Principles

When a mechanical stress is applied to an ionic crystal, in a direction that generates a separation of the centre of positive and negative charges, an electric polarization (dipole) is generated. This is the direct piezoelectric effect. Conversely, the application of an electric field gives place to a mechanical strain. Piezoelectricity is an anisotropic effect, which only takes place in some directions of the crystal.

Fig. above shows the subdivision of piezoelectric crystals into pyroelectrics, in which a spontaneous electrical polarization appears when the crystals are either heated or cooled : this spontaneous polarization changes with the temperature, following the changes of the crystal structure. Pyroelectrics in turn, are subdivided into ferroelectric crystals, whose polarization can be reversed by an electric field. Ferroelectrics present a hysteretic loop of polarization versus electric field, and a transition at a high temperature, to a non-polar, paraelectric phase.

Polycrystalline materials, such as ceramics and many of the functional thin films, have randomly oriented grains. They are isotropic and centrosymmetric materials, thus, non-piezoelectric.

Piezoelectric ceramics and thin films are built from ferroelectric crystallites, whose polarization can be oriented by an electric field, in the so-called poling process. Poling induces a cylindrical symmetry with a preferential axis, commonly referred to as 3 axis, being orthogonal to axis 1 and 2. Poled ferroelectric ceramics are non-centrosymmetric and piezoelectric.

Piezoelectric ceramic domains can be imaged using Piezoresponse Force Microscopy.

Piezoelectric ceramics and thin films, due to their ferroelectricity, are an important class of multifunctional materials whose properties have been exploited in a wide number of applications.

Additionally, ferroelectrics have high dielectric permittivity and are used in capacitors. They present non-linear properties and an electro-optic effect, that can be applied to optical storage and optical switches. The polarization reversal allows the establishment of two stable orientations under an electric field, therefore, ferroelectric materials can be used for the fabrication of computer memories (FERAMs).

Piezoelectric materials

Most piezoelectric devices are based on oxide piezoelectric materials. Lead zirconate titanate (Pb(ZrxTi1-x)O3, abbreviated PZT), is the most widely used piezoelectric ceramic. The crystalline structure of PZT and of many other useful piezoelectrics is the perovskite structure (Fig 3.5) or a derivative of perovskite. For some years, the relaxor-ferroelectrics have attracted much attention. Some of the relaxor-ferroelectrics have very high piezoelectric coefficients and, in mono crystalline form, when used in a specific configuration, they are very efficient in converting electrical energy to mechanical energy and vice versa. Although lead-free materials are preferred, the lead containing perovskites such as PZT are by far the best performing for actuators, transducers and a range of sensors.

Relaxors

A large number of lead containing perovskites, called relaxors, show anomalous characteristics, typified by a strong frequency dispersion of their permittivity.
Examples are lead magnesium niobate (PbMg1/3Nb2/3O3, PMN) and lead zinc niobate (PbZn1/3Nb2/3O3, PZN). These materials have many useful properties: a) very high permittivity over a wide temperature range, b) large electrostrictive strains, c) strong piezoelectric activity when the material is under a dc bias field, and d) strong electro-optic activity.

In the present context, greatest interest is in the first three of these characteristics.

Electrostriction in relaxors like PMN, give strain levels similar to that obtained by strong piezoelectrics like PZT, whilst the response is non-hysteretic. This is advantageous for some actuator applications. On the other hand, the electrostriction is quadratically dependent of the applied field. Therefore, for linear response, which is sometimes desired, additional electronics are needed. Another point requiring attention when using relaxors, is the strong temperature dependence of the obtained strain. However, piezoelectric activity under low E-field can still be of interest in micro-devices, where <10 V DC is sufficient to  provide the necessary bias. This feature opens up some interesting possibilities, in particular, in transducer arrays, where it is possible, in principle, to activate selectively the different elements, thus avoiding cross talk.

The high permittivity of relaxors is also advantageous for impedance matching of miniaturized transducers.

Relaxor-ferroelectrics

Solid solutions of relaxors and normal ferroelectrics, called relaxor-ferroelectrics, (for example PMN-PbTiO3, and PZN-PbTiO3),  show remarkable properties in the single crystal form. In direction <100>, PZN-PT has an exceptionally high coupling coefficient, reaching 0.94 and a d33 coefficient of 2000 pC/N in some off-polar directions. Strains as high as 1.7% have been obtained at high fields. A comparison of obtained strains is illustrative of the exceptional properties of relaxor-ferroelectric single crystals.

In the ceramic form, these materials are similar in performance to soft PZT, although the permittivity is higher.

Lead-free piezoelectrics

Replacement of PZT with lead free piezoelectric material will considerably affect many industrial applications. For example,  fuel injection actuators for engines are today based on PZT, as are nearly all probes for non-destructive testing and ultrasonic medical imaging. Replacement is not straightforward – the alternative lead-free material must possess very specific properties, be reliable and stable during exploitation, and still remain cost-effective.

Research and development of lead-free piezoelectric materials is presently amongst the hottest topics in the field of piezoelectricity.

There are two reasons for interest in lead-free piezoelectric materials, one is scientific and technological, the other is related to environment and health. The scientific challenges in developing lead free materials includes the following questions : (i) is the presence of Pb cation essential for high piezoelectricity? ; (ii) can lead-free systems exhibit properties comparable to those in lead-based systems? ; (iii) is it possible to find other mechanisms, besides MPB (morphotropic phase boundary), which would lead to significant enhancement of desirable properties ?

Presently examined lead-free systems, such as (K,Na)NbO3 and (Bi1/2Na1/2)TiO3 based materials, have properties inferior to that of lead based materials, and are difficult to process.

One huge advantage of PZT and related systems, is that this one family, with minor modifications, is suitable for nearly all applications.

The environmental and health issues related to Pb-based materials are obvious. According to RoHS, any homogeneous component containing more than 0.1 weight % of lead is subjected to restrictions. PZT and PMN-PT contain about 60 weight % Pb ! Presently, piezoelectric devices are exempted from the ban because there is no satisfactory replacement. However, reports from Japan and Europe, show that for some specific applications, it is possible to develop lead-free materials competitive to PZT. If such replacements become more widely available, PZT and related materials will have to be replaced.

The present state of the research indicates that replacement of PbTiO3-based materials is done on a case-to-case basis – this is the present day approach in Japan. That is, a lead-free composition is developed for certain niche applications (one for motors, another one for ultrasonic applications, yet another one for a certain type of actuators, and so on). It is clear that such approach is not cost efficient. Thus, concentrated scientific effort is still required in order to fully understand  the origins of high piezoelectricity, as well as technological effort towards solving the processing and reliability problems associated with lead-free piezoelectric materials.

High temperature piezoelectrics

High temperature piezoelectrics are needed in a number of key industries, including monitoring pressures in nuclear reactors, monitoring vibrations in aircraft engines and rockets, measuring pressures and vibrations in turbines, and in the materials processing industry generally.

There is great advantage in placing sensors directly at the location where a mechanical signal should be measured (eg: in the engine cylinder, for example), rather than away from it.

This is not often possible however, since, for example,  pressure sensors cannot withstand elevated temperatures –  the most widely used piezoelectric materials, based on PbTiO3 (such as PZT), are limited to temperature of around 200°C to 300°C, whilst present day requirements demand operation at temperatures up to 900°C or more.

Development of new, high temperature piezoelectric materials is exceedingly challenging: Not only must the candidate material possess stable piezoelectric properties at high temperatures and high pressures, but they must often also operate in a harsh atmosphere (e.g., low partial pressure of oxygen, to protect electrical parts). In such atmosphere, many piezoelectric materials become conductive and thus the piezoelectric signal can become noisy or lost altogether.

There are number of possible solutions to operation at high temperatures and under other extreme conditions. One approach is operation under resonance conditions, where resonance parameters are monitored as a function of the mechanical signal. This helps reduce sensitivity to conductivity changes – most often the most significant problem at high temperatures.

But the most urgent need is to explore modified and new piezoelectric materials compositions, and to reduce the susceptibility of piezoelectrics to degradation (e.g.  depoling, conductivity).

There are a few lead-free single crystal materials that are possible candidates for high temperature applications (e.g., tourmaline, quartz, GaPO4), but their sensitivity is low and single crystals are expensive. One set of alternatives are Aurivillius structure materials, based on bismuth titanate, which are not only lead-free, but also exhibit piezoelectric coefficients that are several times higher than most single crystal candidates. Boracites are also possible candidates among single crystals.

Piezoelectric single crystals

Recent discovery of large piezoelectric coefficients (d33>2000 pC/N), and high electro-mechanical coupling coefficients (k > 0.9) in single crystals of Pb(Mg,Nb)O3-PbTiO3 and Pb(Zn,Nb)O3-PbTiO3, has caused a revolution in the research and applications of piezoelectric and ferroelectric materials.

These high performance single crystal piezoelectrics exhibit excellent properties for ultrasonic imaging (better bandwidth and resolution than PZT ceramics-based probes), and for actuators (displacement several times higher, absence of strain-field hysteresis), as well as for some other applications. As the production of single crystals becomes more economical, one can expect an even wider use in medical and non-destructive testing probes, as well as in different type of actuators, scanning systems, and sensors.

Interestingly, except for few isolated cases, research and development of high performance single crystals piezoelectrics has been limited to USA, Korea and Japan, while Europe has, so far,  been left behind. Thus, Europe’s major producers of medical ultrasonic probes (e.g. Philips) are acquiring single crystal piezoelectrics in USA or Korea. However, there are still not enough producers to make supply reliable and affordable.

One disadvantage of these materials (PMN-PT, PZN-PT) is that they are Pb-based, and are thus subjected to RoHS restrictions. Lead-free alternative must therefore be investigated, and for niche applications (e.g., single element probes for high frequency ultrasonic medical imaging), such crystals are already available in Europe.

Piezoelectric polymers

Piezoelectric polymers occupy a special place among piezoelectric materials, due to their softness and the possibility of making large area thin films.

Even though many polymers can exhibit piezoelectric and even ferroelectric effect, polyvinylidene fluoride (PVDF) and its copolymers are presently the only polymers used as piezoelectric materials. The reason for this is that PVDF has relatively large piezoelectric activity, is easy to prepare, and is chemically stable.
PVDF and related copolymers are excellent materials for some applications, in which low acoustical impedance is needed (coupling to air, water, biological tissues).

But, besides these classical applications, new needs are emerging in the large area of biotechnology and research of processes within living systems.

The potential here is enormous, both in terms of the possibility of designing new materials based on biological polymers (e.g., polypeptides), as well as in using piezoelectricity and polar properties of living systems for controlling physiological functions (e.g., osteogenesis, gene expression).

It is of particular interest that many biological polymers also exhibit piezoelectric and pyroelectric effects, but little is known as to whether, or how, this electro-mechanical activity plays any physiological role. In fact, some amino acids, the building blocks of living systems, exhibit piezoelectricity, and accordingly, interest in the non-centrosymmetric properties of polymers has been renewed – most particularly because of the promise of polymers suitable for use in for example,   energy harvesting and biotechnology applications. Some polar polymers have even been investigated for possible use in memory devices.

In fact, the field of piezoelectric polymers is a highly multidisciplinary field, that will require close cooperation of materials scientists, chemists, physicists, and biologists. This field is still in its infancy and will require cooperative multidisciplinary efforts in answering questions such as : (i) what is the role, if any, of electro-mechanical activity in physiological processes ? (ii) if this role is not naturally present, can one benefit from mechanically generated charges, to control biological processes ? (iii) is it possible to design synthetic polypeptides with useful piezoelectric, dielectric and electrical properties?

High dielectric constant & high coupling piezoelectric ceramics

Current state of the art, high dielectric constant and high coupling piezoelectric ceramics, are based on perovskite Pb(ZrxTi1-x)O3 (PZT) at the morphotropic phase boundary (MPB), that occurs for x0.53;  rather, the MPB is more of a compositional region, than a simple, well defined x value.

The highest coefficients are obtained by higher valence substitution (for instance : Nb5+ for Ti4+/Zr4+). This is the basis of commercial soft PZT, with d33 coefficients up to 600 pC N-1.

There is an industrial demand for piezoelectric ceramics with dielectric constant and electro-mechanical coupling coefficients similar or higher than those of soft PZT, whilst at the same time, exhibiting lower losses. Such materials are sought after in order to enable the next generation of high sensitivity and low power devices, entering new application fields, like energy harvesting.

This search has fostered an interest in alternative MPB materials, that present higher single crystal piezoelectric coefficients than PZT. Amongst such materiials, relaxor based Pb(Zn1/3Nb2/3)O3-PbTiO3 and Pb(Mg1/3Nb2/3)O3-PbTiO3 stand out, with d33 values above 2000 pC N-1.

Ceramics of (1-x)Pb(Mg1/3Nb2/3)O3-xPbTiO3 at the MPB (x0.35) present a d33 of 700 pCN-1, and this has been further increased, up to 1500 pC N-1, by texture engineering.

Another relaxor based MPB material with similar characteristics is Pb(Ni1/3Nb2/3)O3-Pb(Zr,Ti)O3.

Until now, a large number of systems have been proposed, but ceramic processing is complicated, and their characterisation incomplete. Lead free perovskite solid solutions with MPB are also attracting increasing research activity, and systems like (Bi1/2Na1/2)TiO3-BaTiO3 and (K,Na,Li)(Nb,Ta)O3, are promising candidates to replace PZT.

Such increasingly extensive materials research efforts are of great potential.

Chemical design of MPB systems needs to be further developed in order to enlarge the spectrum of candidate materials. Reliable and up-scalable concepts for the processing of the different materials also have to be defined, most particularly for texturing of low tolerance factor perovskites, and lead-free materials.

Other major research trends include studies of grain size effects, down to the sub-micron and nano scales, that is being driven by the desire for miniaturisation of ceramic devices, and for producing thin film effects, in the context of integrated technologies like MEMs.

Dielectric and pyroelectric materials

The possibility of producing piezoelectric “bulk” ceramics lies in the possibility of breaking the symmetry properties of the ceramic, via the application of a high strength electric field (poling process). This is possible due to the ferroelectric nature of the crystallite grains, of which ceramic is composed. The ferroelectric nature of these crystallites implies an increase in the functional physical properties of the ceramic, including amongst others, pyroelectricity. In addition, ferro-piezoelectric compounds also present characteristics of high dielectric constant (K) due to the strong lattice polarizability, as well as an increased polarizability due to the presence of domain walls.

Dielectric materials

The dielectric properties of the ferro-piezoelectric (FEP) ceramics have been used with success for the fabrication of ceramic capacitors with high volumetric efficiency.

Depending on the material used, “stable Mid – K” class II X7R (EIA Specifications) and “high – K” Z5U and Y5V quality capacitors can be fabricated.

Traditional compositions used are based on the BaTiO3 (BT) perovskite, with tailored doping to reduce the K change, due to the phase transition and conductivity. Ferroelectric relaxor materials can also been used, due to their larger K.

In order to increase the volumetric efficiency, multilayer capacitors (MLCs) were developed early on. The preparation of MLCs implies co-firing of the electrodes with the ceramics, thus lowering of the processing temperature has become a major research topic. This has made necessary the introduction of fluxes to reduce the temperature, in order to open the possibility of using non-noble metal electrodes like Ni or Cu.

The degradation of the conductivity of MLCs is also an important area of research.

The preparation of thin film capacitors based on FEP phases is also interesting for integration strategies, making it suitable for applications in Gbit DRAMs and high frequency devices.

In integration applications, the influence of the interfaces, and also the size effects, are of even greater importance. For these applications, the necessity for high conductivity and for high diffusion barrier capabilities of the electrode arre crucial issues for research.

Lowering the preparation temperature is also important for the integration of FEP materials compatible with silicon technology.

Any increase in the volumetric efficiency of multilayer capacitors is based on a reduction in the thickness of the dielectric layer, which can approach 1 m. Such a  reduction in thickness implies a reduction in the grain size to sub-micron range, approaching the nanoscale, thus justifying detailed study of the size effect on the dielectric properties, as a fundamental research issue.

Other possible applications of the dielectric capabilities of FEP materials are the fabrication of embedded capacitors using polymer ceramic composites and the fabrication of capacitors on flexible polymer substrates. This  has potential for the future development of flexible micro and nanoelectronic systems.

Pyroelectric materials

The characteristic that accounts for the performance of a pyroelectric material is the total pyroelectric coefficient, T. This is defined as the ratio between the change of electrical polarization of the material with respect to the change of temperature.

Usually, pyroelectric devices work under constant stress conditions, thus an important contribution to T comes from a piezoelectric contribution due to thermal expansion. This contribution, termed secondary pyroelectricity, is only possible in poled FEP and pyroelectrics, and is not exhibited by simple piezoelectrics.

Technological applications of pyroelectric materials, in both “bulk” ceramics and in thin films, are mainly within the realm of thermal radiation detection. Thermal detectors measure infrared (IR) radiation through temperature change due to infrared absorption in the 10 mm wavelength range.

The integration of FEP materials in thin films for pyroelectric applications, presents mostly the same research issues as do dielectric applications.

Following a change of temperature of a pyroelectric, a depolarizing electric field appears, both inside the pyroelectric and outside its edges (the edge depolarizing electric field (EDEF)). The EDEF extends outwards, up to a distance of the order of magnitude of the pyroelectric element width. This strong EDEF (104 ¸ 106 V/cm),  penetrating into the surrounding medium, creates a variety of physical effects. Such effects include the acceleration of charged and neutral, polarizable particles, in vacuum, and in gases. As a consequence, pyroelectric crystals have been used to ionize gas and to accelerate ions to energies up to 200keV at room temperature.

Another notable research topic in this field, is the search for lead free materials for pyroelectric applications.  Compositions based on mixtures of Bi0.5Na0.5TiO3 and BaTiO3 (BNT-BT) and alkaline niobates, are examples.

Like piezoelectric conversion,  energy harvesting by pyroelectric energy conversion can also be exploited.

Polymer/FEP particle composites for pyroelectric applications are also the subject of research interest, due to their increased figure of merit.

Multiferroic materials

Multiferroics are an interesting class of materials that display at least two of the ferroic ordered states: electric, magnetic or elastic. Such materials are therefore, able to respond simultaneously to multiple stimuli at once (e.g. magnetic fields, electric fields, mechanical stress, etc), by changing their physical properties.

For example, electric polarization can be altered by the application of a magnetic field, and conversely, the level of magnetization of the material can be altered by the application of an electric field.

The interplay between electric and magnetic states is realized via the magneto-electric (ME) effect, that is defined as the coupling between electric and magnetic fields in matter. Although this coupling can have non-linear components, the ME effect is usually described mathematically by the linear ME coupling coefficient (), which is the dominant coupling term.

The linear ME coupling coefficient () can be induced either electrically (E = dM/dE), or magnetically (H = dP/dH  0rdE/dH), where M is the magnetization, E is the electric field, P is the electric polarization, H is the magnetic field and 0, r, are respectively, the dielectric permittivity of the vacuum and of the medium. Knowledge of these relationships is essential for the design and functionality of ME devices (eg. ME sensors).

Magnetoelectric multiferroics are the most common group of multiferroics, where the two-ferroic orders are ferromagnetic and ferroelectric.

In these materials, the internal magnetic / electric fields are enhanced by the presence of the multiple long-range ferroic ordering, which in turn, produces large ME coupling effects. In addition, this ME coupling can be stress mediated in samples exhibiting piezo-effects, especially when the two ferroic phases are separated, as in multiferroic composites, or when they coexist within a matrix. In such cases, the ME coupling effect is described by an enhanced effective ME coupling coefficient (at  room temperatures) , that entails both the linear effect and the stress mediated component.

Although multiferroic materials have been known for a long time, the field is in its infancy, and multiferroics are re-emerging as materials with strong potential for a range of applications. The recent increased research interest in multiferroics is due to the following trends:

1) The potential for composite forms of such materials, that allow optimisation of their magneto-electric coupling strength and their operating temperature, by tuning the individual phases of the composite.
2) The development of thin film multiferroic structures, that makes them the best candidates for vertical integration at wafer level. The availability of high-quality thin-film multiferroics makes it easier to tailor their properties, through epitaxial strain, atomic-level engineering chemistry, and interfacial coupling, and is a prerequisite for their incorporation into practical devices.

Relaxors

Relaxors are a special class of polar dielectrics that are characterised by their very high, dispersive, polarisability.

The temperature dependencies of the dielectric permittivity, and of the losses, present broad maxima at temperatures Tm, that increases with frequency.

A model relaxor is perovskite Pb(Mg1/3Nb2/3)O3, that shows a maximum in relative permittivity of ~20000eo at -8 ºC and 1 kHz. That shifts to 20 ºC with frequency in the radio range. Moreover, relaxors present slim polarisation hysteresis loops.

These electrical properties are due to the presence of polar nano-regions (PNRs), immersed in a paraelectric matrix, and to their particular dynamics.

One opopular applications of relaxors is electrostrictive actuation in servo-displacement transducers, due to their hysteresis free electromechanical response, that is particularly advantageous for applications such as adaptative optics. 0.9Pb(Mg1/3Nb2/3)O3-0.1PbTiO3 is the material that presents the best electrostrictive characteristics, and it is the basis of electrostrictive multilayer actuators.

A major research trend, is in the development of novel lead free relaxors, with high electrostrictive strain, required to meet both industrial demand, and RoHS regulations. Although materials with polarisation coefficients similar to those of 0.9Pb(Mg1/3Nb2/3)O3-0.1PbTiO3 have been obtained, their permittivity was significantly lower.

There is also a large potential market for integrated technologies, and a lot of activity focuses on relaxor films. Relaxor films generally present lower permittivity than bulk materials however, and this is currently not well understood.

A number of fundamental scientific questions remain. These include: the precise mechanisms responsible of the appearance of PNRs, those responsible for their dynamics, and those responsible for the relaxor to ferroelectric transition.

Tunable Ferroelectrics at microwave frequency

The permittivity of ferroelectrics, especially at temperatures not too far from the ferroelectric phase transition, is strongly dependent on the prevailing electric field. Therefore, a ferroelectric capacitor can in principle serve as a varactor, (voltage dependent capacitor), fulfilling the same function as an rf-MEMS switch or semiconductor varactor.

In comparison to rf-MEMS, (capacitor switches), tunable ferroelectric capacitors do not have mechanical parts and their response is faster.

Whilst such materials are not standard IC materials, and whilst they still require further detailed research and development, the technology itself is really rather simple. Tunable ferroelectrics are attractive for applications above ≈ 10 GHz, where pin diodes show relatively high power consumption, and MEMS switches relatively slow response.

Electro-optic materials

Traditionally, the intrinsic electro-optic (EO) effect in transparent non-centrosymmetric materials is considered as a fundamental phenomena, investigated in crystals, ceramics and thin films. The electro-optic response of materials is indivisible from the piezoelectric, pyroelectric, and non-linear optical properties of piezoelectrics, since these materials essentially belong to the same family of non-centrosymmetric crystallographic classes.

Exploitation of the electro-optic properties of materials represents a major step forward in the development of new mutltifunctional devices. Such devices would  combine various material intrinsic properties (like piezoelectric and electro-optic, magnetic and electric – multiferroic), that would be tailored for specific  applications.