edfas.org ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 26 NO. 2 4 EDFAAO (2024) 2:4-8 1537-0755/$19.00 ©ASM International® SCANNING THERMAL MICROSCOPY FOR LOCALIZING AND MONITORING DEFECTS IN ELECTRONICS Séverine Gomès CNRS, Centre d’Energétique et de Thermique de Lyon (CETHIL) Villeurbanne, France severine.gomes@insa-lyon.fr INTRODUCTION The constant downscaling of electronic components for the integration of electronic devices has increased the need for thermal measurements at micro and nano scales. Heat flow management at these scales is crucial to meet the challenges of thermal management of complete integrated systems, detecting defects and failures in devices, and improving the design and performance of microelectronic technologies. This need has motivated the development of various thermal techniques, such as thermoreflectance, Raman thermometry, fluorescence, and luminescence. However, these optical methods suffer from a spatial resolution limited by optical diffraction, which is not sufficient to char- acterize nanoscale electronic components. To overcome the lack of versatile, high-resolution thermometry techniques, scanning thermal microscopy (SThM),[1] based on scanning probe microscopy, has been developed and is today one of the most effective techniques used for thermal characterization of materials and systems at small scales. SThM can detect and localize heat sources generated in active devices and analyze the thermal conductance of systems with a sub-100 nm lateral spatial resolution. Nevertheless, the correct analysis and evaluation of heat transfer within individual nanostructures, which have recently been successfully implemented in miniaturized devices, remains an experimental challenge. To meet this challenge, an innovative hybrid instrument based on the combination of a SThM and a scanning electron microscope (SEM) has been built at CETHIL.[2] This article presents the principle of SThM instruments and their potential uses for the local thermal analysis of passive and active electronic components and devices. The hybrid SEM-SThM instrument recently developed at CETHIL is then described and demonstrated through a quantitative analysis of heat transport in an indivi- dual nanowire. SET UP The SThM instruments are based on atomic force microscopy (AFM) systems and use AFM cantilevers equipped with a resistive element located on the tip (Fig. 1). Piezoelectric scanners are used to move the sample vertically and laterally. As the sample surface is scanned by the tip, deviations of the cantilever, corresponding to variations in the force of interaction between the tip and the sample, are detected optically. The thermal sensor on the tip is used to analyze the sample’s thermal properties. The sensor’s temperature (Tp) can be determined because its metallic electrical resistance (Rp) depends linearly on temperature: ΔRp = Rp0 · αp · (Tp-Tp0), with αp the temperature coefficient of the metal’s electrical resistivity and Rp0 its electrical resistance at a reference temperature T0. ΔRp is measured using a control unit based on a Wheatstone bridge. In imaging mode, the SThM probe is generally used in the AFM’s contact and constant force modes. The SThM system can provide a topography image and “thermal” mapping simultaneously. The contrast of the thermal mapping reflects the variation in heat flux exchanged between the probe and the sample. Any variation in this heat flux that is linked to the local sample’s properties during the sample’s scanning will induce a variation in the probe temperature Tp and its electrical resistance Rp. DIFFERENT MODES The probe can be operated in both passive and active modes. In passive mode, a very low electrical current is used to measure the electrical resistance, Rp, of the thermal sensor without heating the sample, and the probe acts as a resistive thermometer. SThM in this mode has mainly been applied to imaging the temperature field at the surface of self-heated micro devices. In active mode, the probe is Joule-heated and used simultaneously as an ultra-local heat source for the sample
RkJQdWJsaXNoZXIy MTYyMzk3NQ==