Clocking the Movements of Atoms

Roland Piquepaille writes "With special microscopes, scientists and engineers involved in nanotechnologies have been able to 'see' atoms for a while. But they couldn't clock the atomic response to events which typically occur in nanoseconds. Now, U.S. physicists have found a way to clock the movements of atoms at the nanometer scale. In their experiments, they were able to literally watch atoms switching positions in ferroelectric materials. Adding the dimension of time to the observation of the nanoworld could lead to easier developments of 'materials for improved memory applications in microelectronics.'"

Comments

BS

[Mr. Underbridge]

With special microscopes, scientists and engineers involved in nanotechnologies have been able to 'see' atoms for a while. But they couldn't clock the atomic response to events which typically occur in nanoseconds.

*Femto*second laser spectroscopy has been available for some time now to investigate chemical reactions that happen much faster than nanoseconds. Got the Nobel Prize in Chemistry in 1999 for Zewail.

Re:BS

[kebes]

You're quite right. Femtosecond laser spectroscopy has already pushed the time resolution way beyond what this new report is claiming. In fact attosecond pulses have even been generated, making it possible not only to measure motion of atoms, but motion of electrons within atoms during a chemical reaction.

However, it should be noted that this new report uses X-ray microdiffraction. They obtain picosecond time resolution, simultaneous with structural information. The article summary is not especially exact, but the structural information (positions of atoms) of this technique is indeed unique. Femtosecond laser spectrscopy will return quite a bit of information, but it doesn't really tell you the movement of the atoms in a bulk sample (only movement of atoms in relation to each other, during the chemical reaction). This new technique appears to be applicable to simultaneously deducing the locations and motions of atoms in solids (as opposed to gases/plasmas).

The time resolution is nothing amazing, but it is nevertheless quite impressive that they can deduce the motion of a domain wall in a solid with picosecond time resolution.

For anyone interested, the actually paper in question is:
Grigoriev et al. Physical Review Letters (12 May 2006), 187601. DOI 10.1103/PhysRevLett.96.187601

Re:BS

[XchristX]

The article can be obtained from Grigoriev's webpage itself:

PDF WARNING!!!

http://homepages.cae.wisc.edu/~alexey/PRL06_grigor iev.pdf

thanks, that's helpful

[Goldsmith]

There are already plenty of ways to "see" the motions of atoms which do not require a very expensive national X-ray source. This technique is not comparable to AFM and STM, which are now cheap enough to use in teaching labs.

What about simulations?

[brian0918]

Maybe we couldn't physically see how atoms are moving over short time intervals, but we've been able to simulate it for quite a while. I used to run simulations at Sandia National Labs in New Mexico that looked at the interactions of a couple dozen atoms on a femtosecond scale.