DOC PREVIEW
UW-Madison GEOSCI 777 - A New Monte Carlo Program that Computes X-ray Spectra Obtained with a Scanning Electron Microscope

This preview shows page 1-2-3-4-5 out of 16 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 16 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 16 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 16 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 16 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 16 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 16 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Win X-ray: A New Monte Carlo Program thatComputes X-ray Spectra Obtained with a ScanningElectron MicroscopeRaynald Gauvin,1,* Eric Lifshin,2Hendrix Demers,1Paula Horny,1and Helen Campbell11Department of Metals and Materials Engineering, McGill University, Montréal, Québec H3A 2B2, Canada2College of Nanoscale Science and Engineering, University at Albany, CESTM, 251 Fuller Road, Albany, NY 12203, USAAbstract: A new Monte Carlo program, Win X-ray, is presented that predicts X-ray spectra measured with anenergy dispersive spectrometer ~EDS! attached to a scanning electron microscope ~SEM! operating between 10and 40 keV. All the underlying equations of the Monte Carlo simulation model are included. By simulatingX-ray spectra, it is possible to establish the optimum conditions to perfor m a specific analysis as well asestablish detection limits or explore possible peak overlaps. Examples of simulations are also presented todemonstrate the utility of this new program. Although this article concentrates on the simulation of spectraobtained from what are considered conventional thick samples routinely explored by conventional microanaly-sis techniques, its real power will be in future refinements to address the analysis of sample classifications thatinclude rough surfaces, fine structures, thin films, and inclined surfaces because many of these can be bestcharacterized by Monte Carlo methods. The first step, however, is to develop, refine, and validate a viable MonteCarlo program for simulating spectra from conventional samples.Key words: X-ray microanalysis, Monte Carlo, scanning electron microscopy, energy dispersive spectrometry,electron scatteringINTRODUCTIONA new program is described that predicts full X-ray spectrameasured with an energy dispersive spectrometer ~EDS!attached to a scanning electron microscope ~SEM!.Itisbased on the simulation of electron scattering in solidsusing the Monte Carlo method described by Gauvin andL’Espérance ~1992! for X-ray microanalysis in the t ransmis-sion electron microscope and by Hovington et al. ~1997! forX-ray microanalysis in the SEM; these methods are anextension of previous work by Bishop ~1965! and by Kar-duck and Rehbach ~1991!.Previous attempts that have been made to computecomplete X-ray spectra generally full into two categories,closed form analytical models and Monte Carlo models.Because of uncertainties in either the correctness of thephysical models or the parameters used in both approaches,they must be refined with the use of some adjustable param-eters to achieve a close match with experimental spectra.Examples of the analytical model approach include the Desk-top Spectrum Analyzer ~DTSA! described by Fiori and Swyt~1989! and more recently an approach descr ibed by Dun-cumb et al. ~2001!. The former is readily available from theNational Institute for Standards and Technology ~NIST! atno charge; however its oper ation is limited to Apple com-puters. The general availability of the latter program is notknown at this time although many of the equations used inits development are described in the reference. This refer-ence also specifies that the RMS error between measuredand simulated peak intensities was determined to be 7.1%for 360 K, L, and M peaks from known standards. Verygood agreement was also found for peak to integr ated totalbackground ratios. This approach is therefore very promis-ing for estimating spectra for thick samples and standardsfor conventional electron microprobe analysis where theelectron excitation volume and absorption paths are wellcontained in the region being analyzed. The second ap-proach, that of Monte Carlo modeling, has been describedby Ding et al. ~1994!, who computed the bremsstrahlungusing Monte Carlo simulations; however, their work waslimited to pure elements and normal electron beam inci-dences and furthermore absolute X rays were not computed.More recently the Monte Carlo program PENELOPEhas been applied to the generation of complete X-r ay spec-tra, including characteristic and continuum peaks; however,its initial use was limited to K lines and L lines by Llovetet al. ~2004!. The results are given in photon/electron/steradian as a function of energy and a very large number ofReceived March 18, 2003; accepted July 27, 2005.*Corresponding author. E-mail: [email protected]. Microanal. 12, 49–64, 2006DOI: 10.1017/S1431927606060089MicroscopyANDMicroanalysis© MICROSCOPY SOCIETY OF AMERICA 2006trajectories are required to compute a full spectrum, thustaking many hours of computer time. Because the MonteCarlo approach even with some simplification is inherentlymore time consuming than closed form modeling, i t is notunreasonable to ask “Why bother using it?” if the accuracyand computation time of closed form modeling is consider-ably better. The answer is that, for conventional analysis, itmay not be particularly useful beyond predicting absoluteX-ray spectra and detectability limits. However, many pra c-tical samples examined cannot be described easily or even atall by modifications of conventional models, and MonteCarlo modeling may be the only viable approach. Further-more Monte Carlo calculations may facilitate a more con-ventional closed form approach. An example of the lattermight be the generation of a calibration curve of K ratioversus film thickness. Examples of nonconventional analysisinclude rough surfaces, fine structures, thin films, and in-clined surfaces.Before Monte Carlo methods can truly be shown to beaccurate in addressing the more complex analytical caseslisted, it was felt that a program should first be developedthat accurately predicts the complete spectr a expected fromconventional samples using normal electron beam inci-dence to the specimen surface. It was further felt that a newprogram should be relatively easy to use, fast, generallyavailable, as is DTSA, and that all of the details of theunderlying model be fully disclosed. What is described hereis such a program, Win X-ray, designed to simulate the totalX-ray spectra ~the characteristic lines and the bremsstrah-lung! for homogeneous alloys or compounds at any ang le ofthe incident electron beam and the X-ray detector takeoffangle. Furthermore, this program computes absolute X-r ayintensities in order to simulate real experimental conditionsfor incident electron energies ranging from 10 to 40 keV.Results can then be used for a variety of applications,including the calculation of detection limits,


View Full Document

UW-Madison GEOSCI 777 - A New Monte Carlo Program that Computes X-ray Spectra Obtained with a Scanning Electron Microscope

Documents in this Course
Load more
Download A New Monte Carlo Program that Computes X-ray Spectra Obtained with a Scanning Electron Microscope
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view A New Monte Carlo Program that Computes X-ray Spectra Obtained with a Scanning Electron Microscope and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view A New Monte Carlo Program that Computes X-ray Spectra Obtained with a Scanning Electron Microscope 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?