Aberration-aware robust mask design with level-set-based inverse lithography

Yijiang Shen, Ngai Wong, Edmund Y. LamProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2010 · DOI

TL;DR

This paper introduces a novel method for designing photomasks that are robust to optical aberrations and process variations, a critical challenge in modern semiconductor manufacturing. By incorporating these variations directly into an inverse lithography framework using a level-set approach, the authors demonstrate a significant improvement in pattern fidelity compared to traditional methods.

What Problem Does This Paper Address?

Modern semiconductor manufacturing relies on optical lithography to print intricate circuit patterns onto silicon wafers. As feature sizes shrink, lithography systems become more susceptible to imperfections, particularly optical aberrations (deviations from perfect spherical wavefronts) and process variations like defocus. Traditional Optical Proximity Correction (OPC) methods often assume ideal imaging conditions, leading to degraded pattern quality and increased manufacturing errors when these variations occur. This paper tackles the problem of designing photomasks that maintain high pattern fidelity even in the presence of these unavoidable real-world imperfections.

Methods

The core of the paper's contribution lies in its aberration-aware inverse lithography (ILT) framework. Here's a breakdown of their approach:

  • Modeling Aberrations: They extend the standard point spread function (PSF) model to explicitly include wave aberrations. These aberrations are represented using Zernike polynomials, a standard mathematical tool for describing optical wavefront distortions.
  • Statistical Incorporation of Variations: Instead of assuming fixed aberration values, the method treats defocus and aberrations as random variables with statistical properties (zero mean, non-zero variance). This allows the optimization to consider a range of possible process conditions.
  • Level-Set-Based Optimization: The inverse lithography problem is formulated as minimizing the difference between the desired (target) pattern and the simulated output pattern under various aberration conditions. This optimization is solved using a level-set-based time-dependent model. This approach is known for its stability and ability to handle complex shape evolution, making it suitable for mask pattern design.
  • Robustness Objective: The objective function is designed to minimize the expected pattern fidelity error across a distribution of defocus and aberration conditions, rather than just at a single nominal point. This directly leads to a mask design that is robust to these variations.

Datasets

Not discussed in the paper. The paper focuses on a simulation-based approach using a defined target pattern and simulated lithography system parameters.

Key Insights and Ideas

  • Aberrations are Not Just Noise, They're Design Constraints: The paper moves beyond treating aberrations as something to be corrected after mask design. Instead, it integrates them into the design process itself, acknowledging their fundamental impact on imaging.
  • Statistical Modeling is Key to Robustness: By modeling process variations statistically, the method accounts for the inherent uncertainty in manufacturing, leading to designs that perform well across a range of conditions, not just the ideal.
  • Level-Set Methods Offer a Stable Path to Complex Designs: The use of level-set methods provides a computationally tractable and stable way to solve the complex inverse problem, enabling the synthesis of intricate mask patterns.
  • "Virtual Virtuality" is Being Addressed: The authors acknowledge that current ILT methods are still not fully applicable to real-world manufacturing due to issues like process variation. Their work directly addresses this gap by making ILT more practical.

Key Results

The paper presents simulation results demonstrating the effectiveness of their aberration-aware ILT approach.

  • Improved Pattern Fidelity Against Coma: Figure 2 and Table 1 show that the proposed aberration-aware mask significantly reduces pattern error when subjected to coma aberration (c31=100) compared to a mask designed for nominal conditions. The pattern error for the aberration-aware mask was substantially lower.
  • Robustness to Combined Defocus and Coma: Figure 3 and Table 2 further validate the method's ability to handle combined defocus (f=800) and coma (c31=80) aberrations. The aberration-aware mask shows improved pattern fidelity under these combined conditions compared to a nominal mask.
  • Qualitative Improvements: The figures visually demonstrate how the aberration-aware masks produce cleaner and more accurate output patterns when subjected to the simulated aberrations, preserving critical features that would otherwise be distorted.

Limitations and Future Work

  • Computational Cost: While the level-set method is presented as computationally efficient, the paper implies that solving the inverse problem can still be time-consuming, especially with a large number of aberrations.
  • Model Simplifications: The paper mentions that the PSF can be computed using approximations for speed. The accuracy of these approximations for very large aberrations might be a limitation.
  • Future Work: The authors suggest that their method can be extended to handle more complex scenarios, including partially coherent illumination and phase-shifting masks (PSMs). They also note that when more aberrations are present, the root-mean-square deformation follows a chi-square distribution, and their algorithm is expected to produce similarly robust masks.

Summary updated

Abstract

Optical proximity correction (OPC) is one of the most widely used Resolution Enhancement Techniques (RET) in mask designs. Conventional OPC is often designed for a set of nominal imaging parameters without giving sufficient attention to the process variations caused by aspherical wavefront leaving the exit pupil of the lithography system. As a result, the mask designed may deliver poor performance with process variations. In this paper, we first describe how a general point spread function (PSF) with wave aberration can degrade the output pattern quality, and then show how the wave aberration function can be incorporated into an inverse imaging framework for robust input mask pattern design against aberrations. A level-set-based time-dependent model can then be applied to solve it with appropriate finite difference schemes. The optimal mask gives more robust performance against either one specific type of aberration or a combination of different types of aberrations. © 2010 SPIE.
BibTeX
@inproceedings{shen2010aberration,
  author = {Yijiang Shen and Ngai Wong and Edmund Y. Lam},
  title = {{Aberration-aware robust mask design with level-set-based inverse lithography}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2010},
  doi = {10.1117/12.863973},
}
Figure