2
$\begingroup$

I was watching a video about LCLS 2 and they mentioned that they can shoot and record up to 1 million x-ray pulses allowing them to see chemical reactions. They said that these pulses of x-ray last only for a few femtoseconds and if one femtosecond is 1x10^-15 how can we record that. Light only travels at approximately 300000 m/s so there is no way light can travel that fast for a camera, or sensor to see it. if an electron wanted to travel 1 meter, it would take it 3x10^-6 seconds which is still orders of magnitude slower than a femtosecond. This is especially true considering that electrons also travel at the speed of light. If that is true, how can a computer possibly record data that lasts 1 femtosecond?

The video I refer to: https://www.youtube.com/watch?v=X7CoLCTeqpE

$\endgroup$
5
  • 5
    $\begingroup$ Well, the sensor gets triggered when the light gets there. The time it takes is pretty irrelevant - telescopes take pictures using light generated billions of years ago. $\endgroup$
    – Jon Custer
    Jun 28, 2021 at 20:17
  • $\begingroup$ Correct but the computer needs to be able to take in data coming in at the speed of 1 femtosecond. $\endgroup$ Jun 28, 2021 at 20:20
  • 1
    $\begingroup$ No, no it doesn't. Depends very much on what data it is actually taking and how they are taking it. Often you just build up counts in, e.g., a CCD detector, and then dump to record. I've done pump/probe experiments with femotsecond lasers and the data acquisition was pretty much accumulating counts with a spectrometer. $\endgroup$
    – Jon Custer
    Jun 28, 2021 at 20:28
  • $\begingroup$ I can see why you aren't getting answers. That video wasn't as informative as you might hope. I don't have one that answers your question. But at least here are a few from Don Lincoln at Fermilab on how accelerators work. Accelerator Science: Why RF?, Accelerator Science: Circular vs. Linear, $\endgroup$
    – mmesser314
    Jun 29, 2021 at 1:21
  • $\begingroup$ Accelerator Science: Proton vs. Electron. There are more. Google "don lincoln playlist" $\endgroup$
    – mmesser314
    Jun 29, 2021 at 1:23

1 Answer 1

10
$\begingroup$

There is a lot of confusion in this question. Let me try to clear up some of it.

First, the LCLS II is not yet running. The LCLS II is expected to begin operation in 2022. However, SLAC is doing experiments with the original LCLS (Linac Coherent Light Source). This delivers pulses of X-rays at 120 Hz (120 pulses per second). Each pulse is ten billion times more intense than an X-ray pulse from a synchrotron, so it is possible to take a complete X-ray diffraction pattern for a molecule with each pulse. Experimental groups have used the LCLS to make movies of chemical reactions.

Chemical reactions take place on 100 femtosecond (fs) to picosecond (ps) timescales, so an X-ray pulse with a length of a few fs will take a static diffraction pattern image of a molecule at an intermediate stage of a chemical reaction. These pictures can be converted to pictures of the atoms in real space and then assembled to form a movie. To take the picture, the diffracted X-rays go out from the sample to an array of detectors. Once the X-rays are absorbed by the detectors, there is plenty of time (1/120 sec) for the detectors to process and output these signals. However, it is not unreasonable that properly designed detectors could take 1 million pictures per second. The tracking detectors at the CERN Large Hadron Collider take simpler pictures at the rate of 40 million pictures per second.

The LCLS is a free-electron laser. That is, the X-rays are generated by wiggling an electron beam and forcing it to emit synchrotron radiation. The length of the X-ray pulse is set by the length of the electron pulse that generates it. The speed of light is, in convenient units, 1 foot (0.3 m) per nanosecond, so a few femtosecond pulse corresponds to an electron pulse of length about 1 micron. It is tricky to compress a pulse of electrons to this size, but not as tricky as it might appear. The natural electron bunches in the SLAC linear accelerator are about 100 microns in length.

To summarize: At regular intervals (now 120 per second, soon 1 million per second) the accelerators at SLAC produce very short bunches of electrons that in turn create very short bunches of X-rays. These are intense enough to take a picture of a molecule with each pulse. The pulses are short enough to capture molecules in the intermediate stages of chemical reactions. The resulting pictures can be turned into movies.

The person who asked this question saw only a very brief video about the LCLS, but SLAC has longer videos that provide more information. Many of these are on the SLAC Public Lecture web page:

https://www6.slac.stanford.edu/community/past-lectures

For better understanding of this question, I especially recommend the lectures from 4/11/2017 (by Raymond Sierra) and 9/29/2015 (by Mike Minitti). The lecture from 11/17/2009 (by Daniel Ratner) explains how the LCLS works.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.