MASTER OF SCIENCE
SUPERVISOR LEE. JAICHAN
S
IMULATION
A
ND
F
ABRICATION
O
F
P
IEZOELECTRIC
M
EMS
I
NKJET
P
RINT
H
EAD


P
IEZOELECTRIC
M
EMS
I
NKJET
P
RINT
H
EAD A Thesis Presented
by
PHAM VAN SO Submitted to the Graduate School of SungKynKwan University
in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
in
Materials Science and Engineering June 2007
Department of Materials Science and Engineering
Graduate School of SungKynKwan University


Microelectromechanical systems (MEMS) have played an increasingly important
role in sensor and actuator applications. And its key contribution is that it has enabled
the integration of multi-components (i.e., electronics, mechanics, fluidics and etc) on a
single chip and their integration has positive effects upon performance, reliability and
cost. Compared to conventional electrostatic, thermal or magnetic actuating schemes,
piezoelectric MEMS inkjet has the advantages of lower power consumption, lower
voltage operation and relatively larger driving force.
Based on the primary design and fabrication of piezoelectric MEMS inkjet (1
st

version-InkjetVer1) done in our STD Lab, the computer simulation and validation of
inkjet have been investigated, and then the 2
nd
version (InkjetVer2) with the modified
nozzle shape was fabricated and characterized.
In details, firstly the simulation of piezoelectric MEMS inkjet with the electro-
mechanical-fluid interaction has been performed. In order to verify the simulation
results, a fabrication and characterization of actuator part consisting of PZT-based
actuating membrane and ink chamber was carried out. These treatments are to
determine how much “dynamic force”, in terms of membrane’s maximum displacement,
maximum force and driving frequency, can be produced by the actuator membrane.
Secondly, a simulation of microdroplet generation in inkjet has also been done. This
work gives an understanding about the droplet generation process, and the effects of
driving characteristics, fluid properties and geometrical parameters on droplet
generation. Especially, this simulation helps to predict how much “dynamic force” is
required to generate mirodroplets. The combination of both results (i.e., how much
“dynamic force” produced and required) gives an effective guideline in designing inkjet
structure. Thirdly, in the experimental work, the fabrication of InkjetVer2 was carried
out based on MEMS techniques. And then its electrical, mechanical characteristics as

invaluable help during my MSc course. And my thanks send to my friends in SKKU,
N.T.N. Thuy, N.T. Tien, N.T. Xuyen and N.D.T. Anh, for their helpful discussion and
argument about my results.
Finally, I want to thank my parents and relatives for their constant encouragement
and support.
iii
D
EDICATION

To my parents
Mr. Pham Van Vinh and Mrs. Le Thi Anh

iv
Table of contents

A
BSTRACT
......................................................................................................................... i

E
XPERIMENTAL
S
TUDY
O
N
A
CTUATOR PERFORMANCE
O
F
P
IEZOELECTRIC
M
EMS
I
NKJET
P
RINT
H
EAD
........................................................11
2.1 Introduction ......................................................................................................... 12
2.2 Modeling and simulation settings........................................................................ 13
2.3 Experimental procedure....................................................................................... 16
2.4 Results and discussion......................................................................................... 17
2.4.1 Performance characteristics of PIPH actuator in air................................... 17
2.4.2 Performance characteristics of PIPH actuator in liquid.............................. 18
2.5 Conclusion........................................................................................................... 20
2.6 References ........................................................................................................... 21
C

3.3.2 Effect of actuating characteristics ................................................................. 29
3.3.3 Effect of fluid properties ............................................................................... 30
3.3.4 Effect of geometrical parameters .................................................................. 32
3.4. Conclusion.......................................................................................................... 32
3.5 References ........................................................................................................... 34
C
HAPTER
4.

F
ABRICATION
A
ND
C
HARACTERIZATION
O
F
P
IEZOELECTRIC
M
EMS
I
NKJET
P
RINT
H
EAD
...................................................................................................... 38
4.1 Introduction ......................................................................................................... 39
4.2 Experiments......................................................................................................... 39

Fig. 1-3. Structure of PZT unit cell: (a) Cubic (T≥T
c
) an (b) tetragonal (T< T
c
)............. 4
Fig. 1-4. Phase diagram for the PbZrO
3
-PbTiO
3
system. C: Cubic, T: Tetragonal, R
I
:
Rhombohedral (high temp form), R
II
: Rhombohedral (low temp form), A:
rthorhombic, M: MPB, and T
c:
Curie temperature............................................ 4
Fig. 1-5. Deformation mode of piezoelectric inkjet actuator: (a) squeeze, (b) bend, (c)
push and (d) shear mode.................................................................................... 6
Fig. 1-6. A typical approach to MEMS application from concept to devices................... 7
Fig. 1-7. Steps of overall solution procedure.................................................................... 8
Fig. 1-8. Modeling settings for design of piezoelectric MEMS inkjet. Computations are
performed using CFD-ACE+ package software. .............................................. 9

Fig. 2-1. Model of a piezoelectric inkjet print head (PIPH) structure: (a)
design and (b) CFD-ACE+ symmetric model with meshing grids................. 23
Fig. 2-2. Flowchart of fabrication process (a) and SEM images (b) of PIPH actuator... 23
Fig. 2-3. Maximum displacement of PIPH actuator membrane (300 um): (a) simulation
and (b) experiment. Simulation was extended with membrane width of 500-

and (b) viscosity. High surface tension or viscosity makes cohesive forces
predominant..................................................................................................... 36
Fig. 3-6. Geometrical parameters: (a) relative chamber X1/X2, (b) aspect ratio d/h and
(c) diffuser. ...................................................................................................... 37
Fig. 3-7. Time duration for droplet generation vs.: (a) relative chamber size (A-type) and
(b) aspect ratio (B-type & C-type). ................................................................. 37
Fig. 3-8. Time duration for droplet generation vs. driving characteristics of the selected
structure (B-type). Microdroplet can be generated at an applied voltage of 9V-
21V and frequency above 15 kHz................................................................... 37

Fig. 4-1. Schematic of piezoelectric inkjet print head structure (side view): (a) Inkjet
version 1 and (b) Inkjet version 2 with the modified nozzle shape at locations
marked 1 &2.................................................................................................... 45
Fig. 4-2. Masks used for fabrication of PIPH : M1-M6 (wafer 1) and M7- M10 (wafer2).
......................................................................................................................... 45
Fig.4-3. Fabrication process flow of PIPH: (a) wafer 1-actuator and chamber and (b)
wafer 2-channel and nozzle. Both wafers are bonded by Eutectic bonding
method............................................................................................................. 46
Fig. 4-4. SEM and optical micrographs of the fabricated PIPH structure...................... 47
Fig. 4-5. Preparing for ejection test: (a) 4-inkjet heads on 1 cell and (b) PCB-wire
bonding and tube attachment........................................................................... 48
Fig. 4-6. Ejection testing by high speed digital camera system. .................................... 49
Fig. 4-7. Meniscus vibration under an applied voltage of 10V-40 kHz. ........................ 49

Fig. 5-1. Model of InkjetVer3 (3-silicon wafers)........................................................... 51
Fig. 5-2. Masks used for fabrication of InkjetVer3........................................................ 51

viii
List of tables


The objective of this thesis is to investigate the piezoelectric MEMS inkjet print
head from design to fabrication. Therefore, this chapter will briefly summarize the
background of piezoelectricity, types of piezoelectric MEMS inkjet head and general
principle of numerical simulation.

2
1.1 Piezoelectricity
1.1.1 Piezoelectric effect
All polar crystals show piezoelectricity, since any mechanical stress T will result in
strain because of the elastic properties of the materials. And the strain will affect the
polarization since the polarization is caused by a displacement of the charge centers of
the anions and cations. For small changes of the stress T, the relation
P=d.T
is called the direct piezoelectric effect, where d denotes the piezoelectric coefficient.
Once a force (mechanical stress) is applied to a piezoelectric material, surface charge is
induced by the dielectric displacement and therefore an electric field is built up. On
applied electrodes this field can be tapped as electrical voltage (Fig. 1-1. (a)). If the
electrodes are shorted, the surface charge balances out by a current ((Fig. 1-1. (b)). The
direct piezoelectric effect is employed for mechanical sensors.
Fig.1-1. Direct piezoelectric effect in open circuit (a) and in shorted circuit (b).
Because of the piezoelectric property of polar materials, a converse effect is
observed. If an external electrical field, E is applied, a strain
S=d.E
is observed. If this strain is prevented (blocking totally or partially the material), an
elastic tension T occurs. A force F is thereby applied to the device, which prevent to the
distortion of the piezoelectric body (Fig. 1-2.(a)). In practice, the converse piezoelectric
effect is used in static as well as dynamic operation (Fig. 1-2. (b)) and it is used for

3
mechanical actuators. The first experimental work on piezoelectricity was performed by

ions are moved from the
center of the cube (Fig.1-3(b)). This results in a dipole and a structure that is no longer
cubic but rather than tetragonal.
The significant feature of PZT is its phase diagram, which is characterized by a

4
boundary, known as the morphotropic phase boundary (MPB, i.e. the boundary between
rhombohedral and tetragonal phases at PbZr
0.52
Ti
0.48
O
3
). Figure 1-4 shows the phase
diagram for the PbZrO
3
-PbTiO
3
system. PZT compositions were developed for
moderate power applications. PZT compositions have a low loss tangent resulting in
low power losses as well as high distortion constant and high Curie point. Compositions
near MPB have the largest piezoelectric constants and dielectric constants. This
enhancement is a result of enhanced polarizability arising from the coupling between
two equivalent energy states, i.e. the tetragonal and rhombohedral phases, allowing
optimum domain reorientation during the poling process. In other word, it is due to the
greater ease of polarization near MPB.

Fig. 1-3. Structure of PZT unit cell: (a) Cubic (T≥T
c
) an (b) tetragonal (T< T

i

where d
ij
is the piezoelectric coefficient, P
i
the remnant polarization on poling, k the
dielectric constant, Q
ij
the electrostriction coefficient. Since both Q
ij
and P
i
exhibit little
dependence on composition at temperature below T
c
in ferroelectric ceramics such as
PZT, the d
ij
and k are interrelated, i.e. a ceramic with high piezoelectric coefficient also
exhibits a large dielectric constant. To achieve a high dielectric constant or piezoelectric
coefficient, MPB based ceramics can be further engineered by compositionally
adjusting the T
c
downward relative low temperatures. The lower the T
c
, the higher the
dielectric constant is [4,5].
PZT is one of extremely important materials in mechatronics and MEMS
applications. In its bulk form, PZT is not readily amenable for integration into silicon


Fig. 1-5. Deformation mode of piezoelectric inkjet actuator: (a) squeeze, (b) bend, (c)
push and (d) shear mode.
Piezoelectric MEMS inkjet structures used the piezoelectric thin film as main
component of actuator part which has forcing function, and were integrated fluidic
components such as ink chamber, channel and ink reservoir. In the case of inkjet heads,
print resolution is one of the primary measures of product performance. When printing
with inkjet heads, smaller, more tightly spaced droplets of ink result in sharper print
quality. However, this produces a smaller print area, resulting in an increased printing
time. To optimize both print quality and speed, a printer head must deliver an increased
number of smaller-sized droplets, droplets; this equates directly to an increase in the
density of holes per inch in an inkjet head [7]. Therefore, the contribution of MEMS
technology in fabricating inkjet heads is not only an integration of multi-components as
well as their micro-scales but also an increase of numerous nozzles or array of nozzles
(based on deep reactive ion etching (DRIE) technique).

7
1.3 Numerical simulation
1.3.1 Role of numerical simulation
A typical approach to MEMS application from concept to devices can be shown in
Fig. 1-6. The approach consists of several steps such as specifications of MEMS device,
design, modeling to evaluate performance, fabrication and testing. Reviews of the
modeling and test results enable optimization of the performance of the MEMS device.

Fig. 1-6. A typical approach to MEMS application from concept to devices.
The process of developing a MEMS application starts with determining the
specifications for MEMS device (i.e., inkjet print head). The specifications come from
the nature of physical phenomenon, mechanism of operating principle and etc. With the
specifications in place, the next step is design which is the most important step in the
flow sheet from concept to prototype. Depending on the complexity of the design, it’s

Fig. 1-7. Steps of overall solution procedure.

9

1.3.3 Numerical simulations of piezoelectric MEMS inkjet with CFD-ACE+
Piezoelectric MEMS inkjet print head is a complex device integrated electro-
mechanical-fluidic components. The modeling, therefore, was performed in a separated
way consisting of simulating (1) actuator performance and (2) droplet generation. The
former concentrates on analyzing the piezoelectric actuator characteristics (i.e., driving
force, displacement and frequency) with electro-mechanical-fluidic couplings. The fluid
flow is driven by the vibration of the thin plate and the flow also imposes a resistance to
this vibration. Thus, vibrations of the plate and the fluid flow are inherently coupled.
The vibration characteristics of actuator membrane of piezoelectric MEMS inkjet head
highly depend on this fluid-structure coupling. Therefore, fluid-structure interaction is
one of the primary concerns when studying piezoelectric inkjet head. The later
considers the generation of microdroplets which is influenced by a strong competition
between cohesive and disruptive forces (i.e., driving force, viscous force and surface
tension force). The simulations have been performed by CFD-ACE+ package software
known as a multiphysics modeling tool. The details of simulation settings will be
described in chapter 2 and chapter 3. The goals of these simulations are indicated in Fig.
1-8.

Fig. 1-8. Modeling settings for design of piezoelectric MEMS inkjet. Computations are
performed using CFD-ACE+ package software.

10
1.4 References

A
ND
E
XPERIMENTAL
S
TUDY
O
N
A
CTUATOR PERFORMANCE
O
F
P
IEZOELECTRIC
M
EMS

I
NKJET
P
RINT
H
EAD


Various designs and fabrication techniques of piezoelectric inkjet print head (PIPH)
have been reported in the literatures [4-7]. In this study, we designed a PIPH structure
which could be fabricated from two silicon wafers using MEMS processing. However,
depending on the complexity of PIPH structure integrating actuator-chamber-reservoir
components, it’s difficult to predict the performances of inkjet intuitively. One of
important parts of a PIPH is actuator component which exhibits the ability of PIPH to
work and eject the ink droplets. Therefore, a numerical simulation of PIPH to analyze
its actuator performance has been performed prior to fabrication. In order to verify
simulation results, a fabrication of PIPH’s actuator component was also carried out
before the whole PIPH structure was fabricated. In addition, this treatment helps to
understand physical phenomena in interaction between solid and fluid components such
as piezoelectric actuator and ink chamber. Fan et al revised that no complete coupling
study would lead to that the flow rate increased continuously with increasing the
frequency even at extremely high frequencies [4]. Therefore, structural-fluidic
interaction is one of primary concerns when studying actuator performance of inkjet
head structure. In simulation work, the finite element method (FEM) and computational
fluid dynamics (CFD) with finite volume method (FVM) are employed to study the
membrane-fluid coupling. In experimental work, the piezoelectric actuator component

13
was fabricated using MEMS processing. Then, its characteristics including
displacement and resonance frequency were also monitored using a LK-G10-
KEYENCE non-contact laser displacement measurement and a HP4194A impedance
analyzer, respectively. This paper addresses how the maximum displacement, maximum
force and driving frequency affect to the PIPH actuator performance. Simulation results
in air and in liquid agree well with experimental ones.
2.2 Modeling and simulation settings
The inkjet structure consists of an ink chamber connected directly with a nozzle and
a reservoir. This chamber is covered by a multi-layer membrane which consists of a
primary piezoelectric actuator layer (PZT) and other secondary layers including an

flow. Therefore, the Navier-Stokes equations and the mass continuity equation become
[11]
PVg
Dt
VD
LL
∇−∇+=
r
r
r
2
μρρ
(1)
0).( =∇+
∂
∂
L
L
V
t
ρ
ρ
r
(2)
where
),,( wvuV =
r
is velocity vector, ρ
L
is density of liquid, μ is viscosity of liquid.

E
ijkl
is elastic stiffness constant tensor at constant electric field . C
E
ijkl

is a 6x6 symmetric tensor.
Piezoelectric material used in simulation and fabrication is PZT with ratio
Zr:Ti=0.52:0.48. Its properties are listed in Table 2-2.
According to the elastic plate theory [13], the equation of displacement of
membranes is given by