Abstract
Quantifying energy-dispersive X-ray (EDX) spectra acquired in a transmission electron microscope can produce inconsistent and unreliable results even if correct values for specimen thickness and density are being used for calculating an appropriate absorption and fluorescence correction. This can be due to various factors, such as incorrect k-factors, faulty absorption corrections due to specimen roughness, specimen buckling or wrong estimates of the detector's take-off angle, complicated fluorescence effects not accounted for in the software etc. For a thorough investigation of the InGaAs system, wedge shaped specimens were cleaved from GaAs and InAs wafers and studied by EDX spectroscopy in a JEOL2010F instrument, with thicknesses ranging from a few nanometres up to several microns. The thickness dependence of the relative intensities of X-ray lines that remains after nominal absorption and fluorescence correction were taken into account could be used to perform linear least-squares fits, thereby eliminating to a first order the apparent remaining thickness dependence of the k-factors. The resulting fits demonstrate this correction improves the quantification of the chemical concentrations of indium and gallium in InGaAs samples of known thickness to 1at%.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call