Article
Reference
Physics of thin-film ferroelectric oxides
DAWBER, Matthew, RABE, K. M., SCOTT, J. F.
Abstract
This review covers important advances in recent years in the physics of thin-film ferroelectric
oxides, the strongest emphasis being on those aspects particular to ferroelectrics in thin-film
form. The authors introduce the current state of development in the application of ferroelectric
thin films for electronic devices and discuss the physics relevant for the performance and
failure of these devices. Following this the review covers the enormous progress that has
been made in the first-principles computational approach to understanding ferroelectrics. The
authors then discuss in detail the important role that strain plays in determining the properties
of epitaxial thin ferroelectric films. Finally, this review ends with a look at the emerging
possibilities for nanoscale ferroelectrics, with particular emphasis on ferroelectrics in
nonconventional nanoscale geometries.
DAWBER, Matthew, RABE, K. M., SCOTT, J. F. Physics of thin-film ferroelectric oxides.
Reviews of Modern Physics, 2005, vol. 77, no. 4, p. 1083-1130
DOI : 10.1103/RevModPhys.77.1083
Available at:
http://archive-ouverte.unige.ch/unige:126367
Disclaimer: layout of this document may differ from the published version.
1 / 1
Physics of thin-film ferroelectric oxides
M. Dawber
*
DPMC, University of Geneva, CH-1211, Geneva 4, Switzerland
K. M. Rabe
†
Department of Physics and Astronomy, Rutgers University, Piscataway, New Jersey 00854-
8019, USA
J. F. Scott
‡
Department of Earth Sciences, University of Cambridge, Cambridge CB2 3EQ, United
Kingdom
共Published 17 October 2005兲
This review covers important advances in recent years in the physics of thin-film ferroelectric oxides,
the strongest emphasis being on those aspects particular to ferroelectrics in thin-film form. The
authors introduce the current state of development in the application of ferroelectric thin films for
electronic devices and discuss the physics relevant for the performance and failure of these devices.
Following this the review covers the enormous progress that has been made in the first-principles
computational approach to understanding ferroelectrics. The authors then discuss in detail the
important role that strain plays in determining the properties of epitaxial thin ferroelectric films.
Finally, this review ends with a look at the emerging possibilities for nanoscale ferroelectrics, with
particular emphasis on ferroelectrics in nonconventional nanoscale geometries.
CONTENTS
I. Introduction 1084
II. Ferroelectric Electronic Devices 1084
A. Ferroelectric memories 1084
B. Future prospects for nonvolatile ferroelectric
memories 1086
C. Ferroelectric field-effect transistors 1087
D. Replacement of gate oxides in DRAMs 1088
III. Ferroelectric Thin-Film-Device Physics 1089
A. Switching 1089
1. Ishibashi-Orihara model 1089
2. Nucleation models 1089
3. The scaling of coercive field with thickness 1090
4. Mobility of 90° domain walls 1090
5. Imaging of domain-wall motion 1090
B. Electrical characterization 1092
1. Standard measurement techniques 1092
a. Hysteresis 1092
b. Current measurements 1093
c. Dielectric permittivity 1093
2. Interpretation of dielectric permittivity data 1093
a. Depletion charge versus intrinsic response 1093
b. Domain-wall contributions 1094
c. Dielectric measurements of phase transitions 1094
3. Schottky barrier formation at
metal-ferroelectric junctions 1095
4. Conduction mechanisms 1097
a. Schottky injection 1097
b. Poole-Frenkel 1098
c. Fowler-Nordheim tunneling 1098
d. Space-charge-limited currents 1098
e. Ultrathin films—direct tunneling 1098
f. Grain boundaries 1098
C. Device failure 1099
1. Electrical breakdown 1099
2. Fatigue 1100
3. Retention failure 1102
IV. First Principles 1102
A. Density-functional-theory studies of bulk
ferroelectrics 1102
B. First-principles investigation of ferroelectric thin
films 1104
1. First-principles methodology for thin films 1105
2. Overview of systems 1107
3. Studies of individual one-component
systems 1108
a. BaTiO
3
1108
b. PbTiO
3
1110
c. SrBi
2
Ta
2
O
9
1111
d. SrTiO
3
and KTaO
3
1111
4. Studies of individual heterostructures 1112
5. First-principles modeling: methods and
lessons 1113
6. Challenges for first-principles modeling 1115
V. Strain Effects 1116
VI. Nanoscale Ferroelectrics 1121
A. Quantum confinement energies 1121
B. Coercive fields in nanodevices 1121
C. Self-patterned nanoscale ferroelectrics 1121
D. Nonplanar geometries: ferroelectric nanotubes 1122
VII. Conclusions 1124
References 1124
*
Electronic address: matthew.dawber@physics.unige.ch
†
Electronic address: rabe@physics.rutgers.edu
‡
Electronic address: jsco99@esc.cam.ac.uk
REVIEWS OF MODERN PHYSICS, VOLUME 77, OCTOBER 2005
0034-6861/2005/77共4兲/1083共48兲/$50.00 ©2005 The American Physical Society1083
I. INTRODUCTION
The aim of this review is to provide an account of the
progress made in the understanding of the physics of
ferroelectric thin-film oxides, particularly the physics rel-
evant to present and future technology that exploits the
characteristic properties of ferroelectrics. An overview
of the current state of ferroelectric devices is followed
by identification and discussion of the key physics issues
that determine device performance. Since technologi-
cally relevant films for ferroelectric memories are typi-
cally thicker than 120 nm, characterization and analysis
of these properties can initially be carried out at compa-
rable length scales. However, for a deeper understand-
ing, as well as for the investigation of the behavior of
ultrathin films with thickness on the order of lattice con-
stants, it is appropriate to redevelop the analysis at the
level of atomic and electronic structure. Thus, the sec-
ond half of this review is devoted to a description of the
state of the art in first-principles theoretical investiga-
tions of ferroelectric-oxide thin films, concluding with a
discussion of experiment and theory of nanoscale ferro-
electric systems.
As a starting point for the discussion, it is helpful to
have a clear definition of ferroelectricity appropriate to
thin films and nanoscale systems. Here we consider a
ferroelectric to be a pyroelectric material with two or
more stable states of different nonzero polarization. Un-
like electrets, ferroelectrics have polarization states that
are thermodynamically stable, not metastable. Further-
more, it must be possible to switch between the two
states by the application of a sufficiently strong electric
field, the threshold field being designated the coercive
field. This field must be less than the breakdown field of
the material, or the material is merely pyroelectric and
not ferroelectric. Because of this switchability of the
spontaneous polarization, the relationship between the
electric displacement D and the electric field E is hyster-
etic.
For thin-film ferroelectrics the high fields that must be
applied to switch the polarization state can be achieved
with low voltages, making them suitable for integrated
electronics applications. The ability to create high-
density arrays of capacitors based on thin ferroelectric
films has spawned an industry dedicated to the commer-
cialization of ferroelectric computer memories. The clas-
sic textbooks on ferroelectricity 共Fatuzzo and Merz,
1967; Lines and Glass, 1967兲 though good, are now over
20 years old, and predate the shift in emphasis from bulk
ceramics and single crystals towards thin-film ferroelec-
trics. While much of the physics required to understand
thin-film ferroelectrics can be developed from the under-
standing of bulk ferroelectrics, there is also behavior
specific to thin films that cannot be readily understood in
this way. This is the focus of the present review.
One of the points that will become clear is that a
ferroelectric thin film cannot be considered in isolation,
but rather the measured properties reflect the entire sys-
tem of films, interfaces, electrodes, and substrates. We
also look in detail at the effects of strain on ferroelec-
trics. All ferroelectrics are grown on substrates which
can impose considerable strains, meaning that properties
of ferroelectric thin films can often be considerably dif-
ferent from those of their bulk parent material. The
electronic properties also have a characteristic behavior
in thin-film form. While bulk ferroelectric materials are
traditionally treated as good insulators, as films become
thinner it becomes more appropriate to treat them as
semiconductors with a fairly large band gap. These ob-
servations are key to understanding the potential and
the performance of ferroelectric devices, and to under-
standing why they fail when they do.
In parallel with the technological developments in the
field, the power of computational electronic structure
theory has increased dramatically, giving us new ways of
understanding ferroelectricity. Over the last 15 years,
more and more complex systems have been simulated
with more accuracy; and as the length scales of experi-
mental systems decrease, there is now an overlap in size
between the thinnest epitaxial films and the simulated
systems. It is therefore an appropriate and exciting time
to review this work, and to make connections between it
and the problems considered by experimentalists and
engineers.
Finally, we look at some issues and ideas in nanoscale
ferroelectrics, with particular emphasis on new geom-
etries for ferroelectric materials on the nanoscale such
as ferroelectric nanotubes and self-patterned arrays of
ferroelectric nanocrystals.
We do not attempt to cover some of the issues which
are of great importance but instead refer readers to re-
views by other authors. Some of the more important
applications for ferroelectrics make use of their piezo-
electric properties, for example, in actuators and mi-
crosensors; this topic has been reviewed by Muralt
共2000兲. Relaxor ferroelectrics in which ferroelectric or-
dering occurs through the interaction of polar nano-
domains induced by substitution are also of great inter-
est for a number of applications and have recently been
reviewed by Samara 共2003兲.
II. FERROELECTRIC ELECTRONIC DEVICES
A. Ferroelectric memories
The idea that electronic information can be stored in
the electrical polarization state of a ferroelectric mate-
rial is a fairly obvious one; however, its realization is not
so straightforward. The initial barrier to the develop-
ment of ferroelectric memories was the necessity of
making them extremely thin films because the coercive
voltage of ferroelectric materials is typically of the order
of several kV/cm, requiring submicron thick films to
make devices that work on the voltage scale required for
computing 共all Si devices work at 艋5V兲. With today’s
deposition techniques this is no longer a problem, and
now high-density arrays of nonvolatile ferroelectric
memories are commercially available. However, reliabil-
ity remains a key issue. The lack of good device models
means that the design of ferroelectric memories is ex-
1084
Dawber, Rabe, and Scott: Physics of thin-film ferroelectric oxides
Rev. Mod. Phys., Vol. 77, No. 4, October 2005
pensive and that it is difficult to be able to guarantee
that a device will still operate ten years into the future.
Because competing nonvolatile memory technologies
exist, ferroelectric memories can succeed only if these
issues are resolved.
A ferroelectric capacitor, while capable of storing in-
formation, is not sufficient for making a nonvolatile
computer memory. A pass-gate transistor is required so
that a voltage above the coercive voltage is only applied
to the capacitor when a voltage is applied to both the
word and bit line; this is how one cell is selected from an
array of memories. The current measured through a
small load resistor in series with the capacitor is com-
pared to that from a reference cell that is poled in a
definite direction. If the capacitor being read is in a dif-
ferent state, the difference in current will be quite large
where the displacement current associated with switch-
ing accounts for the difference. If the capacitor does not
switch because it is already in the reference state, the
difference in current between the capacitor being read
and the reference capacitor is zero.
Most memories use either a 1 transistor–1 capacitor
共1T-1C兲 design or a 2 transistor–2 capacitor 共2T-2C兲 de-
sign 共Fig. 1兲. The important difference is that the 1T-1C
design uses a single reference cell for the entire memory
for measuring the state of each bit, whereas in the 2T-2C
there is a reference cell per bit. A 1T-1C design is much
more space effective than a 2T-2C design, but has some
significant problems, most significantly that the refer-
ence capacitor will fatigue much faster than the other
capacitors, and so failure of the device occurs more
quickly. In the 2T-2C design the reference capacitor in
each cell fatigues at the same rate as its corresponding
storage capacitor, leading to better device life. A prob-
lem with these designs is that the read operation is de-
structive, so every time a bit is read it needs to be writ-
ten again. A ferroelectric field-effect transistor, in which
a ferroelectric is used in place of the metal gate on a
field-effect transistor, would both decrease the size of
the memory cell and provide a nondestructive readout;
however, no commercial product has yet been devel-
oped. Current efforts seem to run into serious problems
with data retention.
An example of a real commercially available memory
is the Samsung lead zirconate titanate-based 4 Mbit
1T-1C ferroelectric memory 共Jung et al., 2004兲. The scan-
ning electron microscopy cross section 共Fig. 2兲 of the
device gives some indication of the complexity of design
involved in a real ferroelectric memory.
Lead zirconate titanate 共PZT兲 has long been the lead-
ing material considered for ferroelectric memories,
though strontium bismuth tantalate 共SBT兲, a layered
perovskite, is also a popular choice due to its superior
fatigue resistance and the fact that it is lead-free 共Fig. 3兲.
However, it requires higher-temperature processing,
which creates significant integration problems. Recently
progress has been made in optimizing precursors. Until
recently the precursors for Sr, Bi, and Ta/Nb did not
function optimally in the same temperature range, but
last year Inorgtech developed Bi共mmp兲3—a 2-methoxy-
2-propanol propoxide that improves reaction and lowers
the processing temperature for SBT, its traditional main
disadvantage compared to PZT. This material also satu-
rates the bismuth coordination number at 6. Recently
several other layered perovskites, for example, bismuth
titanate, have also been considered.
FIG. 1. 共a兲 1T-1C memory design. When a voltage is applied to both the word and bit line, the memory cell is addressed. Also
shown is the voltage applied to the capacitor and the current output, depending on whether a one or a zero is stored. The current
for the zero state is pure leakage current and by comparison to a reference capacitor can be removed. 共b兲 A 2T-2C memory cell
in which the reference capacitor is part of the memory cell.
FIG. 2. Cross-sectional SEM image of the Samsung 4 Mbit
1T-1C 3 metal FRAM.
1085
Dawber, Rabe, and Scott: Physics of thin-film ferroelectric oxides
Rev. Mod. Phys., Vol. 77, No. 4, October 2005
As well as their applications as ferroelectric random
access memories 共FRAMs兲, ferroelectric materials have
potential use in dynamic random access memories
共DRAMs兲 because of their high dielectric constant in
the vicinity of the ferroelectric phase transition, a topic
which has been reviewed by Kingon, Maria, and Streif-
fer 共2000兲. Barium strontium titanate 共BST兲 is one of the
leading materials in this respect since by varying the
composition a transition temperature just below room
temperature can be achieved, leading to a high dielectric
constant over the operating temperature range.
B. Future prospects for nonvolatile ferroelectric memories
There are two basic kinds of ferroelectric random ac-
cess memories in production today: 共1兲 the free-standing
RAMs and 共2兲 fully embedded devices 关usually a CPU,
which may be a complementary metal-oxide semicon-
ductor electrically erasable programmable read-only
memory 共CMOS EEPROM兲, the current generation
widely used nonvolatile memory technology, plus a
FRAM and an 8-bit microprocessor兴. The former have
reached 4 Mbit at both Samsung 共using PZT兲 and Mat-
sushita 共using SBT兲. The Samsung device is not yet, as
far as the authors know, in commercial production for
real products, but the NEC FRAM is going into full-
scale production this year in Toyama 共near Kanazawa兲.
Fujitsu clearly leads in the actual commercial use of its
embedded FRAMs. The Fujitsu-embedded FRAM is
that used in the SONY Playstation 2. It consists of 64
Mbit of EEPROM plus 8 kbit of RAM, 128-kbit ROM,
and a 32-kbit FRAM plus security circuit. The device is
manufactured with a 0.5-
m CMOS process. The ca-
pacitor is 1.6⫻1.9
m
2
and the cell size is either
27.3
m
2
for the 2T-2C design or 12.5
m
2
for the 1T-
1C.
The leading competing technologies in the long term
for nonvolatile computer memories are FRAM and
magnetic random access memories 共MRAM兲. These are
supposed to replace EEPROMs 共electrically erasable
programable read-only memories兲 and “Flash” memo-
ries in devices such as digital cameras. Flash, though
proving highly commercially successful at the moment,
is not a long-term technology, suffering from poor long-
term endurance and scalability. It will be difficult for
Flash to operate as the silicon logic levels decrease from
5 at present to 3.3, 1.1, and 0.5 V in the near future. The
main problem for ferroelectrics is the destructive read
operation, which means that each read operation must
be accompanied by a write operation leading to faster
degradation of the device. The operation principle of
MRAMs is that the tunneling current through a thin
layer sandwiched between two ferromagnetic layers is
different depending on whether the ferromagnetic layers
have their magnetization parallel or antiparallel to each
other. The information stored in MRAMs can thus be
read nondestructively, but their write operation requires
high power which could be extremely undesirable in
high-density applications. We present a summary of the
current state of development in terms of design rule and
speed of the two technologies in Table I.
Partly in recognition of the fact that there are distinct
advantages for both ferroelectrics and ferromagnets,
there has been a recent flurry of activity in the field of
multiferroics, i.e., materials that display both ferroelec-
tric and magnetic ordering, the hope being that one
could develop a material with a strong enough coupling
between the two kinds of ordering to create a device
that can be written electrically and read magnetically. In
general multiferroic materials are somewhat rare, and
certainly the conventional ferroelectrics such as PbTiO
3
and BaTiO
3
will not display any magnetic behavior as
the Ti-O hybridization required to stabilize the ferro-
electricity in these compounds will be inhibited by the
partially filled d orbitals that would be required for mag-
netism 共Hill, 2000兲. However, there are other mecha-
nisms for ferroelectricity and in materials where ferro-
electricity and magnetism coexist there can be coupling
between the two. For example, in BaMnF
4
the ferroelec-
tricity is actually responsible for changing the antiferro-
magnetic ordering to a weak canted ferromagnetism
共Fox et al., 1980兲. In addition, the large magnetoelectric
coupling in these materials causes large dielectric
anomalies at the Néel temperature and at the in-plane
spin-ordering temperature 共Scott, 1977, 1979兲. More re-
cent theoretical and experimental efforts have focused
on BiMnO
3
, BiFeO
3
共Seshadri and Hill, 2001; Moreira
de Santos et al., 2002; Wang et al., 2003兲 and YMnO
3
共Fiebig et al., 2002; Van Aken et al., 2004兲.
FIG. 3. 共a兲 ABO
3
cubic perov-
skite structure; 共b兲 strontium
bismuth tantalate 共layered per-
ovskite structure兲.
1086
Dawber, Rabe, and Scott: Physics of thin-film ferroelectric oxides
Rev. Mod. Phys., Vol. 77, No. 4, October 2005
