Structure de mise en forme 2 colonnes

WP4 results for M18

The development of the high-performance 0.5THz HBT’s pursued within the DOTFIVE project requires a continuous device miniaturization. This results in physical effects, which are not captured in existing compact models because they did not play a significant role so far. In Deliverable D4.2.1, the physical effects and other model issues that have been observed so far for experimental data and device simulation of advanced doping profiles are summarized in conjunction with first ideas for solutions. From this, the tasks for developing improved compact model formulations are identified that guide the work of the second project phase.
Within a contribution to Deliverable D4.1.2, the statistical variations of several device characteristics and FoMs across the wafer are visualized and their impact on parameter extraction and device modeling was identified.
Measurements of DC characteristics and RF characteristics for cold and hot states as well as temperature dependent measurements have been performed on the most recent DOTFIVE technology (B3T from ST) using 50GHz and 110GHz measurement systems, respectively. Experimental characterization on high-speed transistors and related de-embedding structures as well as DC test structures has been performed.

Work related to thermal impedance modelling:
The accuracy of the previously developed model of the steady thermal behaviour of SiGe HBTs has been verified for the case of air-filled trenches and ST HBT structures. Preliminary results indicate that a more accurate approach is needed. An analytical model has been developed for the transient case, and compared to numerical FEM simulations. The comparison shows that it is difficult to capture all the complex effects that occur at all time ranges. The behaviour of the thermal impedance in the frequency domain has been investigated. The accuracy of equivalent thermal network has been assessed.

Work related to HF measurements at low temperature:
The first set of HBTs developed within WP2 shows fT/fMAX of 275 / 400 GHz at room temperature. At low temperature (30 K) these frequencies increase to 420 / 500 GHz. Transit times represent 75% of the emitter-collector delay involved in the transit frequency.

Work related to HF EM simulation:
The coupling between probe tips and wafer surface has been investigated through EM-simulation and the simulation results have been compared to measurements. It is pointed out that the results are very dependent on the adjacent structures lying under the probe tips. Different solutions are analyzed to master and/or reduce the coupling and ensure reproducibility.