Real-time observation of electron dynamics in molecules

An important application for ultrashort laser pulses with an impulse duration in the attosecond range (1 as = 10-18s) is the control of electron dynamics in molecules. If it were possible to manipulate electrons on their natural time scale from a few hundred attoseconds to a few femtoseconds (1 fs = 10-15s), it might be possible to control molecular processes with unprecedented precision. Theoretical work of recent years predicts that a charge (positive defect in the electron distribution) caused by sudden ionization of the molecule is capable of moving from one side of the molecule to the other in a few hundred attoseconds. This so-called charge migration, in turn, may favor the subsequent chemical processes, e.g. dissociation, what is referred to as charge-directed reactivity. A first step in experimenting with these predictions will be to make ultra-fast electron motion in molecules experimentally visible. The experiments, which were carried out in the new attosecond laboratories at the Max Born Institute and published in Physical Review Letters, prove that this has now been achieved for the first time.

The shortest laser pulses that scientists can produce today have a pulse duration of only 50-500 as. In a process called "high-order harmonic generation (HHG)", a noble gas is ionized by a near-infrared laser , The released electron is first accelerated away from the ion in the laser field in field polarization direction and then accelerated back to the ion. Recombination occurs here and the energy required to ionize the gas and accelerate the free electron is released in the form of photons. The photon energy is typically in the range of 10-100 eV and corresponds to a multiple of the photon energy of the near-infrared laser. Since the period of the near-infrared field is in the range of only a few femtoseconds and the recombination time is approximately the same for all released electrons, optical pulse durations in the attosecond range can be achieved.

While femtosecond pulses have been successfully used to study structural changes in molecules, attosecond pulses are needed to track the rapid movements of electrons in molecules. After being first observed in 2001, these ultrashort optical pulses have been used in studies of a variety of systems, including atoms, molecules, and solids. Previous work by the MBI team has already used attosecond pulses for pump-probe investigations on molecules. First experimental evidence for the coupling of electron and nuclear motion on attosecond and femtosecond time scales and the effects of entanglement in multielectron systems was found. However, it has not been possible to study pure electron dynamics in neutral molecules with attosecond pulses.

Fig. 1 The attosecond pulse train (blue) is either synchronized with the zero crossing of the near infrared field (red) (a) or its extrema (b). In this case, the molecule is ionized when its electron density distribution has been changed by the near-infrared field. This change in electron density is shown in (c) for molecular nitrogen. Here, red means that the density in this part of the molecule has increased, while blue indicates the opposite.

The latest results for the successful observation of molecular electron dynamics are based on "dynamic alignment". This technique was used by the MBI team some time ago and is now an integral part of many experiments that investigate the structure and dynamics of and in molecules. If a molecule is exposed to laser radiation that is too weak to ionize the system, but strong enough to induce a dipole in the molecule, the most polarizable molecular axis aligns along the laser polarization axis. With the help of this dynamic alignment, processes that take place in the molecule reference system succeed in studying in the laboratory reference system.

In the recently published experiments carried out by the MBI team together with colleagues from Lyon (France) and Lund (Sweden), this oscillating dipole could be observed directly. The molecule was ionized by an attosecond pulse train synchronized to the dipole-inducing near-infrared laser field. Depending on the phase between pump and probe pulse, the ionization probability was significantly different. When the flashes of the attosecond pulse train were synchronized to the zero crossing of the near-infrared field (Figure 1 (a)), the ionization yield was smaller than when the flashes were synchronized with the extremes of the near-infrared field (Figure 1 (b)). The reason for this is the conservation of energy and momentum. Provided the photon energy in the attosecond flash is sufficient, it is much easier to release electrons from the molecule that move close to the atomic nuclei. However, this probability of residence is periodically changed by the near-infrared field (Figure 1 (c)), resulting in a modulation of the ionization interaction cross-section (Figure 2).

Fig. 2 Measured variation in ionization efficiency as a function of pump-sample delay (black). The molecules (molecular nitrogen, carbon dioxide and ethene) were ionized by an attosecond pulse train under the influence of a near-infrared field. The red line is the result of Fourier analysis near double the frequency of the near-infrared laser.

The pump-probe experiment was performed for various molecules, with the o.g. Effect changed linearly with the polarizability of the molecules (Figure 2). The experimental arrangement can be interpreted as the first implementation of the Stark spectroscopy on the attosecond time scale. With this method, the response of the molecule to an external electric field is optically interrogated. Future experiments of the MBI team will deal with transient absorption with attosecond time resolution. In addition, optical attosecond pulses are to be used to observe intrinsic charge movements in molecules that were not induced by an external laser field.

Original publication

Probing time-dependent molecular dipoles on the attosecond time scale

Ch. Neidel, J. Klei, C.-H. Yang, A. Rouzée, M. J. J. Vrakking, K. Klünder, M. Miranda, C. L. Arnold, T. Fordell, A. L'Huillier, M. Gisselbrecht, P. Johnsson, M. P. Dinh, E. Suraud, P.-G. Reinhard, C. Despré, M. A. L. Marques, F. Lépine

Physical Review Letters 111 (2013) 033001/1-5