Extremely low count detection for EELS spectrum imaging by reducing CCD read-out noise

Area of Science:

• Materials Science
• Spectroscopy
• Electron Microscopy

Background:

• Electron Energy Loss Spectroscopy (EELS) spectrum imaging requires extremely low count detection.
• Electron irradiation and limited applications are challenges in current EELS.

Purpose of the Study:

• To systematically study and reduce CCD noise in EELS.
• To enable high signal-to-noise detection for low signal EELS spectrum imaging.

Main Methods:

• Statistical analysis of CCD noise in EELS.
• Proposed a calculation method to estimate noise properties.
• Developed a noise reduction procedure involving dark reference subtraction and spectral summation.
• Applied a gain-averaging method to enhance the signal-to-noise (SN) ratio.

Main Results:

• Demonstrated effective reduction of dominant random noise components.
• Achieved high SN spectra even with single-count core-loss signals.
• Validated the method for low signal EELS spectrum imaging.

Conclusions:

• The developed noise reduction method significantly improves EELS spectrum imaging.
• This technique is beneficial for analyzing radiation-sensitive materials.
• Enables accurate measurements of monochromated spectra and other low signal data.