Hello again,

When you pass light through two slits the wave-like property of the light causes it to diffract and interfere with itself in a similar way to sound waves. The coincidences of two peaks or two troughs cause bright bands (constructive interference), and the coincidences of (partial) matched peaks and troughs causes dark bands (destructive interference).

Of course we don't need two slits to do this. Any two sources of monochromatic light should work. The interference pattern will depend on the frequencies of the light sources and the relative phase angle.

Well then I thought about the light coming from a distant, light-emitting object like a star, where the number of sources of photons is so large that we can assume it tends to infinity, and if those sources of photons are asynchronous, which I'm guessing they are because there doesn't seem to be any logical or practical explanation as to how or why they shouldn't be, then it seems to me that the photons being emitted by the star in our direction should have variable, discrete frequencies (quantised by the energy lost by the electrons), but equally-random phase, and therefore surely the equally-randomly phased photons would cancel each other out by destructive interference?

Sorry for being so complicated.

I hope somebody can understand what I'm getting at.

clop

This "coincidence" is called coherence and is a necessary condition for interference.


This randomness means that the light is not coherent. Without coherence there is no interference, neither constructive nor destructive.

__________________
papageno


"Why waste time learning, when ignorance is instantaneous?" - Hobbes (Calvin and Hobbes)

"It's all about context!" - Vince Noir (The Mighty Boosh)

"I've never heard of such a brutal and shocking injustice that I cared so little about!" - Zapp Brannigan (Futurama)

Of course there is interference without coherence, just not neat static interference bands like you might see in a lab.

If you shine 1,000,000,000,000,000 photons with exactly the same frequency, and random phase. along a miniscule corridor for thousands of light years might you not expect the electromagnetic vector orthogonal to the direction of propogation to average out to zero.

clop

Actually, the answer is, they do interfere destructively. However, there is always some random signal left. To see this, imagine a random walk process. A random walk is like adding a bunch of steps of equal size but different directions, which is just like interference of incoherent photons. If you take N steps, however, you don't end up back where you started-- that would be completely destructive interference. You end up, on the average, the square root of N times the stepsize away from where you started. How does this correspond to photons? With photons, it is the amplitudes from each microscopic source that you are adding. If you have N sources, the destructive interference that you are asking about ends up giving you, on the average, an amplitude size of the square root of N. That sounds weird-- don't the sources just add? Well, yes, because the energy you get, which is the number of photons, depends on the square of the amplitude-- or, N! So N incoherent sources get you N photons, on the average (especially if N is extremely large-- look up "Poisson statistics").

Note how this is different from a laser. With a laser, if you have N sources, they add coherently and you get an amplitude of size N. But that's N squared photons! Now you know why lasers are used in surgery and not incandescent light bulbs...

That "neat static interference" is what we usually call interference.
Random linear superposition of waves is not considered interference.

No, I don't expect the EM field to average to zero because the phases are random.
See Ken G's post.

__________________
papageno


"Why waste time learning, when ignorance is instantaneous?" - Hobbes (Calvin and Hobbes)

"It's all about context!" - Vince Noir (The Mighty Boosh)

"I've never heard of such a brutal and shocking injustice that I cared so little about!" - Zapp Brannigan (Futurama)

Actually, that was a question I had in my earlier physics classes (one of many many unanswered questions about how things worked - my teachers considered me quite annoying). Thanks, KenG.

__________________
http://amssolarempire.blogspot.com

Questions like this one are what keep us really learning. Any teacher annoyed by such questions really ought to reconsider what they think they are doing for a living, although I suppose there more, and less, appropriate moments to broach such seemingly sticky questions...

I don't have a profound answer, but here is some more grist for the mill. Unless quantum physics rejects the idea of a photon; photons can be produced at audio frequencies, but we can't hear them unless they are converted by a radio receiver and speaker, to condensations and rarefractions in air. I'm saying light and sound are fundamentally different forms of enegy, so some of the characteristics likey do not apply to both. I suspect the cancelation and reinforcement occur less in radio waves and light, but they do occur in radio when multipath reception occurs. This is heard as a tremelo applied to the music, more often as a much slower fading in and out.
In the case of a point source in a vacuum. the almost parallel photons diverge over billions of light years, so they do not reinforce or cancel, but stars a few seconds of arc away can (but likely rarely) produce cancelation or reinforcement as this would (I think) blur the image. Neil