Non-equilibrium radical probe technique for measuring spatially resolved fluorine radical density

Le résumé fourni par la source

Fluorine radicals are the primary etchant in plasma-based silicon etching processes, yet spatial measurements of fluorine radical density remain challenging with existing diagnostic techniques. Actinometry, the most widely applied method, provides only a volumetrically averaged measurement, limiting its utility for spatially resolved studies. Here, we present a new, spatially resolved technique based on a non-equilibrium etching radical probe consisting of a tungsten reactive probe and an aluminum inert probe, each instrumented with a thermocouple. By measuring the differential thermal response of the probe pair immediately before and after plasma extinction, the rate of heating due to fluorine radical etching can be isolated from all other heating and cooling mechanisms without requiring thermal equilibrium. An in situ two-step calibration procedure is used to determine the effective heat capacity of each probe and to relate the thermal response to the tungsten etch rate. The tungsten etch probability was characterized from 100 to 700 °C, revealing an Arrhenius activation energy of 0.07 eV below 250 °C and good agreement with previously reported values above 400 °C. Fluorine radical density was measured radially across a silicon wafer in an inductively coupled plasma etcher under several conditions. The measured density profiles correlate directly with independently measured silicon etch rate profiles, confirming that spatial nonuniformities in the silicon etch rate are driven by the fluorine radical distribution. Silicon etch probabilities determined from spatially resolved probe measurements are significantly lower than those derived from actinometry, providing evidence that volumetric averaging in actinometric measurements is a key source of the large disagreements in reported silicon etch probabilities.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.