General · Language · Media · Site

Imaging a light pulse?

Reference: The 2018-2019 Watson Lecture Series, Caltech, Beckman Auditorium

World’s Deepest-Penetration and Fastest Optical Cameras
by Lihong Wang
Wednesday, November 28, 2018, 8:00 PM – 9:30 PM

Lihong Wang will discuss the development of photoacoustic tomography, which allows scientists to peer deep into biological tissue. He will also talk about his lab’s development of compressed ultrafast photography that records 10 trillion frames per second.

I came across some articles recently that ultrafast photography reached a point where a light pulse can be imaged. I was interested because in the past claims were made that atomic events — like photon emission — were instantaneous or essentially so because such events were too fast to ever image. Well, that may change. And there are signs that “quantum leaps” do indeed take a slice of time (so there is some physical process in space-time).

These demonstrations, like in most entanglement correlation experiments, are post hoc visualizations of assumed identical events repeated trillions of times.

YouTube: “Lihong Wang: World’s fastest camera for ultrafast phenomena provides temporal information” (published March 20, 2017). Photonic version of Mach cone — using laser light to image light propagating in a background medium.

[From transcript, somewhat reconstructed.] If we have a background medium of some sort, the speed of light will be reduced in that medium relative to the speed of light in vacuum. We created a tunnel where the speed of light in that tunnel is greater than the speed of light in a medium. We propagate a light pulse — a very short light pulse — in that medium … [which] scatters within the tunnel [and] generates secondary light sources and those light sources will be propagated in the background medium; so, the light source will propagate at a greater speed and that creates a superluminal light source.

YouTube: TED “Imaging at a trillion frames per second | Ramesh Raskar” (published on Jul 26, 2012).

Ramesh Raskar presents femto-photography, a new type of imaging so fast it visualizes the world one trillion frames per second, so detailed it shows light itself in motion. This technology may someday be used to build cameras that can look “around” corners or see inside the body without X-rays.

[Selected comments to clarify context of what was presented.]

1. Who … was his audience? These people just saw the speed of light, the fastest speed there is in our world. And they saw it slowed down enough to see it move in SLOW MOTION! Do they not understand what they just saw? The audience at the iPhone 5 [event] was more excited than these people. This is easily one of the most incredible advances in science! The ability to see the speed of light… it’s pretty … awesome people, get hype!

2. They aren’t really taking a video of a single event. They are repeating the event a trillion times and taking frames at different instances in time to get one full event. So if you think you can use this as a slow-mo camera to capture a water balloon bursting, you better have a LOT of water balloons …

3. I think some of you may be a little bit misled by this video. You are not seeing a particular beam of light move through a particular bottle in a single event. You are seeing the motion of light of that particular circumstance.

4. It took thousands of measurements over several minutes, hours, or days to compile that short clip. Never the less, the clip does show the movement of light and it is backed by data.


Other notes:

Wiki: Picosecond

A picosecond is an SI unit of time equal to 10−12 or 1/1,000,000,000,000 (one trillionth) of a second.

One picosecond is equal to 1000 femtoseconds, or 1/1000 nanoseconds. Because the next SI unit is 1000 times larger, measurements of 10−11 and 10−10 second are typically expressed as tens or hundreds of picoseconds. Some notable measurements in this range include:

  • 1 picosecond – time taken by light in a vacuum to travel approximately 0.30 mm
  • 1 picosecond – half-life of a bottom quark
  • picoseconds to nanoseconds – phenomena observable by dielectric spectroscopy
  • 1.2 picoseconds – switching time of the world’s fastest transistor (845 GHz, as of 2006)
  • 3.3 picoseconds (approximately) – time taken for light to travel 1 millimeter

Wiki: Chirped pulse amplification

Chirped pulse amplification (CPA) is a technique for amplifying an ultrashort laser pulse up to the petawatt level with the laser pulse being stretched out temporally and spectrally prior to amplification [thereby avoiding problematical nonlinear effects impacting the beam or laser’s components].

Chirped-pulse amplification was originally invented as a technique to increase the available power in radar in 1960. CPA for lasers was introduced by Gérard Mourou and Donna Strickland at the University of Rochester in the mid 1980s, a work for which they received the Nobel Prize in Physics in 2018.

In CPA … an ultrashort laser pulse is stretched out in time prior to introducing it to the gain medium using a pair of gratings that are arranged so that the low-frequency component of the laser pulse travels a shorter path than the high-frequency component does. After going through the grating pair, the laser pulse becomes positively chirped, that is, the high-frequency component lags behind the low-frequency component, and has longer pulse duration than the original by a factor of 1000 to 100000.

Then the stretched pulse, whose intensity is sufficiently low compared with the intensity limit of gigawatts per square centimeter, is safely introduced to the gain medium and amplified by a factor of a million or more. Finally, the amplified laser pulse is recompressed back to the original pulse width through the reversal process of stretching, achieving orders of magnitude higher peak power than laser systems could generate before the invention of CPA.

In addition to the higher peak power, CPA makes it possible to miniaturize laser systems (the compressor being the biggest part). A compact high-power laser, known as a tabletop terawatt laser (T3 laser), can be created based on the CPA technique. (October 2, 2018): First experiments at new X-ray laser reveal unknown structure of antibiotics killer

An international collaboration led by DESY and consisting of over 120 researchers has announced the results of the first scientific experiments at Europe’s new X-ray laser, European XFEL.

The 3.4 kilometres long European XFEL is designed to deliver X-ray flashes every 0.000,000,220 seconds (220 nanoseconds). To unravel the three-dimensional structure of a biomolecule, such as an enzyme, the pulses are used to obtain flash X-ray exposures of tiny crystals grown from that biomolecule. Each exposure gives rise to a characteristic diffraction pattern on the detector. If enough such patterns are recorded from all sides of a crystal, the spatial structure of the biomolecule can be calculated. The structure of a biomolecule can reveal much about how it works.

every crystal can only be X-rayed once since it is vaporised by the intense flash (after it has produced a diffraction pattern). So, to build up the full three-dimensional structure of the biomolecule, a new crystal has to be delivered into the beam in time for the next flash, by spraying it across the path of the laser in a water jet. Nobody has tried to X-ray samples to atomic resolution at this fast rate before. … To probe biomolecules at full speed, not only the crystals must be replenished fast enough—the water jet is also vaporised by the X-rays and has to recover in time.

To record X-ray diffraction patterns at this fast rate, an international consortium led by DESY scientist Heinz Graafsma designed and built one of the world’s fastest X-ray cameras, tailor-made for the European XFEL. The Adaptive Gain Integrating Pixel Detector (AGIPD) can not only record images as fast as the X-ray pulses arrive, it can also tune the sensitivity of every pixel individually, making the most of the delicate diffraction patterns in which the information on the structure of the sample is encoded.

” … Experiments that used to take hours can now be done in a few minutes, …”

Serial femtosecond X-ray crystallography (SFX) is a powerful method to determine the atomic structure of a sample, typically a biomolecule like a protein. It builds on classic crystallography, which was developed more than a century ago. … The brief, but extremely bright flashes of X-ray lasers like the European XFEL overcome two problems at the same time: They are bright enough to produce usable diffraction patterns even from the smallest crystals, and they are short enough to outrun the radiation damage of the crystals. A typical X-ray laser flash lasts only a few femtoseconds (quadrillionths of a second) and has left the crystal before it is vaporised. This “diffraction before destruction” method produces high-quality diffraction pattern even from tiny crystals. But as every crystal is vaporised in a single flash, a new crystal has to be X-rayed with every flash. Therefore, the scientists spray thousands of randomly oriented protein crystals into the path of the X-ray laser and record series of until they have gathered enough data to calculate the protein’s structure with atomic resolution.


  1. So, in femto-photography, how many photons are in a laser pulse 1 millimeter (mm) in length, that is, in a 3.3 picosecond pulse?

6 thoughts on “Imaging a light pulse?

  1. Regarding temporal coherence of lasers, LIGO uses a complex laser generation system to achieve required quality: LIGO’s Laser.

    Frequency and Power Stability

    Despite the purity and stability of the initial beam, uponexiting the HPO it is still not stable enough for use. LIGO’s laser needs to be 100-million times more stable than it is intrinsically. To achieve this unprecedented level of stability, the beam’s natural frequency variations (i.e. its inability to continually radiate a single, discrete color of light) and power fluctuations are mechanically reduced by about a factor of 100-million through a series of feedback mechanisms before the laser is used in the interferometer. This whole process is akin to tuning the world’s most complex piano.

  2. My posts Imaging a light pulse? and A photon’s frame of spacetime — no rest for the massless discussed the landscape of light speed. Live Science recently posted an article “Here’s What the Speed of Light Looks Like in Slow Motion” (March 29, 2019) about a visit to the Caltech lab by The Slow Mo Guys where light pulses can be visualized at 10 trillion frames per second. The article includes Gav and Dan’s video “Filming the Speed of Light at 10 Trillion FPS” — a quick overview of the complex equipment and visualization, each of which involved hours of processing — far from instant photography.

    While the camera on your phone takes two-dimensional photos, T-CUP is a type of streak camera, which records images in a single dimension, very very quickly. Unlike prior streak cameras, which create composite images of light by recording different horizontal slices of laser over multiple laser pulses, the T-CUP is able to image an entire laser pulse in a single frame. It does this by diverting the laser beam to two different cameras simultaneously, then using a computer program to combine the two images.

    Light at 10 Trillion FPS

  3. Regarding my speculation that “there are signs that ‘quantum leaps’ do indeed take a slice of time (so there is some physical process in space-time),” I found this article “The discrete-time physics hiding inside our continuous-time world” (April 15, 2019) interesting. The article also was a refresh on Markov chains.

    In a pair of papers, one appearing in this week’s Nature Communications and one appearing recently in the New Journal of Physics, physicists at the Santa Fe Institute and MIT have shown that in order for such two-time dynamics over a set of “visible states” to arise from a continuous-time Markov process, that Markov process must actually unfold over a larger space, one that includes hidden states in addition to the visible ones. They further prove that the evolution between such a pair of times must proceed in a finite number of “hidden timesteps”, subdividing the interval between those two times. (Strictly speaking, this proof holds whenever that evolution from the earlier time to the later time is noise-free—see paper for technical details.)

    “We’re saying there are hidden variables in dynamic systems, implicit in the tools scientists are using to study such systems,” says co-author David Wolpert (Santa Fe Institute). “In addition, in a certain very limited sense, we’re saying that time proceeds in discrete timesteps, even if the scientist models time as though it proceeds continually. …”

    Wiki: A Markov chain is a stochastic model describing a sequence of possible events in which the probability of each event depends only on the state attained in the previous event. … it is common to define a Markov chain as a Markov process in either discrete or continuous time with a countable state space (thus regardless of the nature of time) … Markovian systems appear extensively in thermodynamics and statistical mechanics, whenever probabilities are used to represent unknown or unmodelled details of the system, if it can be assumed that the dynamics are time-invariant, and that no relevant history need be considered which is not already included in the state description. … The paths, in the path integral formulation of quantum mechanics, are Markov chains.

  4. An update on research by Caltech’s Lihong Wang [Bren Professor of Medical Engineering and Electrical Engineering] in phase-sensitive compressed ultrafast photography (pCUP):

    Caltech > About > News > “Ultrafast Camera Takes 1 Trillion Frames Per Second of Transparent Objects and Phenomena” (January 17, 2020).

    Wang explains that his new imaging system combines the high-speed photography system he previously developed with an old technology, phase-contrast microscopy, that was designed to allow better imaging of objects that are mostly transparent such as cells, which are mostly water.

    The fast-imaging portion of the system consists of something Wang calls lossless encoding compressed ultrafast technology (LLE-CUP). Unlike most other ultrafast video-imaging technologies that take a series of images in succession while repeating the events, the LLE-CUP system takes a single shot, capturing all the motion that occurs during the time that shot takes to complete. Since it is much quicker to take a single shot than multiple shots, LLE-CUP is capable of capturing motion, such as the movement of light itself, that is far too fast to be imaged by more typical camera technology.

    In the new paper, Wang and his fellow researchers demonstrate the capabilities of pCUP by imaging the spread of a shockwave through water and of a laser pulse traveling through a piece of crystalline material.

  5. In quantum physics, mathematically discontinuous changes – or jumps – between quantum states are popularly referred to as quantum leaps (remember the TV series). As to whether quantum leaps are instantaneous (zero time) or not – as well as random (without any harbinger), here’s another article which discusses these foundational questions.

    Quanta Magazine > Quantum Physics > “Quantum Leaps, Long Assumed to Be Instantaneous, Take Time” by Philip Ball (June 5, 2019) – An experiment caught a quantum system in the middle of a jump — something the originators of quantum mechanics assumed was impossible. Image tag: “A quantum leap is a rapidly gradual process.

    Ball notes that while Niels Bohr, Werner Heisenberg, et al., assumed jumps were instantaneous, Erwin Schrödinger thought otherwise: wave functions change “only smoothly and continuously over time, like gentle undulations on the open sea.”

    The argument wasn’t just about Schrödinger’s discomfort with sudden change. The problem with a quantum jump was also that it was said to just happen at a random moment — with nothing to say why that particular moment. It was thus an effect without a cause, an instance of apparent randomness inserted into the heart of nature. Schrödinger and his close friend Albert Einstein could not accept that chance and unpredictability reigned at the most fundamental level of reality. According to the German physicist Max Born, the whole controversy was therefore “not so much an internal matter of physics, as one of its relation to philosophy and human knowledge in general.” In other words, there’s a lot riding on the reality (or not) of quantum jumps.

    Ball highlights earlier research exploring this topic in 1986 and 2007.

    The current Yale research is based on a quantum system of “artificial atoms” and their ground and excited states. And also uses an optical cavity, something which I’ve noticed as a mainstay in more and more studies. And a new method for me, namely, tomographic reconstruction. And another reference to quantum trajectories theory.

    A new experiment shows that they aren’t. By making a kind of high-speed movie of a quantum leap [in a single artificial atom], the work reveals that the process is as gradual as the melting of a snowman in the sun. “If we can measure a quantum jump fast and efficiently enough,” said Michel Devoret of Yale University, “it is actually a continuous process.” The study, which was led by Zlatko Minev, a graduate student in Devoret’s lab, was published on Monday in Nature.

    But there’s more. With their high-speed monitoring system, the researchers could spot when a quantum jump was about to appear, “catch” it halfway through, and reverse it, sending the system back to the state in which it started. In this way, what seemed to the quantum pioneers to be unavoidable randomness in the physical world is now shown to be amenable to control.

    The Yale researchers needed to find an indirect way to extract information about the state of the system, since a direct interrogation / measurement on a quantum system destroys the coherence of the wave function – “collapses” the wave function to an observable state. They needed a way to explore the dynamics of the superposed state itself.

    “Devoret and colleagues employ a clever trick involving a second excited state.” They probe an auxiliary state coupled to the primary state of interest. Whether the experimental system is in that auxiliary state or not elicits information about the primary state. In certain contexts, something not occurring can be telling. For example (my metaphor, not Ball’s), the fact that a normally functioning door (sensor) alarm was not heard during a certain period of time tells us that the door remained closed – it was not opened.

    “Absence of an event can bring as much information as its presence,” said Devoret.

    But their strategy also involved detection of precursory signals “somewhat like [for] volcanic eruptions.” There’s a level of randomness as to the exact time but with a lead-up over a probabilistic interval.

    Each [volcanic] eruption happens unpredictably, but some big ones can be anticipated by watching for the atypically quiet period that precedes them.

    That warning allowed the researchers to study the jump in greater detail. When they saw this brief pause, they switched off the input of photons driving the transitions. Surprisingly, the transition to the dark state still happened even without photons driving it — it is as if, by the time the brief pause sets in, the fate is already fixed. So although the jump itself comes at a random time, there is also something deterministic in its approach.

    With the photons turned off, the researchers zoomed in on the jump with fine-grained time resolution to see it unfold. Does it happen instantaneously — the sudden quantum jump of Bohr and Heisenberg? Or does it happen smoothly, as Schrödinger insisted it must? And if so, how?

    The team found that jumps are in fact gradual.

    The techniques developed by the Yale team reveal the changing mindset of a system during a quantum jump. Using a method called tomographic reconstruction, the researchers could figure out the relative weightings of the dark and ground states in the superposition. They saw these weights change gradually over a period of a few microseconds. That’s pretty fast, but it’s certainly not instantaneous.

    Ball concludes, “Yes, it [the quantum world] is shot through with randomness — but no, it is not punctuated by instantaneous jerks. Schrödinger, aptly enough, was both right and wrong at the same time.”

  6. Moving beyond flat imaging up to 70 trillion frames per second …

    Another update on research by Caltech’s Lihong Wang [Bren Professor of Medical Engineering and Electrical Engineering] in single-shot stereo-polarimetric compressed ultrafast photography (SP-CUP), adding 3D movies at 100 billion frames per second.

    • Caltech > About > News > “Ultrafast Camera Films 3-D Movies at 100 Billion Frames Per Second” (October 16, 2020).

    In CUP technology, all of the frames of a video are captured in one action without repeating the event. This makes a CUP camera extremely quick (a good cell-phone camera can take 60 frames per second). Wang added a third dimension to this ultrafast imagery by making the camera “see” more like humans do.

    When a person looks at the world around them, they perceive that some objects are closer to them, and some objects are farther away. Such depth perception is possible because of our two eyes, each of which observes objects and their surroundings from a slightly different angle. The information from these two images is combined by the brain into a single 3-D image.

    The camera is stereo now,” he says. “We have one lens, but it functions as two halves that provide two views with an offset. Two channels mimic our eyes.

    Just as our brain does with the signals it receives from our eyes, the computer that runs the SP-CUP camera processes data from these two channels into one three-dimensional movie.

    SP-CUP also features another innovation that no human possesses: the ability to see the polarization of light waves [“our eyes cannot detect the polarization of light directly”].

    … he hopes that it will help researchers better understand the physics of sonoluminescence, a phenomenon in which sound waves create tiny bubbles in water or other liquids. As the bubbles rapidly collapse after their formation, they emit a burst of light.

    “Some people consider this one of that greatest mysteries in physics,” he says. “When a bubble collapses, its interior reaches such a high temperature that it generates light. The process that makes this happen is very mysterious because it all happens so fast, and we’re wondering if our camera can help us figure it out.”

Comments are closed.