Correction of EELS dispersion non-uniformities for improved chemical shift analysis Robert W.H. Webster,1* Alan J. Craven,1 Bernhard Schaffer,2 Sam McFadzean,1 Ian MacLaren,1 Donald A. MacLaren.1** 1 SUPA, School of Physics and Astronomy, University of Glasgow, G12 8QQ, United Kingdom 2 AMETEK GmbH, Ingolstaedtlerstr. 12, Munich, Germany *Email: rwebster@physics.org **Email: dmaclaren@physics.org What is included? ----------------- The data provided has been divided into three folders. 'Scripts' contains the DigitalMicrograph(DM) user scripts "Dispersion_Measurement_vUltram.s" and "Dispersion_Corrector_vUltram.s" which are used for running the dispersion measurement while on the microscope, and processing measurements/correcting data, respectively. The raw spectra used to produce Figures 3 & 4 of the paper are included, so that others may trial the processing script on real data, if they so wish. The folders 'NiO' and 'Diamond' contain the relevant spectra for those specimens, as well as the appropriate dispersion measurement data. Filenames for spectra follow the style "Sample_dispersion+offset_dualEELSrange.dm3" so that they can be identified by their acquisition conditions, e.g. "NiO_0p5eVpCh+HL.dm3" Filenames for dispersion measurment data follow the style "dispersioneVpCh_disp_meas_ref/scan.dm3" (e.g. "0p5eVpCh_disp_meas_scan.dm3") and separate data are provided for each specimen in the folders 'NiO' and 'Diamond.' The 'ref' or 'scan' in the filename indicates whether the data contains the reference ZLP (low-loss) or the scanned ZLP (high-loss). Transmission Electron Microscopy Data ------------------------------------------------------ Raw data is recorded in Gatan's Digital Micrograph file format (details at the time of writing are advertised at: http://www.gatan.com/products/tem-analysis/gatan-microscopy-suite-software. This file format is also readable by open source software package HyperSpy (http://hyperspy.org/). All data was acquired using a JEOL ARMcFEG instrument at 200kV, equipped with a Gatan Quantum electron spectrometer for EELS measurements. Notes on using the scripts --------------------------- The data-processing described in the paper is achieved through the use of two DM scripts. The first, "Dispersion_Measurement_vUltram.s", is to be used while on the microscope. We recommend first setting up the microscope for STEM using your usual alignment routine at a high magnification (0.5Mx or greater) to optimise standard-focus/eucentric height, as well as tuning the spectrometer focus, before navigating to a hole or off the edge of the specimen so the beam is in vacuum. At this point, set a live EELS spectrum view for dualEELS with no offset and identical exposure times so the ZLP is at the zoom point in both views; you are now ready to start the measurement script. The script will then ask for some user inputs to define step sizes, and whether frames should be dropped to mitigate afterglow effects. We have used a 10 channel step size throughout the paper and find this sufficient. After the initial beam shift, a popup will ask if you are ready, and you should check the condenser focus before commencing. Pressing OK here will start the full measurement, which will take approximately 2 minutes for the 10 channel step. At the end of the measurement, two datasets will need to be saved, the low-loss and high-loss. The second script, "Dispersion_Corrector_vUltram.s", can be used offline to process the measurement and correct the dispersion of real spectra after your microscope session. Simply load the dispersion measurement low- and high-loss data, as well as spectra you wish to correct (optional) and executing the script will take you through all or some of the processing steps.