High energy-resolution electron energy-loss spectroscopy based on electron microscopy

2000 ◽  
pp. 239-272
Keyword(s):  
Electron Microscopy ◽  
Energy Loss ◽  
Electron Energy ◽  
Energy Resolution ◽  
Electron Energy Loss Spectroscopy ◽  
High Energy ◽  
High Energy Resolution ◽  
Resolution Electron ◽  
Electron Energy Loss

Toward 10 meV Electron Energy-Loss Spectroscopy Resolution for Plasmonics

2014 ◽  
Vol 20 (3) ◽  
pp. 767-778 ◽  
Author(s):  
Edson P. Bellido ◽  
David Rossouw ◽  
Gianluigi A. Botton
Keyword(s):  
Energy Loss ◽  
Electron Energy ◽  
Surface Plasmon ◽  
Energy Resolution ◽  
Electron Energy Loss Spectroscopy ◽  
High Energy ◽  
High Energy Resolution ◽  
Surface Plasmon Resonances ◽  
Plasmon Resonances ◽  
Electron Energy Loss

AbstractEnergy resolution is one of the most important parameters in electron energy-loss spectroscopy. This is especially true for measurement of surface plasmon resonances, where high-energy resolution is crucial for resolving individual resonance peaks, in particular close to the zero-loss peak. In this work, we improve the energy resolution of electron energy-loss spectra of surface plasmon resonances, acquired with a monochromated beam in a scanning transmission electron microscope, by the use of the Richardson–Lucy deconvolution algorithm. We test the performance of the algorithm in a simulated spectrum and then apply it to experimental energy-loss spectra of a lithographically patterned silver nanorod. By reduction of the point spread function of the spectrum, we are able to identify low-energy surface plasmon peaks in spectra, more localized features, and higher contrast in surface plasmon energy-filtered maps. Thanks to the combination of a monochromated beam and the Richardson–Lucy algorithm, we improve the effective resolution down to 30 meV, and evidence of success up to 10 meV resolution for losses below 1 eV. We also propose, implement, and test two methods to limit the number of iterations in the algorithm. The first method is based on noise measurement and analysis, while in the second we monitor the change of slope in the deconvolved spectrum.

Sign in / Sign up

Export Citation Format

Share Document