Hostname: page-component-848d4c4894-m9kch Total loading time: 0 Render date: 2024-05-06T20:49:32.476Z Has data issue: false hasContentIssue false
  • English
  • Français

Detecting Clusters in Atom Probe Data with Gaussian Mixture Models

Published online by Cambridge University Press:  26 April 2017

Jennifer Zelenty ,
Andrew Dahl ,
Jonathan Hyde ,
George D. W. Smith  and
Michael P. Moody
Jennifer Zelenty*
Affiliation:
Department of Materials, University of Oxford, Oxford OX1 3PH, UK
Andrew Dahl
Affiliation:
Department of Statistics, University of Oxford, Oxford OX1 3LB, UK
Jonathan Hyde
Affiliation:
National Nuclear Laboratory, Culham Science Centre, Abingdon OX14 3DB, UK
George D. W. Smith
Affiliation:
Department of Materials, University of Oxford, Oxford OX1 3PH, UK
Michael P. Moody
Affiliation:
Department of Materials, University of Oxford, Oxford OX1 3PH, UK
*
*Corresponding author. jzelenty@gmail.com
Get access

Abstract

Accurately identifying and extracting clusters from atom probe tomography (APT) reconstructions is extremely challenging, yet critical to many applications. Currently, the most prevalent approach to detect clusters is the maximum separation method, a heuristic that relies heavily upon parameters manually chosen by the user. In this work, a new clustering algorithm, Gaussian mixture model Expectation Maximization Algorithm (GEMA), was developed. GEMA utilizes a Gaussian mixture model to probabilistically distinguish clusters from random fluctuations in the matrix. This machine learning approach maximizes the data likelihood via expectation maximization: given atomic positions, the algorithm learns the position, size, and width of each cluster. A key advantage of GEMA is that atoms are probabilistically assigned to clusters, thus reflecting scientifically meaningful uncertainty regarding atoms located near precipitate/matrix interfaces. GEMA outperforms the maximum separation method in cluster detection accuracy when applied to several realistically simulated data sets. Lastly, GEMA was successfully applied to real APT data.

Keywords

atom probe tomographycluster identificationGaussian mixture modelsexpectation maximizationmachine learning
Type
New Approaches and Correlative Microscopy
Information
Microscopy and Microanalysis , Volume 23 , Special Issue 2: Atom Probe Tomography & Microscopy 2016 , April 2017 , pp. 269 - 278
Copyright
© Microscopy Society of America 2017 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Bishop, C.M. (2006). Pattern Recognition and Machine Learning. Cambridge: Springer.Google Scholar
Felfer, P., Ceguerra, A.V., Ringer, S.P. & Cairney, J.M. (2015). Detecting and extracting clusters in atom probe data: A simple, automated method using Voronoi cells. Ultramicroscopy 150, 30–36.CrossRefGoogle Scholar
Gault, B., Moody, M.P., Cairney, J.M. & Ringer, S.P. (2012). Atom Probe Microscopy. New York, NY: Springer.Google Scholar
Hirata, A., Fujita, T., Wen, Y.R., Shneibel, J.H., Liu, C.T. & Chen, M.W. (2011). Atomic structure of nanoclusters in oxide-dispersion-strengthened steels. Nat Mater 10, 922–926.CrossRefGoogle ScholarPubMed
Hyde, J.M & English, C.A. (2000). An analysis of the structure of irradiation induced Cu-enriched clusters in low and high nickel welds. Symposium R Microstructural processes in irradiated materials, Boston, MA, 27–30 November 2000.Google Scholar
Hyde, J.M., Marquis, E.A., Wilford, K.B. & Williams, T.J. (2011). A sensitivity analysis of the maximum separation method for the characterisation of solute clusters. Ultramicroscopy 111, 440–447.CrossRefGoogle ScholarPubMed
Jägle, E.A., Choi, P. & Raabe, D. (2014). The maximum separation cluster analysis algorithm for atom-probe tomography: Parameter determination and accuracy. Micros Microanal 20, 1662–1671.CrossRefGoogle ScholarPubMed
London, A.J., Santra, S., Amirthapandian, S., Panigrahi, B.K., Sarguna, R.M., Balaji, S., Vijay, R., Sundar, C.S., Lozano-Perez, S. & Grovenor, C.R.M. (2015). Effect of Ti and Cr on dispersion, structure and composition of oxide nano-particles in model ODS alloys. Acta Mater 97, 223–233.CrossRefGoogle Scholar
Moody, M.P., Ceguerra, A.V., Breen, A.J., Cui, X.Y., Gault, B., Stephenson, L.T., Marceau, R.K.W., Powles, R.C. & Ringer, S.P. (2014). Atomically resolved tomography to directly inform simulations for structure–property relationships. Nat Commun 5, 1–10.Google Scholar
Pollock, T.M. & Tin, S. (2006). Nickel-based superalloys for advanced turbine engines: Chemistry, microstructure and properties. J Propul Power 22, 361–374.Google Scholar
Stephenson, L.T., Moody, M.P., Liddicoat, P.V. & Ringer, S.P. (2007). New techniques for the analysis of fine-scaled clustering phenomena within atom probe tomography (APT) data. Microsc Microanal 13, 448–463.Google Scholar
Styman, P.D., Hyde, J.M., Wilford, K., Morley, A. & Smith, G.D.W. (2012). Precipitation in long term thermally aged high copper, high nickel model RPV steel welds. Prog Nucl Energy 86, 86–92.Google Scholar
Styman, P.D., Hyde, J.M., Wilford, K. & Smith, G.D.W. (2013). Quantitative methods for the APT analysis of thermally aged RPV steels. Ultramicroscopy 132, 258–264.CrossRefGoogle ScholarPubMed
Wells, P.B., Yamamoto, T., Miller, B., Milot, T., Cole, J., Wu, Y. & Odette, G.R. (2014). Evolution of manganese-nickel-silicon-dominated phases in highly irradiated reactor pressure vessel steels. Acta Mater 80, 205–219.Google Scholar