Richard Feynman said about QED that students do not understand this theory, and that is because the professors do not understand it either!
NATURE Venus sometimes is a morning star, and sometimes - an evening star!
INVENTION Feynman: Mayans had hundreds of thousands of books. Of those only 3 were left. This was due to Spanish conquistadores and especially priests who burned them very efficiently.
Light is corpuscular. People thought it was a wave for some time, but in the end it turned out it is corpuscular!
Human eye needs 5 or 6 photons for a nerve to activate. Light is a particle every time we can detect it! It is NOT a wave! When we use two photon detectors, and throw single photons at them, either the first detector fires, or the second. The light particle is either here or there.
Partial reflection is an interesting phenomenon. Assume we have a surface of water, and light shining on that water. When we shine the light from above, about 4% of photons will be reflected. This means that given 21 billion photons, about 1 billion will be reflected.
What decides, which photons will be reflected exactly? We do not see, so the outcome is probabilistic for us. What are the options?
- The whole fenomenon can be entirely probabilistic, with chance ruling the whole universe from the ground up;
- There can be spots in the water - spots that must be hit by a photon for it to be bouced back.
- Electrons coming onto water may be "spinning", like football balls. Depending on how fast and in which direction they are spinning when hitting the surface, they would be bouncing off in different directions. In other words - something inside the photon would determine, which way it goes.
The first explanation is very unfavourable. And the second one is wrong. The invalidity of the second one has been postulated by Newton. Newton said that the sole fact that we can polish glass disproves the second way of understanding of this fenomenon. He said that from some point on smoothing the glass further does not influence how well light can penetrate that piece of glass.
We can now measure things about photons and atoms with the accuracy of $10^-15$ cm. Assuming the second interpretation, with this accuracy we should expect to be abe to find a point where we could focus a beam of photons and find that the reflection coefficient is greater in that particular spot.
It turns out no matter how well we polish a piece of glass, still exactly one photon in 25 will be reflected. The same occurs for various positions of the beam - the same amount is reflected.
The third understanding is also wrong. If it were true, it would mean all light that went through is a certain kind of football, and next time it was reflected you could expect a number other than 4% being reflected.
As we understand it today, the way the reflection experiment works can be understood only by chance.
This is the key, and correct, observation about light - it works by probability.
Do you remember soap bubbles and their colors? Soap itself is a colorless mixture. Take it and see! Where do these colors come from? Recall oil spills. Oil is dark grey. Still, when spilled, you can observe various colors on its surface. What is the cause of that?
It turns out that if we add another layer of glass below the first one, and shine monochromatic light onto it, the amount of light reflected off this two-layer surface will vary depending on the distance between these two layers. At some distances between the layers, the amount of reflected photons will be zero, and sometimes it will be twice as many as in the case of one layer.
In the case of the reflection fenomenon, actually the surface does not reflect the photons - it is the electrons in the material that do. Otherwise, the internal material would not make a difference - the reflection would be the same each time; but we see differences between materials - often, significant differences.
We can predict the outcomes for experiments involving electrons, and photons, by drawing arrows and calculating the probability by squaring the amplitude of the phenomenon (the length of the resulting arrow). The light arrow goes around 10^15 times every second.
Source: QED lecture #1