5 ms·

I still don't get it, 10,000 times more light than its size?

21 9y ago
They say it was a long exposure. The atom was illuminated and irradiated light for a long enough time to register on the sensor.
- thinkloop 9y ago
  Even if the exposure were for a year, if the amount of light being emitted is less than the hardware can perceive, it wouldn't register - hence the need for electron-microscopes and such. There's a story here about the size of the light being emitted that doesn't seem satisfactorily answered (not by you, the article and stuff in general).
  - brazzledazzle 9y ago
    Maybe it's light reflected by the atom at multiple points? The article mentions that it's "held nearly motionless" but over a long exposure perhaps the degree of movement allowed created a composite large enough to show up in the photo.
    - ars 9y ago
      No, that's not it. Just because it is small, doesn't make it invisible. You can't see single atoms because they normally don't send any light toward you. But if they did, and this one is, you would see that light. You aren't seeing the atom exactly, you are seeing what it sends toward you, and the human eye can see single photons. (And so can cameras.)
      - albedoa 9y ago
        > No, that's not it. That has to be at least partially it for the rest of your comment to be right (which it is!). To say the atom was "held nearly motionless" is to say it moved a little during the exposure. It sent the light toward the sensor from all of its positions.
        ars 9y ago
        > That has to be at least partially it No, not at all. The wavelength of the light is about 1,500 times larger than the atom emitting it. So the atom would need to move at least 1,500 diameters to make any difference in the light. It doesn't move that much.
        dingo_bat 9y ago
        > It doesn't move that much. Source?
        ars 9y ago
        Here: https://en.wikipedia.org/wiki/Laser_cooling https://en.wikipedia.org/wiki/Laser_cooling
  - ars 9y ago
    > Even if the exposure were for a year, if the amount of light being emitted is less than the hardware can perceive, it wouldn't register That is not true. It might be noisy, but it would show up. In point of comparison the human eye can see a single photon. A single atom can emit a single photon, ergo you can see a single atom. You won't get any detail out of it obviously, but you'll see it as a dot of light. That's what's happening here. > There's a story here about the size of the light being emitted that doesn't seem satisfactorily answered You are misunderstanding the "size". The size "10,000" is the resolution needed to differentiate between two atoms sitting next to each other. But in order to see the atom (detect the atom), you just need to be able to capture what it emits. And it emits photons, and it's not that hard to capture and record a single photon. So you do, and it shows up as a single dot in the image. If you had two atoms near each other, both emitting photons, you would not be able to distinguish them from each other (i.e. you couldn't say if there were one or two), unless they were farther apart.
    - BuildTheRobots 9y ago
      If we're seeing the single photon emitted by a single atom, then how much larger is the spot of light in the image compared to the atom that emitted it? Why isn't there a ruler somewhere in the picture? My brain is absolutely failing to make sense of the scales and sizes in this image.
      - ars 9y ago
        > then how much larger is the spot of light in the image compared to the atom that emitted it? The atom is about 255 picometers, the light is around 400 nanometers. That makes the light around 1500 times larger than the atom. (Which also helps explain why visible light can not distinguish atoms placed close together - it's so much larger than them.) > Why isn't there a ruler somewhere in the picture? My brain is absolutely failing to make sense of the scales and sizes in this image. The two electrodes are about 2mm apart. About the width (not length) of a sesame seed. It's really small. The photo was taken through a microscope. Another point of reference: the width between the electrodes is about 5,000 times the wavelength of the light being emitted. In theory, if you had a good enough camera, the dot of light would be around 1/5000 of the width. (However the diffraction limit of your lens might enter into play, blurring the image.) And finally, the width between the electrodes is about 8,000,000 times the size of the atom.
  - garmaine 9y ago
    Think about it this way: every photon your eyes perceive came from a single atom.
  - kevin_thibedeau 9y ago
    CCDs can detect single photons. It isn't much of a challenge. I would be suspicious that the surrounding apparatus is Photoshopped in from a separate exposure.
    - klickverbot 9y ago
      Nope, not photoshopped. It's a single exposure; the apparatus is illuminated using flashes. The latter made lighting the shot in a controlled fashion a bit easier than if I had used continuous sources – you'd be looking at using a torch with a bunch of filters or a computer monitor on the lowest brightness otherwise.

21 9y ago

They say it was a long exposure. The atom was illuminated and irradiated light for a long enough time to register on the sensor.

thinkloop 9y ago
Even if the exposure were for a year, if the amount of light being emitted is less than the hardware can perceive, it wouldn't register - hence the need for electron-microscopes and such. There's a story here about the size of the light being emitted that doesn't seem satisfactorily answered (not by you, the article and stuff in general).
- brazzledazzle 9y ago
  Maybe it's light reflected by the atom at multiple points? The article mentions that it's "held nearly motionless" but over a long exposure perhaps the degree of movement allowed created a composite large enough to show up in the photo.
  - ars 9y ago
    No, that's not it. Just because it is small, doesn't make it invisible. You can't see single atoms because they normally don't send any light toward you. But if they did, and this one is, you would see that light. You aren't seeing the atom exactly, you are seeing what it sends toward you, and the human eye can see single photons. (And so can cameras.)
    - albedoa 9y ago
      > No, that's not it. That has to be at least partially it for the rest of your comment to be right (which it is!). To say the atom was "held nearly motionless" is to say it moved a little during the exposure. It sent the light toward the sensor from all of its positions.
      - ars 9y ago
        > That has to be at least partially it No, not at all. The wavelength of the light is about 1,500 times larger than the atom emitting it. So the atom would need to move at least 1,500 diameters to make any difference in the light. It doesn't move that much.
        dingo_bat 9y ago
        > It doesn't move that much. Source?
        ars 9y ago
        Here: https://en.wikipedia.org/wiki/Laser_cooling https://en.wikipedia.org/wiki/Laser_cooling
- ars 9y ago
  > Even if the exposure were for a year, if the amount of light being emitted is less than the hardware can perceive, it wouldn't register That is not true. It might be noisy, but it would show up. In point of comparison the human eye can see a single photon. A single atom can emit a single photon, ergo you can see a single atom. You won't get any detail out of it obviously, but you'll see it as a dot of light. That's what's happening here. > There's a story here about the size of the light being emitted that doesn't seem satisfactorily answered You are misunderstanding the "size". The size "10,000" is the resolution needed to differentiate between two atoms sitting next to each other. But in order to see the atom (detect the atom), you just need to be able to capture what it emits. And it emits photons, and it's not that hard to capture and record a single photon. So you do, and it shows up as a single dot in the image. If you had two atoms near each other, both emitting photons, you would not be able to distinguish them from each other (i.e. you couldn't say if there were one or two), unless they were farther apart.
  - BuildTheRobots 9y ago
    If we're seeing the single photon emitted by a single atom, then how much larger is the spot of light in the image compared to the atom that emitted it? Why isn't there a ruler somewhere in the picture? My brain is absolutely failing to make sense of the scales and sizes in this image.
    - ars 9y ago
      > then how much larger is the spot of light in the image compared to the atom that emitted it? The atom is about 255 picometers, the light is around 400 nanometers. That makes the light around 1500 times larger than the atom. (Which also helps explain why visible light can not distinguish atoms placed close together - it's so much larger than them.) > Why isn't there a ruler somewhere in the picture? My brain is absolutely failing to make sense of the scales and sizes in this image. The two electrodes are about 2mm apart. About the width (not length) of a sesame seed. It's really small. The photo was taken through a microscope. Another point of reference: the width between the electrodes is about 5,000 times the wavelength of the light being emitted. In theory, if you had a good enough camera, the dot of light would be around 1/5000 of the width. (However the diffraction limit of your lens might enter into play, blurring the image.) And finally, the width between the electrodes is about 8,000,000 times the size of the atom.
- garmaine 9y ago
  Think about it this way: every photon your eyes perceive came from a single atom.
- kevin_thibedeau 9y ago
  CCDs can detect single photons. It isn't much of a challenge. I would be suspicious that the surrounding apparatus is Photoshopped in from a separate exposure.
  - klickverbot 9y ago
    Nope, not photoshopped. It's a single exposure; the apparatus is illuminated using flashes. The latter made lighting the shot in a controlled fashion a bit easier than if I had used continuous sources – you'd be looking at using a torch with a bunch of filters or a computer monitor on the lowest brightness otherwise.