Free volume cannot explain the spatial heterogeneity of Debye-Waller factors in a glass-forming binary alloy

Journal of Non-Crystalline Solids

Volumn 352, Issue 42-49 SPEC. ISS., 2006, Pages 5098-5102

  Widmer Cooper, Asaph ^a   Harrowell, Peter ^a  

^a UNIVERSITY OF SYDNEY   (Australia)

Author keywords

Glass transition; Molecular dynamics

Indexed keywords

COMPUTER SIMULATION; CORRELATION METHODS; GLASS TRANSITION; MOLECULAR DYNAMICS;

DEBYE-WALLER FACTORS; FREE VOLUME;

BINARY ALLOYS;

EID: 33750441855     PISSN: 00223093     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.jnoncrysol.2006.01.136     Document Type: Article

References (15)

1. Tsao S.S., and Spaepen F. Acta. Metall. 33 (1985) 881
2. Cohen M., and Turnbull D. J. Chem. Phys. 31 (1959) 1164
3. Falk M., and Langer J.S. Phys. Rev. E 57 (1998) 7192
4. Heggen M., Spaepen F., and Feuerbacher M. J. Appl. Phys. 97 (2005) 033506
5. Nagel C., Ratzke K., Schmidtke E., Faupel F., and Ulfert W. Phys. Rev. B 60 (1999) 9212
6. Widmer-Cooper A., Fynewever H., and Harrowell P. Phys. Rev. Lett. 93 (2004) 135701
7. Buchenau U., and Zorn R. Europhys. Lett. 18 (1992) 523
8. Ngai K.L. Phil. Mag. 84 (2004) 1341
9. Starr F.W., Sastry S., Douglas J.F., and Glotzer S.C. Phys. Rev. Lett. 89 (2002) 125501
10. Perera D.N., and Harrowell P. J. Chem. Phys. 111 (1999) 5441
11. Perera D.N., and Harrowell P. Phys. Rev. E 59 (1999) 5721
12. Sturgeon J.B., and Laird B. J. Chem. Phys. 112 (2000) 3474
13. Sastry S., Corti D.S., Debenedetti P.G., and Stillinger F.H. Phys. Rev. E 56 (1997) 5524
14. Qian J., Hentschke R., and Heuer A. J.Chem.Phys. 111 (1999) 10177
15. See. Press W.H., Flannery B.P., Teukolsky S.A., and Vetterling W.T. Numerical Recipes: The Art of Scientific Computing (2002), Cambridge University Press, Cambridge pp. 633-637. This method is in general more robust than standard linear correlation methods such as Pearson's. Also, because the distribution of the ranks is known (they are uniformly distributed) it is possible to calculate the significance of non-zero values. Except for rare cases where {divides}K{divides} < 0.1, we typically obtain non-zero K values with greater than 99.9% significance. As for Pearson's correlation coefficient, K varies from -1 to 1, with 0 indicating that the two data sets are uncorrelated