A wide range of characterization methods based upon Atomic Force Microscopy (AFM) can be applied for nanometer scale analysis of semiconductor and electronic materials & devices. The standard capability of AFM to image the surface topography with nanometer scale spatial resolution is augmented with the capability to measure a variety of physical properties: electrical (for example, surface potential, work function, conductivity, carrier density), magnetic, thermal (temperature distribution, thermal conductivity), and mechanical. More recently, this capability has also been expanded to chemical identification at the nanometer scale using AFM-IR imaging and spectroscopy.We will cover a structured overview of these operating modes presenting their capabilities and limitations with case studies in semiconductor, ferro-electrics and general nanoelectrical devices. The operating modes are compared in terms of (i) physical properties characterized (carriers, charges, conductivity, ferro-electricity, dielectric constant, ...), (ii) spatial resolution, (iii) sensitivity, (iv) dynamic range, and (v) quantification potential. The capabilities and limitations of each operating mode are illustrated with examples from Si-based devices as well as compound semiconductors (HEMT and other), organic devices, and 1D & 2D materials.Novel approaches and novel prob & cantilevers focus on improvements in spatial resolution, detection sensitivity and the capability to correlate multiple properties efficiently. One such approach consists of datacube modes where one acquires an electrical or chemical spectrum in every pixel of the image: one of the parameters controlling the electrical measurements (for example DC voltage, AC voltage or AC frequency) is varied between two user defined values in each pixel of the 2D image. When combined with force mapping method, one simultaneously obtains a correlated nano-mechanical data cube holding information on modulus, stiffness, and adhesion. At the same time, one avoids contact mode imaging, thus extending electrical measurements to soft and fragile samples and improving measurement consistency as compared to conventional contact mode operation. Optimization of the force mapping movements and the electrical measurement setup allows one to maintain a relatively high imaging speed (typ. 10-100 ms/pixel). The approach is illustrated by multiple modes: in conductive AFM, I-V spectra in each pixel are acquired by ramping the DC bias. Applied to SCM and sMIM, dC/dV-V and C-V spectra are acquired. In PFM either switching loops or contact resonance spectra are obtained. Investigation of a variety of materials illustrates the capability to reveal sample properties which are not accessible or easily missed in conventional methods where maps at only one or a few discrete settings are acquired. The datacube approach can also be combined with chemical characterization offered by the AFM-IR method, in this way, offering a path to correlated mechanical-chemical analysis.
Read full abstract