Slashdot Mirror

Not more photons. See https://www.rp-photonics.com/s...
and http://scienceworld.wolfram.co... for their definition.

They get a peak brightness of about 10^19 photons s^-1 mm^-2 mrad^-2 (per 0.1% bandwidth) at 1 MeV photon energy. Peak means at the peak of a 30fs short pulse.

Actually, by your definition, yes. Their x-ray pulses have a rather low number of photons (about a million per shot). However, physicists use a different definition for brightness: http://scienceworld.wolfram.co...
and
https://www.rp-photonics.com/s...

That definition basically translates to "lots of light in one direction and in a narrow spectral range, per unit time" - squeezing the light in every imaginable way to make it as well-defined/focused as possible. This makes your LED lose in a lot of ways: because it's not pulsed, the light is all over the place instead of well collimated (straight like a laser), and oh yours are not x-ray photons - these carry about 10,000x more energy, each.

Their driving laser (Diocles) is another story though: it has more than 1 Joule of optical(!) pulse energy. That's probably more light in one pulse than your LED will generate in its entire life. And that comes in a pulse that's only 33 femtoseconds short, that's 0.000000000000033 seconds. Not even the biggest laser, though.

I'm not sure we have a transmitter though that we can blast over the distance and still be captured, so we can add a few more years to that.

In order for us to be able to measure a signal from a probe, it would have to be not just bright enough for us to detect it, it would also have to be bright enough to discernably change the light we get from the star.

This page says that it is possible to outshine a star for brief moments (few nanoseconds) using lasers: https://www.princeton.edu/~wil...

I've done some back of the envelope calculations to verify that. And while its totally wrong that one 10 000 th of the output of a star is 4 joules per ns, it should still be possible to build a laser that outshines proxima centauri.

According to wikipedia, proxima centauri has a luminosity of 0.0017 times the luminosity of the sun, which is 382.8 * 10^24 Watts. So it has 6.5 * 10^23 Watts of luminosity.

Let's assume the laser has a beam divergence of .1 millirad.
This page has an example for a red (1064 nm) laser, but we want to shoot a blue one as proxima centauri is mostly red so doesn't have much blue luminosity: https://www.rp-photonics.com/b...

On .1 milirad, the star would emit approx 2.5*10^-10 of its total output (2.2*10^-10 = (.1/(1000*pi*2)^2). That would mean 1.6*^10^14 Watts for proxima centauri.

If you say that .1% of the star's total emitted light is blue at the specific wavelength you are sending, you have to divide by 1000.

Per nanosecond, it would be 163 joule. Theoretically possible, but question is whether you can build a sender and receiver (and get the sender into the right place).

'Green lasers' are typically diode-pumped solid state lasers with a frequency-doubling crystal. It's not easy to make green light directly.

Remember the LASER has mirrors. Many of these mirrors are multi layer tuned reflectors. The mirrors in lasers are much more reflective than your bathroom mirror, but only at the wavelength of operation. More reading here;
http://www.rp-photonics.com/dichroic_mirrors.html
http://en.wikipedia.org/wiki/Dielectric_mirror

Unfortunately making a plane covered with this is not practical.

I thought it was quite common to express the capacity of an optical system by its bandwidth-distance product... Or are we talking about something different here?

The author of that article actually mentioned that we have been able to make green lasers, but that they are not efficient enough to be used.

Actually, the author of that summary mentioned that we have been able to make green diode lasers, but they are not efficient enough to be used for applications that need high efficiency. (they're used all the time for applications that don't need high efficiency-- laser pointers, for example-- take a look at google).

The author of the summary failed to point out that green lasers using technologies other than semiconductor diode lasers have been avalable for decades.

Copper vapor lasers are quite efficient, actually, although argon ion lasers efficiencies are indeed pretty low. Doubled YAG lasers are very commonly used for green-- a diode-pumped doubled YAG can get a wallplug efficiency of around 20%, IIRC.

I'm not worried about the amount of energy getting to the cylinder, that can just be brute forced as you note. I'm more concerned about what the energy that doesn't make it will do. Fiber fuse could be fairly dramatic in such a system. Video of fiber fuse propagating.

I don't doubt that they'll work it out in the end, engineers have a long history of being clever like that; but it is going to take a giant pile of tweaks on top of the naive implementation.

wouldnt the mirror in the laser get damaged if it absorbed ~20%

im pretty sure when i took 3rd yr optics and we made lasers (out of all the bits..) in the lab we used front silvered mirrors and i think they were over 99% reflective. i think more like >99.9%

http://www.rp-photonics.com/dielectric_mirrors.htm l
states laser mirrors can reflect over 99.9%

further it mentions 'supermirrors' with 99.9999%

(maby solid state ones might have been but they require a whole different set of tech) lasers werent possible until engineering of mirrors reached this level, although i believe einsteins paper on light where he discussed the three mechanisms of interaction between light and matter was a crucial idea which leads to the laser. so they were theoretically predicted decades before the mirrors were available.

of course they wouldnt last long outside of an extremely clean environment as they are front mirrored. although it might be possible to coat them with something.

Capturing and reemitting some of the energy at the spacecraft would be useful to alter the path, or else you only could go away from the homeworld (not exactly the most appealing flight profile :-) But that would require lining the sail with photoelectric cells, and making them so lightweight and so numerous that it looks to be even more impossible than the sail itself. Besides, this method won't allow you to reverse the thrust whatever you do.

I am not aware of any microwave lasers that the USSR might have been developing. Generally masers have very low power. The demo of the laser-powered flight was set up by some US scientists and shown on TV in North America.

This statement, and your criticism of the experiment, is based on out of date (or simply ill-researched) information, and it worries me that it got modded up to 5.

In this case, the group velocity is indeed faster than the speed of light - the form of the wavepacket peak (the speed of which is the definition of the group velocity [1]) travels through the fibre almost instantaneously, much faster than c. This is one of the two things about this experiment is interesting, as by the old-fashioned definition you are championing, information has just been transmitted faster than the speed of light (as has been done before [2], although I believe it was generally in quantum-tunneling type situations, rather than something as normal-seeming as a optical fibre.)

The significant point to take home from that part is that the "It's the group velocity that carries information" mantra is not strictly true. In this case, the leading edge of the pulse is all that is needed to reconstruct the whole thing, and then suddenly we're faced with a battle between our definition of information transportation at the group velocity (with the wave peak) and causality. Causality obviously wins, and information transportation needs a more complex definition than is covered in introductory optics courses.

References, cos I like that sort of thing:

[1] http://www.rp-photonics.com/group_velocity.html - definition of group velocity

[2] http://www.rp-photonics.com/superluminal_transmiss ion.html - article on superluminal transmission, including a reference to situations where the group velocity is greater than c.

This statement, and your criticism of the experiment, is based on out of date (or simply ill-researched) information, and it worries me that it got modded up to 5.

In this case, the group velocity is indeed faster than the speed of light - the form of the wavepacket peak (the speed of which is the definition of the group velocity [1]) travels through the fibre almost instantaneously, much faster than c. This is one of the two things about this experiment is interesting, as by the old-fashioned definition you are championing, information has just been transmitted faster than the speed of light (as has been done before [2], although I believe it was generally in quantum-tunneling type situations, rather than something as normal-seeming as a optical fibre.)

The significant point to take home from that part is that the "It's the group velocity that carries information" mantra is not strictly true. In this case, the leading edge of the pulse is all that is needed to reconstruct the whole thing, and then suddenly we're faced with a battle between our definition of information transportation at the group velocity (with the wave peak) and causality. Causality obviously wins, and information transportation needs a more complex definition than is covered in introductory optics courses.

References, cos I like that sort of thing:

[1] http://www.rp-photonics.com/group_velocity.html - definition of group velocity

[2] http://www.rp-photonics.com/superluminal_transmiss ion.html - article on superluminal transmission, including a reference to situations where the group velocity is greater than c.

No, it'd be you who has the two terms reversed. Group velocity is the one which defines the rate of information travel (usually), and is the one which is tricky (but not impossible) to get above c.

Phase velocity is not the 'speed of light itself', but the speed of an individual point on the wave profile, and it is trivially simple to get values greater than c - they arise naturally during X-ray propagation in metals, for example (it would even theoretically be possible using a mechanical system, if you had the time and inclination to build one).

See: http://www.rp-photonics.com/superluminal_transmiss ion.html

http://www.rp-photonics.com/group_velocity.html

No, it'd be you who has the two terms reversed. Group velocity is the one which defines the rate of information travel (usually), and is the one which is tricky (but not impossible) to get above c.

Phase velocity is not the 'speed of light itself', but the speed of an individual point on the wave profile, and it is trivially simple to get values greater than c - they arise naturally during X-ray propagation in metals, for example (it would even theoretically be possible using a mechanical system, if you had the time and inclination to build one).

See: http://www.rp-photonics.com/superluminal_transmiss ion.html

http://www.rp-photonics.com/group_velocity.html

http://www.rp-photonics.com/fiber_fuse.html

Basically light is absorbed at a point of damage in the fiber, creating a tiny plasma ball that burns backwards towards the source and destroys the fiber, preventing further output. The size of the plasma ball is very small (on the order of the size of the fiber core).

Maybe you just stuff some graphite between the bare fiber and the jacket. Then when the fiber core breaks it superheats a tiny bit of the graphite and begins the fiber fuse process - preventing further transmission of laser light.

And then maybe I just gave away a $1M invention.

-m