Piezoresponse Force Microscopy – PFM

Piezoresponse Force Microscopy mode (PFM) is the primary technique for imaging nanoscale ferroelectric domains in thin films and ceramic materials. PFM is based on the inverse piezoelectric effect inherently present in all ferroelectrics, resulting in mechanical deformation under an applied electric field. Specifically, a small AC voltage is applied between the metallic scanning probe microscopy tip in contact with the film surface, and the bottom electrode underlying the film, giving rise to an oscillating mechanical response without inducing polarization switching, measured using a lock-in technique (Shown in the figure below). The phase of this piezoresponse with respect to the excitation voltage depends on the orientation of the polarization, while its amplitude is related to the magnitude of the polarization. The field induced strain Sj in the inverse piezoelectric effect is given by: Sj=dijEi where dij are the components of the piezoelectric tensor and Ei the applied field.


One of the advantages of using PFM in the framework of the project is the ability to obtain immediate feedback on the quality of the piezoresponse, and an indicative estimate of the piezoelectric constant, extremely useful when optimizing the growth parameters for thin films. In addition, this technique allows piezoelectric behavior to be studied locally at the nanometer level. Although a fully quantitative interpretation of PFM-measured piezoelectric coefficients is limited, in particular by the highly inhomogeneous field under the PFM tip, parasitic resonances of the PFM system, and by the changing geometry of the tip, modified by wear during measurements, these effects may be mitigated by depositing a metallic electrode onto the surface. The ability to use a broad frequency range for the measurements (1 kHz-1 MHz) also allows optimization of the driving signal, while contact-resonance PFM techniques permit the use of very low applied voltages, particularly important for ferroelectric films presenting low coercive fields. Access to PFM measurements at cryogenic temperatures may also lessen the effects of switching, as well as allowing intrinsic versus domain-wall enhanced piezoelectric contributions to be explored. Quantitative characterization of piezoelectric materials using PFM measurements relies on reproducing the measured lateral and vertical sample strains using finite element simulations by fitting for the piezoelectric coefficients.

Setup for AFM measurement of piezoelectric properties of thin films using a metallic electrode on the film surface.
Setup for AFM measurement of piezoelectric properties of thin films using a metallic electrode on the film surface.



The operation of PETMEM devices involves generation and propagation of strain transients within nanoscale structures fabricated in heterogeneous thin films. The metrological challenges include:

  1. The film thickness, lateral dimensions of structures and length scale of heterogeneities are small compared with the acoustic wavelength used in current methods,
  2. The sound propagation is highly anisotropic, requiring methods which characterise both the in-plane and surface-normal propagation,
  3. Large amplitude strains in the PE layer may be required to drive the phase transition in the PR layers, resulting in anharmonic behaviour such as different velocities for compressive and tensile transients,
  4. The multi-layer PE/metal/PR stack will present acoustic impedance mismatches which will affect the efficiency of coupling of strain transients from the PE into the PR layer,
  5. The acoustic propagation will be affected by multiple factors (substrate stiffness, film thickness, crystallinity, orientation, microstructure and defects) whose contributions are not fully understood,
  6. Mechanical ageing and fatigue may lead to properties which change during the measurement.

 For these reasons, a detailed study of acoustic propagation characteristics will be important in assessing the materials for application in the PETMEM device.

Conventional Surface Acoustic Waves (SAWs) generated with electrically-driven interdigitated electrodes as strain transducers can be used to characterise thin films since their penetration depth is a small fraction of the acoustic wavelength. ‘Laser ultrasonics’ provides non-contact, in-situ metrology of mechanical properties of materials and structures, and has seen rapid development over the last decade.. A commercial apparatus (LAwave), using pulsed laser illumination to generate a SAW at frequencies up to several 100MHz, is used at NPL to provide information on mechanical properties in thin films of thickness down to a few nm. In the research lab, the generation and characterization of SAWs in silicon was recently extended to 90GHz, corresponding to wavelengths of order 50nm, using metal gratings produced by electron beam lithography and optical excitation with ~50fs laser pulses. NPL will apply pulsed-laser acoustic methods to measure the elastic properties and speed of sound in new piezo materials in complex environments (thin microstructured films and nanoscale pillars with multi-layers). Thin epitaxial piezoelectric films are highly anisotropic due to strain, substrate clamping, crystalline orientation and microstructure (even columnar growth), therefore it is essential to measure both in-plane and out-of-plane (surface normal) properties. In the metrology work package, the team will develop existing capability using laser impulse excitation of the fundamental acoustic pulse to excite an in-plane SAW (using commercial LAWave apparatus at NPL). The SAW measurement is an industry-standard technique and we will assess its effectiveness for the thin films produced in this project. Detection using SAW and/or pvdf piezoelectric knife edge sensing technology (used in the LAWave system) permits assessment of the elastic properties of PE and PR materials.



Double beam laser interferometry


AixACCT will decrease the spot size of its unique double beam laser interferometer which has become a kind of unofficial standard to the piezoelectric MEMS industry. This virtually small technical change, which is quite difficult, has a large benefit for the community, because it allows measurement of the true d33 of the material. By this it will be possible to extend the number of coefficients that can be characterized. This fact will strengthen aixACCT's market position as the most innovative company in this market. This extension of the capability of the DBLI tool goes clearly beyond state of the art.

 AixACCT will measure the resistivity change through the film. Existing instruments measure the resistivity change in plane of the piezoresistive material. Furthermore, aixACCT will integrate this functionality in its DBLI tool. This will offer a comprehensive characterization of piezoelectric as well as piezoresistive film properties on a full wafer and therefore guarantee quality assurance of the full process from film quality to device quality. It will be essential for any company going into this direction to do such kind of characterization and offering a tool that offer almost the whole characterization features in a single tool will be clearly beyond state of the art.

 Finally, aixACCT will develop a cell test feature for the new and quite complex device, which is not a digital but analogue signal test. Because this is a completely new technology it is beyond state of the art test. Also this test will be integrated into the functionality of the DBLI tool.