Method for detecting defects in depth features

Abstract

A method for detecting defects in high aspect ratio channel holes, vias, or trenches is disclosed. First, a substrate is provided having a film stack thereon and a plurality of depth features in the film stack. At least one depth feature of the plurality of depth features includes a defect. Then, an optical inspection process is performed on the substrate. The substrate is illuminated by a broadband light beam. Some broadband DUV beams scattered and / or reflected from the substrate are collected by a detector, thereby producing a bright field illumination image of a plurality of depth featuresin the film stack.

Description

technical field [0001] The present disclosure relates to defect inspection methods. More specifically, the present disclosure relates to a non-destructive method for trapping defects at the bottom of deep (high aspect ratio) features such as holes, vias, slots and / or trenches. Background technique [0002] Three-dimensional (3D) NAND memory continues to evolve, with the critical dimension (CD) shrinking as the stack becomes thicker, the cell density becomes higher and higher. In 3D NAND memory manufacturing, as the number of layers in the film stack increases and multi-stack technology emerges, inspect the lower stack, especially at depth (high aspect ratio) features such as holes, vias, gaps, and / or trenches defects have become increasingly critical. [0003] However, due to the high aspect ratio (40-100) of the channel holes, it is difficult to implement conventional methods for detecting defects, such as under-etch defects at the bottom of the channel holes. Existing m...

AI Technical Summary

Problems solved by technology

Existing methods are expensive and destructive with low precision and difficult to meet production requirements

Images