling effect.
It is observed from the simulation results of the temperature distribution of the
sensor chip that when the pressure sensor is thermally heated to a temperature of
say 350˙C due to ambient temperature, then due to conduction of heat through the
lateral direction, the temperature sensor senses a temperature of 335˙C as shown in
the graph. But the position of the temperature sensor has to be optimized so that it
can sense approximately the same temperature as the ambient temperature. So as
per the simulation results of the temperature distribution of the sensor , it is placed at a
distance of 50 µm from the pressure sensor then it can sense a temperature of 345˙C
as shown in figure.
56
Figure 13 : Top view of the nanocrystalline porous silicon based integrated MEMS
pressure and temperature sensor as designed by MEMSPRO
Figure 14 : Back side of the nanocrystalline porous silicon based integrated MEMS
pressure and temperature sensor as designed by MEMSPRO
57
Figure 15 : Temperature distribution on the sensor, distance between the pressure
and temperature sensor is 100 μm.
Figure 16 : Temperature distribution graph, when the distance between the pressure
58
Figure 17 : Stress distribution on the sensor chip on applied pressure on the membrane
Figure 18 : Temperature distribution on the sensor, distance between pressure

59
Figure 19 : Stress distribution on the sensor chip on applied pressure on the membrane
Distance between the temperature and pressure sensor is 50 μm.
60
Figure 20 : Model of the piezoresistive pressure sensor membrane

As done by ANSYS 10.0
61
Figure 21 : Current density distribution on applied voltage on the piezoresistor
Figure 22 : Current density distribution on applied pressure and

voltage on the piezoresistor
62
Figure 23 : x-component of stress(σl) on the membrane of pressure sensor
Figure 24 : x-component of stress(σl) along the two sides of the piezoresistors,

which are perpendicular to the edges.
63
Figure 25 : y-component of stress (σt) on the membrane of pressure sensor
Figure 26 : y-component of stress (σt) along the two sides of the piezoresistors, which
are perpendicular to the edges.
64
The variation of the stress along the other two piezoresistors (i.e., the one
parallel to the edges) were found to be equal but opposite to the ones shown in
Figure-24. Using the values of x-component of stress and the y-component of stress,
it is possible to work out a fractional change in the value of resistance of a P-type
piezoresistors below:
∆R π44 (σl – σt )
R 2
The Wheatstone bridge configuration converts the resistance change directly to
a voltage signal. The differential output voltage (∆V) of an ideally balanced
bridge with assumed identical (but opposite in sign) resistance change, ∆R, in
response to a differential pressure change ∆P on the sensor, is given by:
∆V ∆R Vbias
R
where R is the zero-stress resistance and Vbias, the bridge supply voltage. The
pressure sensitivity (S) is then defined as the relative change of output voltage per unit
of applied differential pressure and is given by:
S= ∆V1 ∆R 1
∆ P Vbias ∆P R
Note that stress is maximum at the at the edges and minimum at the centre of
the diaphragm. The stress at the edges is maximum because the edges are
constrained (i.e., no displacement along x, y and z- direction)[14].
65
Figure 27 : x-component of the stress on the surface of the diaphragm in a direction

perpendicular to the applied pressure. Note that the stress is maximum at the edges.
Figure 28 : Displacement of the piezoresistors perpendicular to the direction of applied

pressure. Note that displacement varies from one end of the piezoresistor to the other.
66
Figure 29 : Current density distribution on the nanoporous silicon pressure sensor

On applied voltage on the piezoresistor
Figure 30 : Current density distribution on the nanoporous silicon pressure sensor

with applied pressure on the membrane
67
References
[1] Optimized technology for the fabrication of piezoresistive pressure sensors

Merlos A, Santander J, Alvarz M D and Campabadal F 2000 J. Micromech.
Microeng. 10 204–8
[2] A suspended microchannel with integrated temperature sensor for high pressure
Flow studies Wu S, Mai J, Zohar Y, Tai Y C and Ho C M 1997–1998
[3] Fabrication of microdiaphragm pressure sensor utilizing micromachining Fujii T,

Gotoh Y and Kuroyanagi S 1992 Sensors Actuators A 34 217–24
[4] Single crystal silicon piezoresistive nanowire bridge Toriyama T and Sugiyama
S 2003 Sensors Actuators A 108 244–9
[5] Piezoresistive pressure sensing by porous silicon membrane Pramanik C and

Saha H 2006 IEEE Sensors 6 301–9
[6] Design, fabrication, testing and simulation of porous silicon based smart
MEMS pressure sensor Pramanik C, Islam T, Saha H, Bhattacharya J,
Banerjee S and Dey S 2005 18th Int. Conf. on VLSI Design
[7] Effects of uniaxial stress on hole subbands in semiconductor quantum wells

Lee J and Vassell M O 1988 I. Theory Phys.Rev. B 37 8855
[8] Temperature compensation of piezoresistive porous silicon pressure sensor

using ANN Pramanik C, Islam T and Saha H 2006 Microelectron. Reliab.46
343–51

porous silicon C Pramanik, H Saha and U Gangopadhyay, J. Micromech.
Microeng.16 (2006) 1340–1348
[10] A Silicon Piezoresistive Pressure Sensor Ranjit Singh, Low Lee Ngo,Ho Soon
Seng, Frederick Neo Chwee Mok
[11] ANSYS 10.0 Help File and Tutorial.
[12] MEMSPRO Software.
68
Chapter 5 : Summary and Conclusion
5.1 Summary :
Piezoresistive pressure sensors have wide applicability in various systems including
instrumentation, automobiles, biomedical and process control systems. But
conventional MEMS piezoresistive pressure sensors possess low sensitivity, low
resolution and high nonlinearity for low-pressure vacuum applications. To circumvent
the problem of sensitivity and nonlinearity, ribbed and bossed structured diaphragms
have been used in piezoresitive pressure sensors. Capacitive type pressure sensors
have also substituted piezoresistive type sensors in some vacuum applications. But
both the above techniques require complex fabrication steps. A simpler solution would
be to replace monocrystalline silicon by nanocrystalline silicon, which has been
observed to possess one and a half times greater piezoresistive coefficient and three
times higher sensitivity.
A detailed design of a novel integrated sensor on nanocrystalline porous silicon has
been done with a view to minimize the temperature difference between the pressure
and temperature sensors and also avoid the cross-coupling effect. The design of the
pressure sensor has been done for typical specifications of low-pressure applications.
An added advantage of the integrated pressure and temperature sensor is that the
optimum porosity is obtained to be around 55% for both the presssure and
temperature sensors leading to a reduction in the fabrication steps. This integrated
structure thus provides the advantage of increased pressure sensitivity, reduction in
the number of fabrication steps, precise temperature monitoring for temperature
69
Chapter 5 : Summary and Conclusion
compensation of the pressure sensor and compatibility for post-CMOS processing
since porous silicon formation is a low-temperature process.
Various methods to improve the sensitivity paved the way to increased nonlinearity
along with the requirement of complex fabrication steps .The increasing urge towards
improving the performance of pressure sensors without considerably increasing the
cost and complexity of fabrication has stimulated the use of nanocrsytalline silicon,
which is also IC compatible as an active material for pressure sensor.
5.2 Conclusion :
Through simulation by ANSYS 10.0 we have found that the temperature sensor must
be placed at a distance of 50 µm from the pressure sensor in order to sense the
maximum temperature of the ambient and also to avoid cross-coupling effect. Hence,
the distance between the temperature and pressure sensor is optimized to show good
sensitivity results.
For the piezoresistive nanoporous pressure sensor the effect of pressure have been
studied for the particular design of the sensor. The simulation results shows the x and
y components of stress which calculates the sensitivity of the sensor with applied
voltage.
Based on these simulation results we can fabricate the sensor and hence obtain
optimized results.
70
Appendices
Calculations :
Conventional pressure sensor :-

Design Specifications-
Resistivity of the piezoresistor=0.1 Ω-cm=0.001 Ω-m
Length of the piezoresistor=75 × 10-6 m
Cross-sectional area of the piezoresistor= (20 × 2) × 10-12 m2
We know that,
R= ρ( L÷A)
ρ = resistivity of the piezoresistor
R = 0.001× (75÷ 40)×106 = 1875 Ω =1.875 K Ω
Applied Voltage = 5V
Hence, current flow= I = 5÷1.875= 2.666 × 10-3 A
Current Density= I÷A= (2.666 × 10-3 )÷ (40 × 10-12 )
= 6.666 × 10 7 A/m2= 0.666 × 108
pA / (μm)2
Simulation Results :-
Current density of the piezoresistor before application of
pressure
= 0.628 × 10 8
pA / (μm)2
Cross-sectional area of the piezoresistor = 40 (μm)2
Hence, current flow in the piezoresistor = 0.628 × 108 × 40 =
25.12 × 108 pA
71
= 25.12 × 10-4 A =
2.512 × 10-3 A
On application of pressure , current density of the
piezoresistor
= 0.596 × 10 8 pA
/ (μm)2
Hence, current flow in the piezoresistor = 0.596 × 108 × 40 =
23.84 × 108 pA
= 23.84 × 10-4 A =
2.384 × 10-3 A
Appendices
Nanoporous silicon pressure sensor :-

Design Specifications-
Resistivity of the piezoresistor = 10 Ω-m
Length of the piezoresistor = 75 × 10-6 m
Cross-sectional area of the piezoresistor= (20 × 2) × 10-12 m2
We know that,
R= ρ( L÷A)
ρ = resistivity of the piezoresistor
R = 10× (75÷ 40)×106 = 18750 K Ω
Applied Voltage on the piezoresistor = 5V
Hence, current flow = I = 5÷18750= 2.666 × 10-4 × 10-3 A =
2.666 × 10-7 A
Current Density= I÷A= (2.666 × 10-7 )÷ (40 × 10-12 )
72
= 6.666 × 103 A / m2 = 6666 pA /
(μm)2
Simulation Results :-
Current density of the piezoresistor before application of
pressure
= 6284 pA /
(μm)2
Cross-sectional area of the piezoresistor = 40 (μm)2
Hence, current flow in the piezoresistor = 6284 × 40 = 251360
pA
= 251360 × 10 -12 A =
2.512 × 10-7 A
On application of pressure , current density of the

piezoresistor
= 5866 pA /
(μm)2
Hence, current flow in the piezoresistor = 5866 × 40 =
234640 pA
= 234640 × 10 -12 A =
2.346 × 10-7 A
Appendices
Steps of Modelling and Simulation by ANSYS 10.0:
A) Simulation of Conventional Pressure Sensor by ANSYS 10.0 software
To open the work-plane point to
73
Start > ansys 10.0
Two windows will be opened…. 1) ansys utility menu
2) output window.
Aim is to:
1. Apply a constant voltage to the piezoresistors and plot current density distribution.
1. Define Element Type and option:
–Main Menu > Preprocessor > Element Type > Add/Edit/Delete
•Select [Add …]
•Select “Coupled Field” (on left), and “Scalar Tet 98” (on right), then [OK]
•Select [Options …]
Select ‘Degrees of freedom selection’ as “VOLT”, the [OK]
•Select [Close]
2. Define material Properties:
–Main Menu > Preprocessor > Material Props > Material Models
Select New Material
Double Click on electromagnetics,resistivity,constant.
Enter value for resistivity
Appendices
OK to define material property set and close the dialog box.
Material > close.
3. Define the material Structure:
74
Main menu > preprocessor> Modelling > Create > volumes> block > by dimension.
Enter proper values.
Click apply to create more volumes or ok to finish .
Now that the appropriate pieces of the model are defined (volumes),you need to
add/subtract them together so that the model becomes one continuous piece.
Main menu > preprocessor >Modeling > Operate > Boolean >sub volumes/Add
volumes.
Main menu > preprocessor >Modeling > Operate > Boolean>glue volumes.
4. Generate Mesh:
–Main Menu > Preprocessor > Meshing > MeshTool
•Select [Mesh], then [Pick All]
•Then [Close]
5. Define Load :
A) Plot areas to identify where to impose boundary conditions:
–Utility Menu > PlotCtrls > Numbering
•Select “Area numbering”, then [OK]
–Utility Menu > Plot > Areas
B) Define voltage boundary conditions:
Main Menu > Preprocessor > Loads > Define Loads > Apply > Electric > boundary >
Voltage > On Areas
Appendices
75
•Pick proper area where to apply voltage, then [Apply]
•Enter ‘Constant Value of voltage (VOLT)’ as “5”, then [Apply]
•Pick the second area, then [OK]
•Enter ‘Constant Value of voltage (VOLT)’ as “0”, then [OK]
6.Solve electric problem(aim 1):
Main Menu > Solution > Solve > Current LS>OK
7.A) Review current density distribution:
Main Menu > General Postproc > Plot Results > vector Plot >predefined >current
density>coupld source >OK…
7B) Review electric potential distribution:
Main Menu > General Postproc > Plot Results > Contour Plot > Nodal Solution
Select “DOF solution” (on left), and “Elec poten VOLT” (on right), then [OK]
2. To determine the change in current density with applied pressure on the
piezoresistor
1. Redefine element type :
Main Menu > Preprocessor > Element Type > Add/Edit/Delete
Select [Add …]
Select coupled source (on left), and “solid 226” (on right)
Click [Close]
76
Appendices
2. Define material properties :
Main Menu > Preprocessor > Material Props > Material Models
For “Material Model Number 1’ click on ‘Structural’
Choose ‘Linear’, then ‘Elastic’, then ‘Isotropic’
Enter the values of EX and PERX, then [OK]
Choose piezoresistivity, piezoresistive stress matrix then OK
Enter the value of electrical resistivity, then [OK]
On Main Menu, Select ‘Material’, then [Exit]
3. Redefine the structure:
In the same way as mentioned above
4.
Generate Mesh
In the same way as above
5. Apply load
Apply structural constraints on areas connected to base
Utility Menu > Select > Everything , –Utility Menu > Plot > Areas
Main menu >preprocessor>Loads>Define Loads>apply>structural>displacement>on
areas
Select the area and pick it
Apply displacement in x,y,z direction as zero>OK
77
Appendices
6. Specify applied pressure
Main Menu > Preprocessor > Loads > Define Loads > Apply > Structural > pressure >
on areas
Apply a constant pressure of -0.01MPa to the membrane surface.
9. Specify applied voltage
Main Menu > Preprocessor > Loads > Define Loads > apply>boundary>voltage >on
areas
Enter the constant voltage value of 5V on the picked area
10. Solve problem
11. Review results
Main Menu > General Postproc > Plot Results > vectorplot>current density
78
Appendices
B) Simulation of nanocrystalline porous silicon based Pressure Sensor by
ANSYS
Same as above only replacing the material properties of silicon by porous silicon
C) Simulation of integrated pressure and temperature based nanoporous
MEMS sensor
1) Apply a constant temperature to the pressure sensor and study the temperature
distribution
1. Define Element Type and option:
Select [Add …]
Select thermal mass as solid 90 (on right), then [OK]
2. Define material Properties:
Select New Material
Double Click on thermal conductivity,density,specific heat.
Enter their value.
OK to define material property set and close the dialog box.
79
Material > close.
3. Define the material Structure:
Import the structure of the sensor chip from MEMSPRO .
Main menu > preprocessor >Modeling > Operate > Boolean>glue volumes.
Appendices
4. Generate Mesh:
Main Menu > Preprocessor > Meshing > MeshTool
Select [Mesh], then [Pick All]
Then [Close]
5. Define Load :
A) Plot areas to identify where to impose boundary conditions:
Utility Menu > PlotCtrls > Numbering
Select “Area numbering”, then [OK]
Utility Menu > Plot > Area
B) Define temperature conditions:
Main Menu > Preprocessor > Loads > Define Loads > Apply >thermal>convection>
areas
Pick proper area where to apply temperature, then [Apply]
Enter ‘Constant Value of TEMP as 300K’to the boundary.
Enter film coefficient value as 2.2e6,OK
80
Enter an applied temperature of 350 K on the picked area(membrane)
6. Solve
7 Review temperature distribution:
Main Menu > General Postproc > Plot Results > contour Plot >nodal temperature>OK
Appendices
2. To determine the stress distribution with applied pressure we do the
following steps:
A) Redefine element type
Select [Add …]
Select coupled source (on left), and “solid 226” (on right)
Click [Close]
B) Define material properties
•For “Material Model Number 1’ click on ‘Structural’
Choose ‘Linear’, then ‘Elastic’, then ‘Isotropic’
Enter the values of EX and PERX, then [OK]
81
Choose piezoresistivity, piezoresistive stress matrix then OK
Enter the value of electrical resistivity, then [OK]
On Main Menu, Select ‘Material’, then [Exit]
C) Redefine the structure: As mentioned earlier using MEMSPRO software.
Main menu > preprocessor >Modeling > Operator > Boolean > glue > picked volumes
D)
Generate Mesh
Main Menu > Preprocessor > Meshing > Mesh attributes >picked volume
Main Menu > Preprocessor > Meshing > mesh tool
Select [Mesh], then [Pick All]
Then [Close]
Appendices
E) Apply load
Apply structural constraints on areas connected to base
Utility Menu > Select > Everything , –Utility Menu > Plot > Areas
Main Menu>Preprocessor>Loads > Define Loads>apply>structural>displacement>
on areas
Select the area and pick it
Apply displacement in x,y,z direction as zero>OK
F) Specify applied pressure
Main Menu > Preprocessor > Loads > Define Loads > Apply > Structural >
pressure > on areas
Apply a constant pressure of -0.01MPa to the membrane surface.
82
G) Solve problem
Main Menu > Solution > Solve > Current LS
Review status window then select [OK]
H) Review results
Main Menu > General Postproc > Plot Results > contour plot>stress>von mises
stress>OK
83

M.Tech Thesis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

M.Tech Thesis

Uploaded by

Copyright:

Available Formats

“NANOCRYSTALLINE POROUS SILICON BASED

INTEGRATED MEMS PRESSURE AND

Under the guidance of

Thesis submitted in partial fulfillment

IC Design and Fabrication Centre

I would like to take this opportunity to express my gratitude and heartiest

thanks to Prof. H. Saha, IC Design and Fabrication Center, Department of Electronics

& Telecommunication Engineering, Jadavpur University, for his inspiration,

encouragement and guidance in the fulfillment of my project work.

the same department of their help when ever I needed it.

for her constant help throughout the progress of my work.

I am grateful to Prof.S.Basu, Ex Professor, I.I.T Kharagpur and at present

necessary suggestions and helpful scientific discussions to improve the quality of my

Telecommunication Engg, Jadavpur University, Kolkata.

encouragement and support at different level of my work.

1.1 General Introduction

1.2 Objective of this project

1.3 Literature Review

1.4 Scope of this work

Chapter : 2 Overview of micromachining and micromachined pressure

2.2 Overview of Micromechanical Structure

2.2.2 Anisotropic Etching of Silicon

2.2.3 Top view of the micromachined silicon membrane

Chapter : 3 Fabrication of the sensor

3.2 Fabrication steps

3.2.5 Preparation of Porous Silicon

Piezoresistive Pressure Sensor

3.4 Flowchart For Fabrication of porous Silicon Pressure Sensor

3.5 Flowchart for fabrication of nanocrystalline porous silicon based

pressure and temperature sensor

Chapter: 4 Design and simulation of the sensor

4.1 Design of nanocrystalline porous silicon based integrated pressure

and temperature sensor

4.2 Sensing Principles

4.2.1 Pressure sensing principles

4.2.2 Temperature sensing principles

4.2.3 Integrated temperature and pressure sensor

4.3 Equivalent model of the sensor

4.4 Design of Conventional Pressure sensor using lateral contacts

4.5 Results and Discussions

Chapter : 5 Summary and Conclusion

1.1 General Introduction:

combining electrical and mechanical components fabricated using Integrated Circuit

(IC) compatible batch-processing techniques and range in size from micrometers to

function individually or in arrays to generate effects on the macro scale.

Using similar fabrication techniques as building microprocessors, we are now able to

viscosity sensors, resonators, levers, gears, transmission systems, micro-mirrors,

processing unit. They indeed compose a “system on a chip”[1].

Monocrystalline silicon MEMS pressure sensor suffers from high temperature

been conceived[2]. The extremely large surface-to-volume ratio (>=500m2/cm3) of a

morphology through variation of the formation parameters, its compatibility of silicon

IC technology and hence amenability to the development of smart systems-on-chip

sensors—all these properties have made it a very attractive sensing material[3,4].

1.2 Objective of the Work :

There are two objectives of the present work :

• Design and simulation of nanocrystalline porous silicon based MEMS

integrated pressure and temperature sensor by ANSYS 10.0

• Fabrication of nanocrystalline porous silicon based MEMS integrated

pressure and temperature sensor.

1.3 Literature Review :

the piezoresistive effect of silicon ands germanium, which is a change of resistance

with applied stress[5].This discovery enabled production of Semiconductor - based

1958[2]. Piezoresistive pressure sensors have piezoresistors mounted on or in a