CAD for VLSI Manufacturability and Reliability

在本研究中，我们提出了考虑超大规模集成电路（VLSI）的可制造性和可靠性的计算机辅助设计（CAD）技术。首先，我们考虑了光学光刻技术，这是制造当今超大规模集成电路的使能技术。由于特征尺寸已经变得比用于打印它们的光源的波长小得多，设计后的分辨率增强技术（ret），如光学接近校正（OPC）和相移掩模（PSM）被用来提高它们的可打印性。理想情况下，应该在设计步骤（例如，物理设计）中考虑这些ret。我们建议开发支持OPC/ psm的CAD技术。其次，我们考虑下一代光刻（NGL）技术，这将有可能在未来取代光学光刻。两种主要的NGL候选者是电子投影光刻（EPL）和极紫外光刻（EUV）。由于EPL膜掩膜的结构，设计需要进行分区，然后需要“缝合”在一起。我们提出研究这种掩模布局划分问题。对于极紫外光刻，一个主要问题是由掩模对极紫外光的吸收和散射引起的（即“耀斑”问题）。我们提出了CAD技术来解决“耀斑”问题。最后，我们考虑了天线问题，这是一种等离子体诱导栅极氧化物降解的现象。我们计划继续进行二极管插入/路由的研究，以解决天线问题。

In this research, we propose computer-aided design (CAD) techniques that consider manufacturability and reliability of very large scaled integrated circuits (VLSI),First, we consider optical lithography which is the enabling technology for the fabrication of today's VLSI. Since feature sizes have become much smaller than the wave length of the light sources used to print them, post-design resolution enhancement techniques (RETs) such as optical proximity correction (OPC) and phase shifting mask (PSM) are used to improve their printability. Ideally, these RETs should be considered during the design steps (e.g., physical design). We propose to develop OPC/PSM-aware CAD techniques.Second, we consider next generation lithography (NGL) techniques which will potentially replace optical lithography in the future. The two leading NGL candidates are electron projection lithography (EPL) and extreme ultra-violet (EUV) lithography. Due to the structure of the membrane mask for EPL, a design needs to be partitioned and later needs to be "stitched" together. We propose to study this mask layout partitioning problem. As for EUV lithography, a major problem is caused by the absorption and scattering of EUV lights by the mask (i.e., the "flare" problem). We propose CAD techniques for solving the "flare" problem.Finally, we consider the antenna problem which is a phenomenon of plasma induced gate oxide degradation. We plan to continue our ongoing research on diode insertion/routing for fixing antenna problems.