Actuator Materials for Small-scale Devices
The fabrication of a microsystem requires the
processing of a large variety of different materials
in order to realize their different electronic, sensor,
and actuator components. The fabrication processes
of these materials should be compatible with micro-
system technologies, i.e., with silicon micromachin-
ing. One possibility for the realization of sensors and
actuators in microsystems is the use of ‘‘smart’’ ma-
terials that directly transduce electrical, magnetic, or
thermal energy into mechanical energy or vice versa.
The related physical effects are the piezoeffect, mag-
netostriction, or the shape memory effect.
In general, thin film technologies are an attractive
approach to integrate these materials into microsys-
tems as this approach has almost no materials lim-
itations, offers easy down-scaling into the micrometer
range by a cost-effective manufacturing technology,
avoids assembly and interconnection processes, al-
lows the protection of the microsystem by protective
layers (Quandt and Holleck 1998), and provides the
possibility to design new materials as, for example,
metastable phases, multilayers or gradient layers.
Physical vapor deposition (PVD) techniques, in par-
ticular magnetron sputtering, offer the opportunity of
tailoring the constitution of the thin films. Addition-
ally, the interaction with the substrate enables fabri-
cation of highly textured films which, in the case of
piezoelectric materials, show improved properties
compared to conventional polycrystalline bulk mate-
rials. Furthermore, it can be expected that the inte-
grated sensing features of smart film actuators—e.g.,
the resistive effects present in all three classes of
materials—will provide means to realize ‘‘intelligent’’
actuators by using thin film ‘‘smart’’ materials.
Thin film actuators in the case of piezoelectric or
magnetostrictive materials are always used as com-
pounds with micromachined substrates working as
bending transducers, whereas in the case of shape
memory thin films both free-standing thin film actu-
ators and compounds with metallic or silicon subst-
rates are realized. Since free-standing shape memory
films have to be trained for the two-way effect to
introduce internal stresses into the material, which is
quite complicated in small dimensions, the film sub-
strate compounds represent the generally more im-
portant approach for all smart film materials. In this
case the materials development target has to be modi-
fied compared to bulk actuator materials. Whereas in
bulk materials the strain has to be maximized, the
resulting bending stress of a thin film substrate com-
pound is proportional to the strain and Young’s
modulus of the film. As a consequence, this stress
of the thin film actuator material should be as large
as possible. In the following sub-sections the main
results for magnetostrictive, piezoelectric, and shape
memory thin films for actuators will be discussed in
view of their main application areas.
1. Magnetostrictive Films
The development of room temperature giant magne-
tostrictive bulk and thin film materials is based on the
rare earth Fe
2
Laves phase, while at low temperatures
the highest strains are found in hexagonal Tb
x
Dy
1x
and body-centered Tb
x
Dy
1x
Zn alloys. The giant
magnetostrictive materials were optimized in terms of
their magnetostriction to magnetic anisotropy ratio
in order to attain large strains at reasonable magnetic
fields (see Magnetostrictive Materials). Different ap-
proaches have been taken based on the Laves phases
(Tb,Dy)(Fe,Co)
2
with positive or Sm(Fe,Co)
2
with
negative magnetostriction. The fabrication of these
rare earth based materials is restricted to PVD meth-
ods, the most prominent being magnetron sputtering.
Because the magnetic saturation field is propor-
tional to the magnetic anisotropy constant and in-
versely proportional to the saturation magnetization,
its reduction is achieved either by decreasing the
magnetic anisotropy constant and/or by increasing
the saturation magnetization. To reduce the macro-
scopic anisotropy, amorphous rare earth–iron thin
film materials have been fabricated in a wide com-
position range. It was found that increasing the rare
earth content compared to the Laves phase compo-
sition results in the highest low-field magnetostriction
in amorphous thin films. For amorphous TbFe films
with an inplane magnetic easy axis an important im-
provement of the low-field magnetostriction and the
hysteresis can be observed in comparison to nano-
crystalline (Tb,Dy)Fe
2
films.
Assuming the same local environment in the amor-
phous compared to the crystalline state, a further
approach to lower the remaining anisotropy is by Tb/
Dy substitution, which was investigated over a wide
composition range for rare earth–iron or –cobalt thin
film materials. This approach leads to a further re-
duction in the magnetic saturation field, but the lower
Curie temperature due to the dysprosium alloying
significantly reduces the saturation magnetostriction,
resulting in negligible gain in low-field magneto-
strictive strain of these ternary films.
The second route to lower the macroscopic aniso-
tropy is by increasing the saturation magnetization.
The magnetostrictive thin films discussed so far have
rather low magnetizations, owing to their ferri-
magnetic nature. Unfortunately, an increase of the
saturation magnetization using amorphous homoge-
neous rare earth–transition metal materials is impos-
sible, because the compositions of interest contain
A
1