Using 550 previously calculated vibrational energy levels and dipole moments we performed simulations of the HCN-->HNC isomerization dynamics induced by sub-one-cycle and few-cycle IR pulses, which we represent as Gaussian pulses with 0.25-2 optical cycles in the pulse width. Starting from vibrationally pre-excited states, isomerization probabilities of up to 50% are obtained for optimized pulses. With decreasing number of optical cycles a strong dependence on the carrier-envelope phase (CEP) emerges. Although the optimized pulse parameters change significantly with the number of optical cycles, the distortion by the Gaussian envelope produces nearly equal fields, with a positive lobe followed by a negative one. The positions and areas of the lobes are also almost unchanged, irrespective of the number of cycles in the half-width. Isomerization proceeds via a pump-dumplike mechanism induced by the sequential lobes. The first lobe prepares a wave packet incorporating many delocalized states above the barrier. It is the motion of this wave packet across the barrier, which determines the timing of the pump and dump lobes. The role of the pulse parameters, and in particular of the CEP, is to produce the correct lobe sequence, size and timing within a continuous pulse.

The magnetic anisotropy energy of Ni n ͞Cu͑100͒ is calculated in terms of the spin-polarized fully relativistic Korringa-Kohn-Rostoker method including surface relaxation by using 2D structure constants originally described for low-energy electron diffraction calculations. Investigating different relaxations, an explanation for the reorientation transition from in-plane to perpendicular can be given. For a relaxation of 25.5% (c͞a 0.945) this reorientation occurs at about seven layers of Ni and yields second order terms to the magnetic anisotropy energy that are in excellent agreement with experiment.[S0031-9007 (98)08322-7] PACS numbers: 75.30.Gw, 75.70.Ak, 75.70.Cn Thin films of Ni on Cu (100) show an unexpected behavior of magnetic phase transitions [1][2][3][4][5][6][7][8][9][10][11][12]. In contrast to quite a few other magnetic thin films on noble metal surfaces, where with increasing film thickness the moments first line up perpendicular to the surface and then reorientate to an in-plane direction, for n , 7 monolayers of Ni on Cu(100) an in-plane direction of the magnetization M is observed [5,6,10] that reorientates [5,6,10,11] at about n 7 to perpendicular to plane, switching eventually back to in plane for n . 37 [5]. The anisotropic part of the free energy for magnetic multilayer systems (see, e.g., Ref.[2] and references therein) can phenomenologically be described by E 2pM 2 cos 2 u 2 K 2 cos 2 u ,(1)where K 2 refers to the second order term of the magnetic anisotropy energy (MAE) and u denotes the angles of M with respect to the surface normal. As indicated in Eq.(2), where d refers to the film thickness, K 2 is thought to consist of two parts, namely, a thickness-independent "volume"-like contribution K y 2 and a thickness-dependent "interface-surface"-like contribution K s 2 . It is more or less the interplay of these two constants that together with the shape anisotropy (magnetic dipole-dipole interaction) determines the unique features of the reorientation properties of thin films of Ni on Cu(100).In the present paper the fully relativistic spin-polarized screened Korringa-Kohn-Rostoker (KKR) method [13,14] was applied using the spin-polarized local density functional as given by Vosko et al. [15]. In order to be able to treat layer relaxation, the occurring screened structure constants [16,17] have been derived for a system of layers which have only the same in-plane translational symmetry [18][19][20], but otherwise can differ in the respective interlayer distance. Self-consistency for the effective potentials and effective exchange fields (with a uniform orientation perpendicular to plane) was obtained using 45 k k points in the irreducible part of the surface Brillouin zone (ISBZ). In all cases three layers of Cu served as "buffer" at the Cu͞Ni interface and three "empty" layers as buffer to the Ni͞vacuum interface; i.e., for a given number n of Ni monolayers the total number L of atomic layers investigated self-consistently is L n 1 6. It should be noted that the present calculations refer...

No abstract

We investigate population transfer across the barrier in a double-well potential, induced by a pair of time-delayed single-lobe half-cycle pulses. We apply this setup both to a one-dimensional (1D) quartic model potential and to a three-dimensional potential representing HCN-->HNC isomerization. Overall the results for the two systems are similar, although in the 3D system some additional features appear not seen in the 1D case. The generic mechanism of population transfer is the preparation by the pump pulse of a wave packet involving delocalized states above the barrier, followed by the essentially 1D motion of the delocalized part of wave packet across the barrier, and the eventual de-excitation by the dump pulse to localized states in the other well. The correct timing is given by the well-to-well passage time of the wave packet and its recurrence properties, and by the signs of the field lobes which determine the direction and acceleration or deceleration of the wave packet motion. In the 3D system an additional pump-pump-dump mechanism linked to wave packet motion in the reagent well can mediate isomerization. Since the transfer time and the pulse durations are of the same order of magnitude, there is also a marked dependence of the dynamics and the transfer yield on the pulse duration. Our analysis also sheds light on the pronounced carrier envelope phase dependence previously observed for isomerization and molecular dissociation with one-cycle and sub-one-cycle pulses.

In the present article we investigate the influence of the contact region on the distribution of the chemical potential in integer quantum Hall samples, as well as the longitudinal and Hall resistance as a function of the magnetic field. First we use a standard quantum Hall sample geometry and analyse the influence of the length of the leads where current enters/leaves the sample and the ratio of the contact width to the width of these leads. Furthermore we investigate potential barriers in the current injecting leads and the measurement arms in order to simulate non-ideal contacts.