Evaluation of the Extent of Electron Scattering in a Low Vacuum Scanning Electron Microscope

Low vacuum scanning electron microscopes utilize gas molecules in the specimen chamber to neutralize the specimen charging from the impinging electron beam. Electrons scattered out of the primary electron beam can generate x-rays from areas far from the intended analysis location. Measurements under low vacuum conditions reveal x-ray production hundreds of micrometers away from the analysis spot contributing signal to the collected spectrum. Experimental data show a wider radius of x-ray production than Monte Carlo simulation of electron scattering in the gas above the specimen predicts. Use of the electron beam gas path length shortening cone for low vacuum x-ray analysis significantly reduces the scattered electron induced x-ray contributions.

The growth in the market share and the use of low vacuum scanning electron microscopes has been steadily increasing over the last two decades. The concept of having gas molecules in the specimen chamber was pioneered by Electroscan but is now employed by all scanning electron microscope vendors. The presence of the gas molecules can help dissipate electrons that accumulate on the specimen surface during examination in the scanning electron microscope (SEM). This has been extremely valuable in making SEM accessible to specimens that did not lend themselves to being coated with a conductive film for a host of reasons. This advantage is not without compromise, the gas molecules that help eliminate specimen charging also can scatter electrons out of the primary electron beam. As the primary electrons travel from the pole piece to the specimen, if they interact with a gas molecule, there is a chance that they will be scattered out of the focused beam onto the specimen at some distance from the intended analysis spot. The amount of scattering depends on many factors, including accelerating voltage, gas pressure in the chamber, the length that the electron beam must travel through the gas to reach the specimen, and the gas composition. A small amount of electron scattering does not impact the electron imaging because the majority of the primary electron beam remains in the fine focused beam. The scattered electrons are detrimental to x-ray microanalysis done in the low vacuum SEM.

The goal of this work is to better understand the extent of electron scattering in the low vacuum SEM under different operating conditions. This understanding can lead to more complete x-ray microanalysis performed under low vacuum conditions. The use of field emission electron gun sources to produce a smaller, more finely focused primary electron beam is expected to improve the x-ray microanalytical performance of the low vacuum SEM.

X-ray spectrum maps were collected of an area that spanned a binary interface. The interface used in this experiment was silicon and carbon. X-ray spectral maps were collected in high vacuum and in several low vacuum conditions. The spectral data cubes were then processed for line scans of silicon x-ray intensity perpendicular to the interface. The silicon intensity over the carbon is the result of electrons scattered out of the primary electron beam and onto the silicon where an x-ray can be generated and detected. These electron scattering profiles demonstrate the extent that spurious x-ray contributions can get into x-ray spectra under low vacuum conditions. This data can be used to plan x-ray microanalysis experiments under low vacuum conditions to minimize the effect of the gas. This work may lead to an algorithm that could be used to correct x-ray data collected under low vacuum conditions to remove the unwanted signals.

• Comparison of the experimental x-ray data with a Monte Carlo simulation reveals a larger radius of x-ray generation than predicted by electron scattering in the gas.
• Reducing the electron beam path length through the gas with the x-ray cone significantly reduces spurious x-ray generation.