Temperature effect on charge density of silicon nitride films deposited in SiH4–NH3–N2 plasma

Improved surface passivation of silicon nitride films requires a high positive charge density. A neural network model of SiN charge density was used to investigate temperature effects on charge density. For a systematic modeling, the deposition process was characterized by means of a face-centered Box Wilson experiment. Prediction performance of neural network model was optimized by using genetic algorithm. Interestingly, charge density was varied little with the substrate temperature regardless of SiH4 flow rates. Charge density variation was not sensitive to [Si–H] variation. A pronounced temperature effect at higher NH3 flow rate or lower radio frequency (rf) power was attributed to a relatively large [N–H]. For NH3 or rf power variation, charge density was strongly correlated to [N–H]/[Si–H].