Application Notes

Photomask Inspection with SEM-EDS - A Critical Part of the Semiconductor Manufacturing Process

Published: 10 Jun 2023 · Last updated: 10 Jun 2023

Tags: EDS

Aim

To inspect photomask quality in terms of defects and cleanliness – including those defects that cannot be identified using optical methods.

Challenge

A photomask is, essentially, a stencil for printing patterns on a wafer during the semiconductor device manufacturing process; it contains the patterns used to create integrated circuits and can be used to manufacture features that exist as small as the nanoscale (in current generation devices). The photomask must have accurately controlled patterns and zero defects, because mask errors will be amplified and repeated on the wafer. Optical inspection methods lack the required spatial resolution and defect detection sensitivity, especially for EUV (Extreme ultraviolet lithography) masks.

Solution

E-beam analysis with an SEM + EDS can help quickly locate different types of defects on a mask. By using Oxford Instruments' Ultim Extreme detector, which is optimised for low kV analysis, a user can acquire valuable hyperspectral map data in a timescale of a few minutes. This compositional data aids in finding the cause of defects and ultimately helps to eliminate their occurrence and increase manufacturing yield.

EDS map of a photomask using Ultim Extreme showing defects in circled areas

Figure 1: Top: EDS map of a (prior generation) photomask using Ultim Extreme.

EDS map showing defect particle rich in carbon and smaller contaminants 100 nm in size acquired at 3 kV, 800 pA in 5 minutes

Data revealing defects in circled areas (particulates 200 nm–5 μm in size). Bottom: map showing defect particle rich in C, and smaller contaminants 100 nm in size (circled) acquired at 3 kV, 800 pA in only 5 minutes.

Conclusion

SEM-EDS inspection with Ultim Extreme offers both the high sensitivity and speed needed to find defects on photomasks in high throughput semiconductor manufacturing environments. The approach offers flexibility so that different defects, such as pattern imperfections and particulates can be found. Compositional EDS data acquired at very high spatial resolution can help to determine the seriousness of defects and if they fall into a catastrophic category which will cause a circuit not to function correctly.

← Back to Learning