Projection print quality is dependent upon the response of the resist /developer system to the optically projected image.For this reason, it is critical, during process optimization, to relate final pattern quality to both the chemical and physical contrast terms. One method for doing this is to monitor resist /developer response in the absence and then the presence of diffracted light.These results are then compared to the predicted optical requirements.For best performance, the resist /developer (or chemical) contrast must exceed that of the potential optical (or physical) contrast.To do rapid contrast comparisons it is convenient to relate the chemical contrast term,r, to the physical contrast term MTF, via the CMTF where the CMTF = (101 /r -1) /(101 /+ 1) = (Ethreshold -Einitial)/ (Ethreshold + Einitial) and the MTF =(max -Emin)/(Emax + Emij.Perfect imaging potential occurs chemically when the CMTF = 0, Einitial = Ethreshold, and physically when the MTF = 1.0, Emin = 0. As a result, acceptable patterning occurs when the MTF /CMTF ratio is greater than one. This ratio is a comparative, accept /reject test only final process optimization is done by monitoring the accepted resist /developer systems' performance on projected images. Optimum process latitude is predicted by determining focal tolerances at various mask sizing biases and by measuring linewidth necking over steps.This study demonstrates the dependence of the final print quality on the chemical contrast.A method is examined for approximating minimum allowable chemical contrast required for adequately printing desired device design sizes. Then chemical contrast results for various metal-ion -free develop processes are shown. Next, the combined resist /developer/ optical performance is optimized by using focal tolerance and linewidth control over steps as process constraints. Finally, it is shown that focal tolerances are dependent on chemical contrast and that the focus budget is tunable.

No abstract

No abstract

No abstract

No abstract