VOLUME 78, NUMBER 24 P H Y S I C A L R E V I E W L E T T E R S 16 JUNE 1997 Comment on ``Evidence of High Frequency vco and the intensities to vary. Our excellent fits, with Propagating Modes in Vitreous Silica'' a mean x2 1, are illustrated in Fig. 1. As seen, the crossover model predicts a strong q dependence of the In a recent Letter [1], Benassi et al. report an inelas- inelastic line shapes even for excitations that no longer tic x-ray scattering (IXS) investigation of vitreous silica propagate, for q up to a few times qco. In view of the small y-SiO2 at high temperature T . They assume that the inelastic contribution on top of the wings of the huge elastic inelastic structure factor, S q, v , is a damped harmonic peak, this dependence could easily be taken for a shift in a oscillator (DHO) response. From their fits, they conclude DHO peak frequency. that high frequency acoustic modes propagate at frequen- The issue addressed here is not the likely significance of cies v well above the boson peak. Another recent Letter an increase of vco with T [5]. Our main point, illustrated [2] reports combined inelastic neutron scattering (INS) and in Fig. 1, is that the propagation of acoustic modes at IXS data on y-SiO2 at room T. These were analyzed with v beyond the boson peak of glasses is not demonstrated an S q, v that implies strong acoustic-phonon scattering from available IXS data [1,5­7]. A corresponding remark leading to a Ioffe-Regel crossover [3]. The crossover wave of caution applies to the use of the DHO line shape at high vector qco 1 nm21 and frequency vco 4 meV are v, also for liquids [8]. The interpretation of the DHO derived from INS spectra, but agree with the (quite weak) in terms of propagating excitations [1,6­8] is arbitrary IXS signals. The crossover energy corresponds to the at this stage. The latter points rejoin one aspect of an boson peak. Although by themselves measurements of earlier Comment on water [9]. Further progress of IXS, S q, v cannot demonstrate localization [4], we feel that application to well selected test cases, and results from it is rather meaningful to investigate the applicability of alternate spectroscopies such as INS, will presumably all the models. be needed to clarify this central issue. Hence, we analyzed the data presented in Fig. 2 of [1] We thank M. Soltwisch for correspondence on glycerol with the model and parameters of [2]. We simply allowed [7], F. Sette for sending Ref. [6], and G. Ruocco for a preprint of [1]. M. Foret,1 E. Courtens,1 R. Vacher,1 and J.-B. Suck2 1Laboratoire des Verres, UMR 5587 CNRS Université de Montpellier II F-34095 Montpellier Cedex 5, France 2Institut für Physik T.U. Chemnitz-Zwickau D-09107 Chemnitz, Germany Received 29 October 1996 [S0031-9007(97)03360-7] PACS numbers: 63.10.+a, 61.10.Eq, 63.50.+x, 78.70.Ck [1] P. Benassi, M. Krisch, C. Masciovecchio, V. Mazzacurati, G. Monaco, G. Ruocco, F. Sette, and R. Verbeni, Phys. Rev. Lett. 77, 3835 (1996). [2] M. Foret, E. Courtens, R. Vacher, and J.-B. Suck, Phys. Rev. Lett. 77, 3831 (1996). [3] G. Polatsek and O. Entin-Wohlman, Phys. Rev. B 37, 7726 (1988). [4] For example, D. J. Thouless, Phys. Rep. C 13, 94 (1974). [5] Two IXS spectra of y-SiO2 at room T with better resolution recently became available [6]. These are also well fitted using the crossover model with vco 4 meV. Our fits are of a quality similar to the DHO ones in [6]. [6] C. Masciovecchio et al., Phys. Rev. B 55, 8049 (1997). [7] C. Masciovecchio et al., Phys. Rev. Lett. 76, 3356 (1996). [8] F. Sette et al., Phys. Rev. Lett. 75, 850 (1995); 77, FIG. 1. IXS data taken from [1] (points) and their fits to the 83 (1996); G. Ruocco et al., Nature (London) 379, 521 model of [2] (line) with v (1996); Phys. Rev. B 54, 14 892 (1996). co 6.1 meV, or qco 1.5 nm21. The calculated inelastic signal (solid line) and the elastic peak [9] F. J. Bermejo et al., Phys. Rev. Lett. 76, 3656 plus background (dashed line) are also shown separately. (1996). 0031-9007 97 78(24) 4669(1)$10.00 © 1997 The American Physical Society 4669