VOLUME 85, NUMBER 4 PHYSICAL REVIEW LETTERS 24 JULY 2000 Comment on “Quantum Wave Packet Dynamics with Trajectories” In a recent Letter [1], Lopreore and Wyatt (LW) con- cluded that the usual view of particle tunneling within the Bohm interpretation (the quantum potential lowers the bar- rier so that the Bohm particles can travel classically over it [2,3]) is “misleading and incorrect for smooth barrier penetration.” They claim that the Bohm particles feel only a significant quantum force during a short boost phase just after the launching of the wave packet (far from the bar- rier), and only those particles that acquire enough kinetic energy (KE) during this boost phase can pass over the bar- rier and contribute to the tunneling transmission. Con- sequently, they propose a decoupling approximation that consists of ignoring the quantum force after a short ini- tial decoupling time. In this Comment, we argue that the results of LW correspond to a particular case in which tun- neling is not significant. Thus, their conclusions about the tunneling mechanism are not valid. For comparison with LW results, we consider the scat- tering of Gaussian wave packets, jCt , by an Eckart po- tential V x V 0 sech 2 ax 2 x b . If the transmission coefficient of the eigenstate jC k is T k , the total trans- mission probability T can be decomposed into pure tun- neling (T tun ) and over-the-barrier (T OB ) components: T T tun 1 T OB Z k B 0 jak j 2 T k dk 1 Z ` k B jak j 2 T k dk , (1) where ak C k j Ct , and k B q 2mV 0 ¯ h 2 is the wave vector associated with the top of the barrier. Figure 1 shows T , T tun , and T OB as a function of the barrier width. For wide barriers, T tun tends to be negli- gible and T converges to T OB and to the result given by the LW decoupling approximation. The case analyzed by LW corresponds to this limit, i.e., T OB ¿ T tun . However, for thin enough barriers, tunneling tends to dominate the total transmission, and the decoupling approximation fails (it underestimates T ). The increase of transmission due to tunneling can only be accounted for by particles that have KE lower than the barrier height V 0 after the boost phase. Since Bohm trajectories are classical, we must conclude that the quantum potential lowers the classical barrier, as is usually claimed [2,3] and has been clearly shown for Hamiltonian eigenstates [4,5]. The transmission coeffi- cient calculated under the decoupling approximation de- pends on the barrier height but not on the barrier width (see Fig. 1). The decoupling approximation is reasonable for thick barriers but underestimates the transmission by FIG. 1. Transmission coefficient of a Gaussian wave packet (central energy of 0.03 eV, spatial dispersion of 5 nm) impinging upon an Eckart barrier as a function of the barrier width at half maximum (BWHM) for a fixed barrier height, V 0 0.04 eV. The bold line corresponds to the total transmission coefficient (T ), the dashed line to the tunneling component (T tun ), the dotted line to the over-the-barrier component (T OB ), and the continu- ous line to LW decoupling approximation. Inset: transmission coefficient versus central energy (wave vector) for two barrier widths: BWHM 1.3 nm (squares) and BWHM 18.8 nm (circles). The continuous line corresponds to the decoupling approximation. orders of magnitude when tunneling dominates (inset of Fig. 1). In conclusion, the interpretation of tunneling given by LW is misleading and incorrect. Tunneling requires that the quantum potential lowers the classical barrier as pre- viously claimed [2,3]. The decoupling approximation is valid only when quantum effects (tunneling and over-the- barrier resonances) are unimportant. Support by the DGES project PB-97-0182 is acknowledged. Jordi Suñé and Xavier Oriols Departament d’Enginyeria Electrònica Universitat Autònoma de Barcelona 08193-Bellaterra, Spain Received 24 September 1999 PACS numbers: 34.10.+x [1] C. L. Lopreore and R. E. Wyatt, Phys. Rev. Lett. 82, 5190 (1999). [2] D. Bohm and B.J. Hiley, The Undivided Universe (Rout- ledge, London, 1993). [3] P. R. Holland, The Quantum Theory of Motion (Cambridge University Press, Cambridge, England, 1993). [4] X. Oriols, F. Martín, and J. Suñé, Phys. Rev. A 54, 2594 (1996). [5] X. Oriols, F. Martín, and J. Suñé, Solid State Commun. 99, 123 (1996). 894 0031-900700 85(4) 894(1)$15.00 © 2000 The American Physical Society