For many years traditional 193i lithography has been extended to the next technology node by means of multi-patterning techniques. However recently such a 193i technology became challenging and expensive to push beyond the technology node for complex features that can be tackled in a simpler manner by the Extreme UltraViolet Lithography (EUVL) technology. Nowadays, EUVL is part of the high-volume manufacturing device landscape and it has reached a critical decision point where one can push further the single print on 0.33NA full field scanner or move to a EUV double patterning technology with more relaxed pitches to overcome current 0.33NA stochastic limits. In this work we have selected the 28nm pitch dense line-space (P28) as critical decision check point. We have looked at the 0.33NA EUV single print because it is more cost effective than 0.33NA EUV double patterning. In addition, we have conducted a process feasibility study as P28 in single print is close to the resolution limit of the 0.33NA EUV full field scanner. We present the process results on 28nm dense line-space patterning by using Inpria’s metal-oxide (MOx) EUV resist. We discuss the lithographic and etching process challenges by looking at resist sensitivity, unbiased line edge roughness (LER) and nano patterning failures after etching (AE), using broad band plasma (BBP) and e-beam (EB) defectivity inspection tools. To get further understanding on the P28 single patterning capability we have integrated the developed EUV MOx process in a relevant iN7 technology test vehicle by developing a full P28 metallization module with ruthenium. In such a way we were able to carry on electrical tests on metallized serpentine, fork-fork and tip-to-tip structures designed with a purpose of enabling further learning on pattern failures through electrical measurements. Finally, we conclude by showing the readiness of P28 single exposure using Inpria’s MOx process on a 0.33NA EUV full field scanner.
We propose the use of machine learning based analytics to simplify OPC (Optical Proximity Correction) model building process which demands concurrent optimization of more than 70 parameters as nodes shrink. We first built a deep neural network architecture to predict the RMS error, for a given set of model parameters. The neural network was trained on existing OPC model parameters and corresponding output RMS data of simulations to achieve an accurate prediction of output RMS for given set of OPC model parameters. Later, a sensitivity analysis-based methodology for recursive partitioning of OPC modelling parameters was employed to reduce the total search space of OPC model simulations. This resulted in reduction of the number of OPC model iterations performed during model tuning by orders of magnitude.
EUV single patterning opportunity for pitch 28nm metal design is explored. Bright field mask combined with a negative tone develop process is used to improve pattern fidelity and overall process window. imec N3 (Foundry N2 equivalent) logic PNR (place and route) designs are used to deliver optimized pupil through source mask optimization and evaluate OPC technology. DFM (Design For Manufacturing) related topics such as dummy metal insertion and design CD retarget are addressed together with critical design rules (e.g. Tip-to-Tip), to provide balanced design and patterning performance. Relevant wafer data are shown as a proof of above optimization process.