The Design of a Reflection Electron Energy Loss Spectrometer Attachment for Low Voltage Scanning Electron Microscopy

This paper presents the design of a reflection electron energy spectrometer (REELS) attachment for low voltage scanning electron microscopy (LVSEM) applications. The design is made by carrying out a scattered electron trajectory ray paths simulation. The spectrometer attachment is small enough to fit on the specimen stage of an SEM, and aims to acquire nanoscale spatially resolved REELS information. It uses a retarding field electrostatic toroidal sector energy analyzer design, which is able to lower the kinetic energies of elastically backscattered electrons to pass energies of 10 eV or less. For the capture of 1 keV BSEs emitted in the polar angular range between 40 to 50°, direct ray-tracing simulations predict that the spectrometer attachment will have an energy resolution of around 0.4 eV at a pass energy of 10 eV, and 0.2 eV at a pass energy of 5 eV. This predicted performance will make it a suitable REELS attachment for SEMs that use field emission electron sources.

Quantification of Isomer-Resolved Iodide CIMS Sensitivity and Uncertainty Using a Voltage Scanning Approach

Chemical ionization mass spectrometry (CIMS) using iodide as a reagent ion has been widely used to classify organic compounds in the atmosphere by their elemental formula. Unfortunately, calibration of these instruments is challenging due to a lack of commercially available standards for many compounds, which has led to the development of methods for estimating CIMS sensitivity. By coupling a Thermal desorption Aerosol Gas chromatograph (TAG) simultaneously to a flame ionization detector (FID) and an iodide CIMS, we use the individual particle-phase analytes, quantified by the FID, to examine the sensitivity of the CIMS and its variability between isomers of the same elemental formula. Iodide CIMS sensitivities of isomers within a formula are found to generally vary by one order of magnitude with a maximum deviation of two orders of magnitude. Furthermore, we compare directly measured sensitivity to a method of estimating sensitivity based on declustering voltage (i.e., “voltage scanning”). This approach is found to carry high uncertainties for individual analytes (half to one order of magnitude), but represents a central tendency that can be used to estimate the sum of analytes with reasonable error (~30 % differences between predicted and measured moles). Finally, GC retention time, which is associated with vapor pressure and chemical functionality of an analyte, is found to qualitatively correlate with iodide CIMS sensitivity, but the relationship is not close enough to be quantitatively useful and could be explored further in the future as a potential calibration approach.