T2.4 Characterization

• Electrical characterization of both substrates & devices (non-stationary transient phenomena, their effect on device performance; extraction of complete small-signal equivalent circuits; substrate assessmen,t –) - UCL, CEA
• Physical characterization of both substrate & devices (stress, roughness, TEM/SEM microscopy –) - UPB, CEA
• Assessment of device performances / Figures of Merit (analogue, digital, RF) – UCL, CEA
• Benchmarking of various technological options (P18, P28, 28FDSOI) and booste–s - All

Task 2.4 aims at in-depth comprehensive characterization of FDSOI, on both substrate and device levels, as well as critical assessment of various innovations/booster options proposed within WP2. Partners of Task 2.4 will cover both electrical characterization (UCL, CEA) and physical characterization (UPB, CEA)thus allowing for comprehensive assessment and deep understanding of device and substrate behaviour. UCL, UPB and CEA will compare outputs/conclusions and exploit complementary information coming from electrical and physical measurements to make a link between final device/substrate performance/behaviour and fabrication process/use of boosters. This will allow for not only critical assessment, but even more important, to identify and propose routes for further process or device architecture/configuration optimization (which might be depending on the final application).

Task 2.4 will cover both substrate and device performances assessment as well as define Figures of Merit (FoM) (analogue, digital, RF) and benchmarking various technological options (P18, P28, 28FDSOI) and proposed boosters.

Our methodology will assure a link between substrate performance and device performance allowing for co-optimization towards circuit operation and final applications (incl. temperature and rad environment considerations). Device / substrate optimization for radiation/self-heating immunity perspective will be considered (UCL). The outputs of T2.4 are thus beneficial for both simulation/modelling development (T2.5), process optimization (T2.1, T2.2, T2.3) and circuit design (WP3) works.

In this task electrical characterization (by UCL) will largely exploit wide frequency band measurements (covering a range from DC to 170 GHz) and low frequency noise measurements at various temperatures in order to:

• Identify and separately analyse different non-stationary effects, transient phenomena in these new advanced devices. Non-stationary effects (being related to the e.g. defects, noise, self-heating, thermal and/or capacitive couplings, substrate-network response, ...) may strongly degrade/alter expected device performance/FoM. These effects, while often neglected, becomes crucial for deeply downscaled technologies, especially for analogue and RF applications, affecting their performance. The challenge is to separately analyse those effects, which is mandatory for fair benchmarking, proper optimization and modelling.
• Extract complete small-signal equivalent circuits (and thus separate analyse and optimization of both intrinsic device and surrounding parasitic elements) to assure a fair comparison of various technological options not only on inner device parameters, but also on the external parasitic elements, which become critically important for advanced downscaled performance, where they may become dominant (wrt. inner elements) and alter the device performance.
• Characterize extensively the substrate, also to support T2.5. Substrate needs to be considered as an important part of entire device (and not as a support of active part) in advanced devices affecting device/circuit behaviour/FoM and thus worth of optimization/ characterization /modelling (particularly in a link with T2.5).

Electrical characterization will be complemented by the physical-chemical characterization performed (by UPB and CEA) in terms of stress/strain, morphology, roughness, composition, critical dimensions, defects and interface (e.g. Si/SiO2) quality inspection on different substrates (e.g. 28FDSOI, 18FDSOI, 22FDX) and devices, at different development stages. UPB and CEA will involve different tools and techniques available on site to assess the complementarity/interdependence of the gathered properties (morphological, structural and elemental/compositional), resulting from cross-linked AFM (PSD), HR and UHR TEM-SEM-ZC-EDX (including high quality and purity FIB based TEM lamella generation), HR SEM-EDX, µRaman and HRXRD analysis. For a better understanding of the roughness parameters, such as the differences in the surface profil–s - distribution of features, shapes and/or sizes, Power Spectral Density (PSD) studies will be performed. The elemental/compositional analysis via EDX (in TEM) will be corroborated with the interplanar distances of the materials (atomic resolution TEM). The TEM lamellae will be generated by removing/mitigating the Ga contamination through nm-level accuracy polishing using an Ar gun inside the FIB-SEM system (BSOTA approach). For some techniques (like the TERS), CEA will develop the methodology in a way to define the more adapted protocols. The data generated by physical characterization will be linked to electrical characterization results to help gathering a refined picture of FDSOI device operation.

The planned outcome of this task will be:

• a benchmarking of various characterization and metrology techniques to increase the knowhow on substrate and device
• the development of the more adapted metrology and characterization protocols to support the development advanced substrates, FDSOI based evices and PCM.