The investigators performed their experiment on Beamline 7.3.3. A femtosecond laser pulse synchronized to the individual electron bunches in the storage ring with a jitter of less than 5 picoseconds was made to overlap--in both space and time--a single x-ray pulse on the crystal. A monochromator selected x rays of wavelength 2.4 Å that illuminated a crystal oriented to diffract from the (111) planes onto the detector, a streak camera with 3-picosecond resolution. They then followed the structural dynamics of the crystal by monitoring the intensity of the diffracted x rays as a function of time with the camera. Since the length of the x-ray pulse was hundreds of times longer than the laser pulse, it served as a "continuous" source of x rays for diffraction before, during, and after the laser excitation of the sample.

Streak camera image of a single 60-picosecond x-ray pulse. Time runs from the upper left to the lower right.

Recording Atomic Vibrations

We all know that one cannot record the flapping wings of a hummingbird in flight with a camera with a slow shutter speed. The same principle holds for scientific experiments. To record the progress of dynamic processes, whether physical, chemical, or biological, requires that the detector be faster than the process; otherwise, the "picture" is fuzzy. Many of the processes of interest to those at the frontiers of research take place on a time scale of molecular vibrations, typically around 100 femtoseconds (100 quadrillionth of a second). X rays are increasingly being used to probe such processes. A group working at the ALS has directly observed atomic vibrations, as well as ultrafast structural changes, in solids with a high-speed x-ray detector known as a streak camera. During recording, the camera places data gathered at each instant along a line (hence the streak camera name), thereby converting temporal information to spatial and effectively recording the time evolution of the process all at once. Time-resolved observations of structural changes in this way should lead to greater understanding of the driving mechanisms behind the changes in many important materials and possibly hint at how to control them.

Impulsive laser excitation of the crystal initially created "hot" electrons whose energy quickly was transferred to lattice vibrations, thereby heating the crystal. Following the excitation, the intensity of diffracted x rays decreased as expected, owing to a shift in the wavelength of the Bragg peak as the heated lattice expanded. But following the decrease, the team observed distinct temporal oscillations in the diffracted intensity, indicative of coherent lattice motion, in contrast to the incoherently excited lattice vibrations of a crystal in thermal equilibrium. By slightly changing the angle of the crystal with respect to the incident x rays, different vibrational modes, or phonons, could be selected, thus mapping out part of the acoustic phonon dispersion relation. This new technique thus allows one to probe the vibrational properties of solids at frequencies up to 0.1 THz, even under extreme, highly nonequilibrium conditions, by directly watching the atoms collectively ring.

The oscillatory signal following laser excitation (left) is indicative of coherent, large-amplitude lattice vibrations whose frequency spectrum can be found by changing the diffraction angle relative to the Bragg peak (inset). Above a critical laser fluence, the sample is coherently driven into a disordered state (right).

Because the investigators were able to resolve in real time the transfer of energy from the carrier system to the lattice, important physical parameters such as the electron-acoustic phonon coupling time could be extracted by quantitatively fitting the data to models of the processes involved. In addition, the researchers found that a significant contribution to the excitation of the coherent phonon state was also due to a direct coupling between the carriers and the acoustic phonons through the deformation potential interaction.

Furthermore, the team found a close relationship between the excitation of this coherent phonon state and the disordering transition that occurred above a critical laser fluence. In particular, the diffracted x-ray signal disappeared on a time scale determined by a vibrational period, implying that each mode took one final collective swing in one direction before disordering. Further time-resolved observations of phase transitions in other materials (for example, strongly correlated systems) should lead to greater understanding of the driving mechanisms behind them.

Research conducted by A.M. Lindenberg, I. Kang, and S.L. Johnson (University of California, Berkeley); T. Missalla (ALS and Lawrence Livermore National Laboratory); P.A. Heimann and H.A. Padmore (ALS); Z. Chang and P.H. Bucksbaum (University of Michigan); J. Larsson (Lund Institute of Technology, Sweden); H.C. Kapteyn (University of Colorado); R.W. Lee (Lawrence Livermore National Laboratory); J.S. Wark (University of Oxford); and R.W. Falcone (University of California, Berkeley, and Berkeley Lab).

Research funding: Office of Basic Energy Sciences (BES), U.S. Department of Energy; ILSA at Lawrence Livermore National Laboratory; Lawrence Berkeley National Laboratory; and National Science Foundation. Operation of the ALS is supported by BES.

Publication about this research: A.M. Lindenberg, I. Kang, S.L. Johnson, T. Misalla, P.A. Heimann, Z. Chang, J. Larsson, P.H. Bucksbaum, H.C. Kapteyn, H.A. Padmore, R.W. Lee, J.S. Wark, and R.W. Falcone, "Time-Resolved X-Ray Diffraction from Coherent Phonons during a Laser-Induced Phase Transition," Phys. Rev. Lett. 84, 111 (2000).