Mingan on 3/10/2007 at 03:40
I'm most probably oversimplifying, but when an atom absorbs a photon (usually by an electron) it's because that photon was at a precise energy level (thus wavelength) that allow the electron to shift to a higher orbital. The photon is absorbed (it dies. OH NOES), but the 'excited' electron cannot stay on it's excited orbital, so it shifts back to it's original orbital by losing the energy gained from the former photon. That energy's got to go somewhere, thus the electron "reemits" a photon (how photons are borned).
Yakoob on 3/10/2007 at 04:31
Quote Posted by Mingan
I'm most probably oversimplifying, but when an atom absorbs a photon (usually by an electron) it's because that photon was at a precise energy level (thus wavelength) that allow the electron to shift to a higher orbital. The photon is absorbed (it dies. OH NOES), but the 'excited' electron cannot stay on it's excited orbital, so it shifts back to it's original orbital by losing the energy gained from the former photon. That energy's got to go somewhere, thus the electron "reemits" a photon (how photons are borned).
Why doesn't a cone light just flow into random directions? The electrons move around in all sorts of random orbits...
heywood on 3/10/2007 at 13:23
Quote Posted by Mingan
I'm most probably oversimplifying, but when an atom absorbs a photon (usually by an electron) it's because that photon was at a precise energy level (thus wavelength) that allow the electron to shift to a higher orbital. The photon is absorbed (it dies. OH NOES), but the 'excited' electron cannot stay on it's excited orbital, so it shifts back to it's original orbital by losing the energy gained from the former photon. That energy's got to go somewhere, thus the electron "reemits" a photon (how photons are borned).
You've described spectral absorption and emission well. But the electrical potential energy of an orbiting electron is only one possible source of energy for photon emission. There are other sources of energy for photon emission that are not quantized. For example, when two moving atoms collide, some of their kinetic energy can be released in the form of a photon. And thermal energy, which is kinetic energy stored in the vibration of atoms, is the source of energy for broadband photon emission.
Quote Posted by Yakoob
Why doesn't a cone light just flow into random directions? The electrons move around in all sorts of random orbits...
Photon emission is omni-directional. That's why when light passes through a gas, we observe absorption lines in the spectrum. The atoms in the gas absorb and re-emit some of the photons at specific wavelengths, and the emitted photons go in every direction and most don't end up at the observer.
When you observe light passing through a transparent material like glass, the fact that almost all of the light passes through means that almost none of it is being absorbed and re-emitted by the atoms in the glass.
Pyrian on 4/10/2007 at 00:44
Quick, small example on coin flips:
If you flip two coins, you can get 2H0T, 1H1T, 0H2T. You might call that three possibilities, and wonder why the middle one is twice as likely to occur as either of the other two. But it's really four possibilities, as follows: HH, HT, TH, TT. Those possibilities are all equally probable. We just normally add HT and TH together.
This effect scales to any degree. So, if you flip ten coins, each unique possibility has a 1/1024 chance of happening. I.e., HHHHHHHHHH, TTTTTTTTTT, HHTTHTHTTH, and THTTTHHTHH are all equally improbable - we just fail to distinguish HHTTHTHTTH and THTTTHHTHH and add them together. I think it's that addition which is really throwing you off.
So, an exact given set of flips is just as probable as any other set. The different probabilities in the sums just occur because there are many sets of flips that result in 5H5T and only one that results in 10H0T.
RocketMan on 4/10/2007 at 16:00
You're basically saying all "permutations" have equal probability but all "combinations" do not, right?
I just realized something, rather I understand what everybody is trying to say now. If I flipped 50 heads in a row, vanishingly small probability of that happening, but if I flipped ANY other sequence of 50 trials that I knew ahead of time, the outcome would have the same probability because P(H) = P(T) so whether its 50 heads or 1 head, 1 tail, and 48 heads, its still P(H) x P(T) x P(H)^48 = P(H)^50. My apologies for being so stubborn about this as I didn't see it until now.
As for the entropy thing, I still believe it comes in somewhere. It also has an effect on things like the air, microstructure defects, all sorts of things so until real life resembles an "ideal" coin toss with "ideal" coins and "ideal" environment, entropy can still affect the outcome of a coin toss I believe.
Martek on 4/10/2007 at 20:46
Quote Posted by Pyrian
Quick, small example on coin flips:
If you flip two coins, you can get 2H0T, 1H1T, 0H2T.
...
Basically exactly what I was saying on page two of this thread. :)
Pyrian on 4/10/2007 at 22:43
Quote Posted by Martek
Basically exactly what I was saying on page two of this thread. :)
Yes. But my post reads better. ;)
Quote:
As for the entropy thing, I still believe it comes in somewhere.
Entropy comes in everywhere. :p
I have a half-formed hypothesis that randomness (in the results of quantum waveform collapse) is fundamental to entropy and further to the very essence of the arrow of time. It's kind of the opposite causality of what you were trying to argue; rather than entropy causing randomness, randomness causing entropy.
Most laws of physics work the same way forward and backwards in time. Given an asteroids' position and velocity, you can work out where it's been and where it's going with equal validity. Contrast that with a photon fired into a idealized lattice of atoms, which is absorbed and then re-emitted by each atom it contacts. Once it's been bounced around a bit, you can theoretically figure out where it's been (which atoms have been moved by it - each atom will gain momentum equal to the difference between the direction the photon arrived and the direction the photon was emitted in), but cannot even in theory work out where it will go next. That, right there, is your arrow of time and the fundamental reason why we remember the past and not the future.
Chade on 4/10/2007 at 23:23
Here is what my Thermodynamics textbook has to say on the matter:
"not does it [thermodynamics] introduce a new fundamental law analogous to Netwon's or Maxwell's equations. ... the hallmark of thermodynamics is generality. ... thermodynamics applies to all systems in macroscopic aggregation ...
... thermodynamics is the study of the restrictions on the possible properties of matter that follow from the symmetry properties of the fundamental laws of physics.
... Thermodynamics inherits its universality, its nonmetric nature, and its emphasis on relationships from its symmetry parentage."
The implication, I believe, is that entropy occurs in any system whose fundamental laws have certain symmetrical properties.
Unfortunately, I lost interest in my physics degree by the time I got around to studying thermodynamics (I quit shortly afterwards), which is something I've always regretted. So I don't know enough about the subject to say anything with confidence.
RocketMan on 4/10/2007 at 23:24
Pyrian:
Agreed, however I have read that even entropy is bi-directional. The fact that time goes forward means that entropy always increases in the future, but if time were to flow backward (in theory) entropy would STILL increase. The laws of entropy dictate that entropy will increase with any transient, regardless of the direction of that transient. I don't remember the reason for that though.
Martek on 5/10/2007 at 01:24
Quote Posted by Pyrian
Yes. But my post reads better. ;)
:p
