© 1997 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media,
including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution
to servers or lists, or reuse of any copyrighted component of this work in other works.
Development of
ii
MEMS Testing Methodology'
Abhijeet Kolpekwar and R. D. (Shawn) Blanton
Center for Electronic Design Automation
Department of ECE
Carnegie Mellon University
Pittsburgh, PA 15213-3890
Ablstract
Macroe1ectromechanica:l systems (MEMS) are miniature electromechanical sensor and actuator sysiems
developed from the matwe batch-fabricated processes
of VLSI technologies. Projected growth in the MEMS
market requires significant advances in CAD and manufacturing for MEMS. These advances must be accompanied with testing methodologies that ensure both high
quality and reliability. W e describe our approach for
developing a comprehensive testing methodology for a
class of MEMS known as surface micromachined !;ensors. Our first step involving manufacturing process
and low-level mechanical simulations is illustrated b y
studying the effects of realistic contamination:s on the
folded-flexure comb-drive resonator. The sinzulal'ion
results obtained indicate dhat realistic contaminations
can create a variety of defective structures that result
in a wide spectrum of faulty behaviors.
1
Introduction
Microelectromechanical systems (MEMS) are miniature electromechanical sensor and actuator systems
developed from the mature batch-fabricated processes
of VLSI technologies. MElMS have wide applications
such as miniature inertia,l measurement units, biochemical analysis on a chip, arrayed micromanipulation of parts, optical displays and micro-probes for
neural recording. The current and increasing success of
MEMS stems from its proimise of better performance,
low manufacturing cost, miniaturization and its capacity for integration with electronic circuits.
The low cost of MEMS, as with the integrated circuit (IC), is attributed to the amortization of capital
cost over millions of individual batch-fabricated clevices. This has substantiallly reduced the cost of sensors and actuators by several orders of magnitude. For
'This research effort is sponsored by the National Science
Foundation under grant MIP-9702678 and the Defense Research
Projects Agency under Rome Laboratory, Air Force Materiel
Command, USAF, under grant F30602-97-2-0323.
example, Analog Devices' MEMS accelerometer used
extensively in airbag systems sells for about $5 a unit
when sold in large quantities. An accelerometer of similar quality manufactured from a competing technology
sells for more than $1000.
The smaller and reduced-mass characteristics of
MEMS structures also allow higher operating frequencies and lower power consumptions. These properties
make MEMS ideal for many portable and remote applications. For example, MEMS-based pressure sensors
are being used to monitor truck tire pressure in order to extend tire life, while optical gratings are being
developed for heads-up color displays.
The ability to use the same manufacturing technology to integrate MEMS and electronic systems adds
another dimension of utility. Single-chip systems of
electronics and mechanical sensors are being produced
and envisioned for a variety of applications. Texas Instruments, for example, has developed a micromirror
projection display consisting of a two-dimensional array of aluminum mirrors, each individually controlled
by on-chip RAM cells.
By the early 1980s, progress in microelectronics had
reduced the cost of a microprocessor to less than that
of a typical silicon sensor; today the peripheral functions of sensing and actuation continue to represent the
principal bottlenecks in the application of microelectronics to many emerging systems, not only in terms
of cost but in terms of reliability and accuracy as well.
It is thus evident that continued progress in sensors,
actuators, and MEMS is likely to exert considerable
leverage in the microelectronics industry beyond the
direct markets for these products. Accelerometers,
pressure sensors and ink-jet printers are the most matured MEMS applications today accounting for almost
all of the commercial MEMS market of around $1 billion in 1994. The MEMS market is projected to reach
between $12 and $14 billion by the year 2000 [l].The
technology trend is towards higher degrees of integration that result in greater capabilities by MEMS pro-
INTERNATIONALTEST CONFERENCE
0-7803-4209-7/97 $1 0.00 0 1997 IEEE
Paper 36.3
923
Mechanical
simulation
Figure 2: Contamination analysis of microelectromechanical systems.
Figure 1: A SEM of a surface-micromachined combdrive microresonator.
ducts to asses and manipulate their surrounding environment. Realizing these projections requires that
the open problems in design methodology and process
and package technology be confronted and adequately
addressed.
There is a desperate need for computer-aided design
(CAD)tools that shorten the design and development
time for MEMS-based products. Current design cycles
are measured in years [a]. Success in this area depends greatly on new design methodologies that allow
complex microsystems of mixed domain (mechanical,
electrical, thermal, etc.) to be hierarchically represenled arid simulated. Irnproverrients in packaging and
process technology will also be required to support the
mixed-system applications required by MEMS iritegralion. Currently] most MEMS are made horn commonly
used IC fabrication techniques. Process and equipment development specifically for MEMS will expand
the design space and result in more opportunities for
completely integrated systems.
Advances in the above areas alone will not fuel
MEMS growth. Continued success for MEMS will require cost-effective methods of manufacturing. Similar to digital ICs, success in this area must includc a
testing methodology that allows products to be economically tested while ensuring high quality and reliability. This is especially important in applications
where MEMS-based devices are integral parts of safetycritical systems. MEMS testing methodologies must be
Paper 36.3
924
developed in concert with new design, packaging] and
process technologies. To our knowledge] no prior work
has been made in this area. This paper addresses the
need for a MEMS testing methodology. Most industrial practices focus on just checking the functionality of the MEMS-based products by performing certain electrical measurements. No proper mapping is
made between failure modes and the underlying physical mechanisms causing them. Such mappings are critical for providing important guidelines for modifying
process controls that result in fast yield improvements.
We believe that MEMS CAD tools capable of assessing
faulty behavior will lead to test stralegies and designfor-testability (DFT) techniques that improve and ensure the end quality of MEMS-based products.
Faulty MEMS behavior can result from process contaminations that affect the structure or material properties of the microstructure. Process simulation can be
used to predict the effects that contaminations have
on the physical geometries of sensing devices. The
perturbed properties of the resulting device can then
be mapped to the corresponding faulty behavior(s) by
electromechanical simulation. Thcse dcrivcd behaviors
can then be used to develop DFT techniques (in the
form of structural or material modifications) that make
harmful defects less likely, tolerable (fault tolerance),
and easily detectable while preserving the specified operation of the microstructure. This approach can also
be used to form links between defects and faulty behaviors. Such links would aid in diagnosis by helping
to identify culprit processing steps that are likely t o
produce observed faulty behavior.
In this paper, we describe the impact of realistic
spot contaminations on the operation of one class of
MEMS devices, surface- micromachined sensors. We
will illustrate our approach by considering a foldedflexure comb-drive microresonator. Figure 1 shows
such a microresonator. We have chosen the resonator
as our research vehicle lbecause it possesses many of
the basic structures (beams, joints, springs, etc.) that
many MEMS CAD met,hodology researchers believe
will form the core primitives of a MEMS design library. Thus, our approach to developing a complete
MEMS testing methodollogy relies on a fault inductive
approach that :
Completely characterizes the faulty behavior of
MEMS primitives.
Supports the generation of fault macro-models for
simulation and test generation.
Provides techniques for testable and reliable
MEMS design.
The seminal work on the folded-flexure electrostatic
comb-drive microresonator is given in [3]. The resonator is used as a characterization structure in polysilicon surface-micromachining processes to determine
Young’s modulus of the structural polysilicon film. Analytic models have been derived for the foldedl-flexure
spring constants [4], viscous damping [5], and elec1:rostatic force and capacitance of the lateral comb drive
[6]. The structure is a mature case study in the design
of “suspended MEMS” which are now used iin commercial accelerometers [7 81, gyroscopes (soon), and
micromirror optical beam steering [9]. Futuire commercial applications are in resonator oscillators [ LO],
IF mixers, high-Q IF filters for communications, and
microstages for probe-based data storage [ll].
Prototype surface micromachining processes are
available from MCNC (Multi-user MEMS Processes
service (MUMPs)) [12], Analog Devices’ iMEMS process and from Sandia National Labs [13]. We have
chosen MUMPs for our contaminations simulations
of the resonator.
The contamination simulations
have been performed using a modified version of the
Contamination-Defect-Fault simulator (CODEF) tool
[14]. Behavior of the defective resonators generated
by CODEF are manually analyzed to develop ithe defective wire models used by the mechanical simulator
ABAQUS [15]. Our overall approach is illustrated in
Figure 2.
The rest of this paper is organized as follow,s. Section 2 describes the normal operation of the microresonator. Section 3 describes the process simulator
Figure 3: Top view of a comb-drive microresonator.
CODEF and its modification for the MUMPs process.
In Section 4 , we describe our process and mechanical
simulation experiments and the results obtained. Section 5 presents our conclusions and outlines our future
work in this area.
2
Micromachined Microresonator
The folded-flexure comb-drive microresonator that
we consider for our analysis can be fabricated using the
three-layer polysilicon MUMPs process. (Appendix
A lists the complete sequence of the MUMPs process
steps.) Figure 3 shows the top view of a resonator. The
resonator is a mass-spring-damper system made only
from the first and second polysilicon layers. A majority of the resonator’s structure is suspended above
the wafer surface. It is connected to the wafer only
at the four anchor locations shown. The shuttle mass
is supported by symmetrical folded flexures that are
designed to be compliant in the 2 direction and rigid
in the y direction. The shuttle is electrostatically actuated in the 2 direction by applying a voltage across
the comb drives. Each comb drive is formed from interdigitated fingers on either side of the shuttle mass.
Lateral motion in the 2 and y directions are modeled
by the second order equations:
where Fe,,and Fe,yare the electrostatic forces generated by the comb drives in the 2 and y directions,
respectively. The parameter values for the effective
masses (m, and my), damping coefficients ( B , and
By),and spring constants (kZ and ky) are calculated
from the material properties and geometries of the resonator.
The structural parameters El61 that define the resonator’s operation are described in Table 1 and illustrated in Figure 4. These parameters along with properties of the materials are used to model the operation
of the resonator. The effective masses m, and m y are
Paper 36.3
925
- 4 L
Parameter
Typical
description
value
length
- of flexure beam
width of flexure beam
length of truss beam
width of truss beam
length of shuttle yoke
width of shuttle axle
width of shuttle yoke
length of comb fingers
gap between comb fingers
comb finger overlap
number of comb fingers
thickness of resonator
Young’s modulus of poly
density of poly
shuttle mass
long beam mass
truss section mass
300 pm
2 Pm
81 pm
4 Pm
49 pm
11 pm
11 pm
11 pm
2 pm
4 pm
87
2 Pm
165 GPa
2330 Kg/m3
4.35e-1° Kg
2.24e-11 Kg
9.06e-12 Kg
Table 1: Dimensional parameters (and typical values)
that define the modeled operation of the microresonator.
for the
2:
IC,
IC,
(c)
Figure 4: Pertinent dimensional parameters of the (a)
shuttle, (b) folded-flexures, and (c) comb drives of the
microresonator.
and y directions are taken from [4]
Fe,r
1
12
4
35
8
my = m,+-mt+mb
35
= m,+-mt+-mb
(3)
(4)
f,
Xmax
Paper 36.3
926
(6)
+
(7)
=
t
1.12~~N-V~
9
(8)
where E , is the permittivity of air and V is the voltage
applied to the N-finger comb drives. The above relationships define the resonator’s maximum deflection
x,,,
and resonant frequency f, [16]
where m, is the shuttle mass, and mt and mg are the
total masses for the all the truss sections and long
beams, respectively. From [SI, the damping coefficient
in the 2 direction is
where p is the viscosity of air, d is the spacer gap, 6
is the penetration depth of airflow above the resonator
structure, g is the gap between comb fingers, and A , ,
At, Ab, and A, are the surface areas for the shuttle,
truss beams, flexure beams and comb-finger sidewalls,
respectively.
The linear equations for the flexure spring constants
+
+
where E is Young’s modulus for polysilicon, t is the
polysilicon thickness and a = ( W t / W b ) 3 . For the special case of w, = g = t = d, we have from [6] the force
generated by each of the comb drives
given by
m,
+
+
+
+
2 E t ~ ;L: 14aLtLb 36a2Li
L t 4L: 4laLtLb 36a2Li
2EtwB 8L,Z 8CXLtLb 4-a 2 L ?
= L; 4L: 10aLtLb 5a2L:
= -
=
-
2n
m,
BZ
k,
It may, however, be noted that the above mathematical models have limited accuracy because they
do not account for beam compression and axle bending
effects.
3
CODEF
The contamination-defect-fault (CODEF) mapper
[14] is a comprehensive process simulation tool that
account for the MUMPS “release” step (i.e. the HF
etch of the sacrificial phosphosilicate glass (PSG) that
releases the suspended microstructure) which does not
exist in the standard CMOS process. Defect analysis of MEMS cannot be accomplished using the normal CODEF flow of circuit extraction and simulation.
Electromechanical simulation is required to analyze the
effects of the contaminations on the mechanical structure of the device. We have used the mechanical simulator ABAQUS [15] t o analyze the critical parameters
of defective microresonators. ABAQUS is a finite element analysis tool that requires an input structure
to be defined in terms of nodes and elements (i.e. a
mesh). Thus, mechanical simulations were carried out
by defining two-dimensional meshes for each of the defective microresonators analyzed.
4
Simulation Results
In this section, we describe our process and electromechanical simulations of the resonator.
CODEF Simulations
Figure 5: Three-dimensional views of microresonistors
with (a) shuttle, (b) comb, and (c) beam defects generated by CODEF.
maps spot contaminations to layout defects. Contaminations are unwanted foreign particles in the fisbrication process that cause defects in the material and
structural properties of tlhe intended device. E’au1t.s are
the possible changes in the behavior caused by defects.
CODEF accepts layout information, a process description, and statistics that diescribe contamination parameters in each processing sl,ep of interest including particle size, density and condluctivity. As output, COIIEF
produces cross-sections or the resulting microstructure
a t any step of the fabrication process. Given the contamination parameters, CODEF simulates all the fabrication steps and creates a three-dimensional st cucture of the defective device. For the case of electrical
circuits, an extractor is then used to extract information about transistor sizes and connectivity in the form
of a standard HSPICE netlist [17].
To make use of CODElF for MEMS contanninaltion
simulations, we have defined a complete MUMPS fabrication process as a sequence of steps in the IPREDITOR format [MI.CODEF itself was also modified to
A microresonator using the design values listed in Table l was provided to the modified version of CODEF.
Several thousand process simulations involving single
contaminations were performed. Three-dimensional
views generated by CODEF for three of the defective
resonators are shown in Figure 5.
Shuttle Defect: Figure 5a shows a defective
shuttle, resulting from a 2pm diameter, conducting contamination with density 3000 Kg/m3, occurring during deposition of the second layer of
polysilicon (polyl). Notice the formation of a
small bump on the shuttle surface. The defect
increases the shuttle mass m, and hence, the effective mass m,. This implies that the resonant
frequency fz (Eq 9) and the maximum deflection
x,,,
(Eq 10) will change from their nominal values. The actual change in f, and xmar caused by
the increase in m, is dependent upon the size of
the resonator. For the design parameters of Table 1, mechanical simulation has shown negligible
changes in fx (0.035%). (See shuttle defect 12 of
Table 2).
Comb Finger Defect: Figure 5b shows the effect of another 2pm contamination occurring after
the deposition of PSG. As shown, the resulting defect welds together two overlapping comb fingers
which can result in a catastrophic failure. (For
example, see comb defect 5 of Table 2).
Paper 36.3
927
-comb
defects
shuttle defects 12,13
truss defects
2
Figure 7: The locations of 13 different defects analyzed
using CODEF and ABAQUS.
Figure 6: A microresonator with broken beams.
Flexure Defect: Figure 5c shows a defective
resonator that is neither catastrophic nor possesses behavior that is readily predictable from
the mathematical model presented. It is the result of a 2pm diameter contamination occurring
after deposition of PSG. If we assume the defect forms a weak link between beam parts, then
it is likely complete breakage will occur causing
gross changes in the resonator’s critical parameters (beam defect 1 of Table 2 ) . An example of
such a situation is illustrated in Figure 6 for a
real single-finger microresonator. Conversely, if
the defect forms a strong but defective beam, the
resulting parameter changes may be less severe or
non-existent. (For example, see beam defect 2 of
Table 2).
AB AQUS Simulations
Figure 7 identifies the location of 13 different defects
resulting from the process simulation of 2pm contaminations. (Contaminations of 2pm are typical given the
total silicon area of the resonator.) Table 2 gives the
process step of introduction for each of the 13 contaminations and their effects on the critical parameters of
the resonator which were obtained using the mechanical simulator ABAQUS. The second row of the table
lists the nominal values resulting from simulation of
the defect-free resonator. Table 2 indicates that some
Paper 36.3
928
defects can cause a complete failure of the microresonator (defect 5) while others (defects 2, 3, 6, and 12)
do not affect any of the critical parameters. However,
such defects may pose a threat to the long-term reliability of the microresonator similar t o active redundant
faults in digital circuits. There are also defects that
perturbed the resonator in such a way that prevents
mechanical simulation from being performed (defects
4, 7, 11, and 13). This is due to the fact that the exact nature of the contamination effect is uncertain and
cannot be predicted or modeled accurately.
There are a few observations that can be made
about the defects that do affect the resonator. For example, defect 1 affects all the critical parameters while
defects 8 and 9 affect only a subset of the parameters.
Some parameter changes are quite small. For instance,
defect 8 and 9 decrease k , and fy only by 2.0% and
1.3%, respectively. Other changes are large and catastrophic. Defect 5, for example, has several orders of
magnitude change in all parameters. Moreover, there
are defects located at different sites that cause the same
faulty behavior. For example, defects 8 and 9, located
at different locations of the truss (Figure 7), map to
the same parameter changes in k,, k,, and f y .
Defects which posed difficulty for mechanical simulation (defects 4, 7, 11, and 13) all resulted from
contaminations occurring after the poly0 resist etch
(step 8 of the MUMPS process), Contaminations at
this step (affecting the comb, beam and shuttle structures) creates a contact between the poly0 and polyl
layers. We postulate that these unintended contacts
probably cause friction between the two layers during
shuttle movement. Another possibility is that the contamination forms a weak anchor between the poly0 and
polyl layers which prevents shuttle movement entirely.
Defect
No.
None
1
2
3
4
5
6
7
8
9
10
11
12
13
Defect
location
None
Beam
Beam
Beam
Beam
Comb
Comb
Comb
Truss
Truss
Truss
Truss
Shuttle
Shuttle
1
I
Cc
ad
stc
Nc
6202
6015
6202
6202
uncertain
0
6202
uncertain
6119
6119
6202
uncertain
6202
uncertain
9
19
20
8
9
20
8
9
9
19
8
20
8
0
-3.2
0
0
100
0
-1.3
-1.3
0
0
-
Table 2: Critical parameters of defec1;ive resonators derived from mechanical simulations.
Conclusions arid Future Work
5
Process simulations of the microresonator illustrate
that spot defects can have a significant impact on the
structure of a surface-micromachined resonator. The
mechanical simulations ,then performed on the defective resonators has shown that :
to classify these defective structures into an appropriate fault model, we are building an extractor tool that automatically derives (when possible) the structural and material parameters that
define behavior.
o
1. Some defects, as expected, cause catastro'phic
changes in resonator behavior.
2. Other defects cause subtle or negligible changes
in behavior but may have some implicattions on
long-term reliability.
0
3. The impact of some defects cannot be predicted
with mechanical simulation alone and ma;y require
experiments with real defective MEMS.
We pian to conduct a large number of randlom process simulations using real contamination data at every
step of the manufacturing process in order to produce
a large spectrum of defective MEMS structures. Lowlevel mechanical simulations will be performed to categorize these defective structures into a smaller set of
faulty behavior classes. 'These fault classes will form
the basis of MEMS fault models and serve as our jirst
step in developing a comprehensive testing methodology for such systems.
There are several open research areas that we plan
to address concerning the development of a MEMS
testing methodology.
0
Structure extraction: Our testing methodology
depends on gaining a comprehensive understanding of the effects of spot contaminations. Process
simulations of random contaminations will create
a large variety of perturbed structures. In order
0
0
Fault model verification: We plan to verify the
accuracy of the developed fault models using actual fabricated systems. Our approach will use
defective devices to measure and compare actual
and predicted faulty behaviors.
Test methodology grading: Given an accurate fault model, one can measure the effectiveness
of any given test methodology. Presumably, the
fault model will provide an enumeration of possible faulty behaviors along with their likelihood of
occurrence. Coverage figures for current MEMS
testing methods can then be determined through
systematic fault simulations.
Test methodology development: Any possible shortcomings in current testing methodologies will be exposed by test methodology grading. Even if shortcomings do not exist, knowledge
about MEMS faulty behavior may lead to more effective test or design techniques that reduce cost
and increase quality.
MEMS diagnosis: Finally, more effective testing methodologies will undoubtedly lead to better
diagnosis. We plan to create formal links between
observed faulty behavior, testing methods, and
process contaminations for diagnostic purposes.
Paper 36.3
929
Acknowledgements
We would like to thank Prof. Gary Fedder and Dr.
Tamal Mukherjee for their help with the basic design
and model of the microresonator. We are indebted to
Dr. Jitendra Khare and Chris Kellen for their help
in modifying CODEF. We would also like to express
our regards to Prof. Wojciech Maly for his guidance.
This work would not have been possible without their
expertise.
References
G. K. Fedder, “Future Trends in MicroElectroMechancial
Systems (MEMS)”, Memo., E C E Department, Carnegie
Mellon University, Sept. 1996.
R. E. Ham, “Design for Large Scale Integration of Mixed
Technology Microdevices”, http://eto.sysplan.com/ETO/
CompCAD, R 63 D Program of D A R P A , EIectronics Technology Ofice, 3701, North Fairfax Drive, Arlington, VA
22203-1714, 1996.
W. C. Tang, T. C. H. Nguyen, M. W. Judy and R. T .
Howe, “Electrostatic Comb Drive of Lateral Polysilicon
Resonators”, Sensors and Actuators A , Vol. 21, Nos. 1-3,
pp. 328-331, Feb. 1990.
G . K. Fedder, “Simulation of Microelectromechanical Systems”, PhD thesis, University of California a t Berkeley,
Sept. 1994.
Z. Zhang and W. C. Tang, “Viscous Air Damping in Laterally Driven Microsonators”, I n PTOC.of the IEEE Micro
Electro Mechanical Systems Workshop, pp. 199-204, Jan.
1994.
W. A. Johnson and L. K. Warne, “Electrophysics of Micromechanical Comb Acutators”, Journal of Microelectromechanical Systems, Vol. 4, No. 1, pp. 49-59, March 1995.
Analog Devices Inc., “ADXL Series Accelerometer
Datasheets”, http://www.analog.com, One Technology
Way, P.O. Box 9106, Norwood, MA 02062-9106,1996.
Motorola, “MMAS40GlOD Accelerometer Datasheet”,
http://mot-sps.com/senseon/ad.html,Semiconductor Products Sector, 6501, William Cannon Drive, Austin, T X ,
1996.
M. A. Mignardi, “Digital Micromirror Array for Projection
T V ” , Solid State Technology, Vol. 37, No. 7, pp. 63-64, July
1994.
C. T. C. Nguyen and R. T. Howe, “CMOS micromechanical
resonator oscillator”, I n Technical Digest of the IEEE Int.
Electron Devices Meeting, pp. 199-202, Washington, D.C.,
1993.
G. K. Fedder, S. Santhanam, M. L. Reed, S. Eagle, D.
Guillou, M. S.-C. Lu and L. R. Carley, “Laminated highaspect-ratio microstructures in a conventional CMOS process”, I n PTOC.of the IEEE Micro Electro Mechanical Systems Workshop, pp. 13-18, San Diego, CA, Feb. 1996.
D. A. Koester, R. Mahadevan and K. W. Markus, “Multiuser MEMS proceeses (MUMPS) introduction and design
rules” , ht tp:/ /mems.mcnc .org/mumps. html, Oct. 1994.
Sandia National Laboratories, “Intelligent MicroMachine
Initiative” http://www.mdl.sandia.gov/Micromachine,
P.O. Box 969, Livemore, CA 94551.
J. Khare and W. Maly, “From Contamination to Defects, Faults and Yield Loss”, I(1uwer Academic Publishers,
Boston, 1996.
Hibbit, Karlsson & Sorensen, Inc., “ABAQUS User Manual”, Vol. 2 , Pawtucket, RI, 1995.
Paper 36.3
930
G. K. Fedder and T. Mukherjee, “Physical Design for
Surface-Micromachined MEMS“, I n PTOC. of the 5th
A C M / S I G D A Physical Design Workshop, p p . 53-60, April
1996.
Meta-Software, “HSPICE Circuit Simulaltor” , Tech. report, http://www.metasw.com/products/sim/hspice.
html,
1300 White Oaks Road, Cambell, CA 95008.
D. M. H. Walker, C. S.Kellen and A. J. Strojwas, “The
PREDITOR process editor and statistical simulator” I n
Proc. of the 1991 International Workshop on VLSI PTOcess and Device Modeling, pp. 120-123, May 1991.
Appendix
Table 3 lists the complete process recipe for the
MCNC MUMPS process [12] described in the PREDI-
TOR format [18].
Step #
Name
1
2
Initialize
Nitride Deposit
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Deposit PolyO
Litho SpinOn
Litho Expose
Litho Develop
Etch PolyO
Etch PolyO Resist
Delposit Oxidel
Litho SpinOn
Litho Expose
Litlho Develop
Etch Oxidel
Etch Dimple1 Resist
Litlho SpinOn
Litho Expose
Litho Develop
Etch Oxidel
Etch Anchor1 Resist
Deposit Polyl
Litho SpinOn
Litho Expose
Litho Develop
Etch Polyl
Etclh Polyl Resist
Deplosit Oxide2
Litho SpinOn
Lithto Expose
Lithto Develop
Etch Oxide2
Etch Oxide2 Resist
Litho SpinOn
Litho Expose
Litho Develop
Etch Oxide12
Etch Anchor2 Resist
Deposit Poly2
Litho SpinOn
Litho Expose
Litho Develop
Etch Poly2
Etchb Poly2 Resist
Deposit Metal
Litho SpinOn
Litho Expose
Litho Develop
Etch Metal Resist
Release
-
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
___
Description
Heavy doping of wafer surface using phoshous (POCl3)
Deposition of 600 nm silicon nitride layer using a low stress,
low pressure chemical vapor deposition (LPCVD)
Deposition of 500 nm polysilicon f
ilm (polyo) using LPCVD
Coating of wafer with photoresist (PR)
Exposure of PR with CPZ mask
P R development
Etching of poly0 layer with reactive ion etch (RIE)
PR strip
Deposition of phosphosilicate glass (PSG), oxide1 by LPCVD
Coahing of wafer with PR
Exposur,~
of PR with COS mask
PR development
Etching (of PSG layer with RIE to form DIMPLES
P R strip
Coating of wafer with the PR
Exp~osurc:of PR with COF mask
PR devellopment
Etching of PSG layer in RIE to form ANCHORS (anchorl)
P R strip
Deposition of 2 pm of polysilicon blanket (polyl) using LPCVD
Coating of wafer with PR
Exposure of PR with CPS mask
P R development
Etching of Polyl layer in RIE
P R strip
Depositicln of 0.75 pm of PSG
Coating of wafer with PR
Exposure of P R with COT mask
PR development
Etching cd oxide2 with RIE to form polyl-poly2 vias
P R strip
Coating cd wafer with PR
Exposure of PR with COL mask
PR development
Etching of oxidel and oxide2 with RIE to form ANCHORS (anchor2)
P R strip
Deposition of 1.5 pm of polysilicon blanket (poly2)
Coating of wafer with PR
Exposure of PR with CPT mask
PR development
Etching of Poly2 with RIE
PR strip
Depositioin of metal (gold with adhesion layer)
Coating of wafer with PR
Exposure of PR with CCM mask
P R dievelopment
Removing the unwanted metal and PR in a solvent bath
Release of sacrificial PSG layer by immersing in 49% HF solution
~-
I'able 3: The complete MCNC MU'MPis process described in the PREDITOR format [18].
Paper 36.3
931