Einstein was awarded the Nobel Prize in 1921 not for his theory of relativity, as many suppose, but for his explanation of the photoelectric effect. Although Planck had made the first suggestion, it was Einstein who showed convincingly that energy is quantized. Quanta of light energy were later given the name photons.

But just what are photons? To begin our explanation, let’s return to the experiment that showed most dramatically the wave nature of light—Young’s double-slit interference experiment. We will make a change, though: We will dramatically lower the light intensity by inserting filters between the light source and the slits. The fringes will be too dim to see by eye, so we will replace the viewing screen with a detector that can build up an image over time.

What would we predict for the outcome of this experiment? If light is a wave, there is no reason to think that the nature of the interference fringes will change. The detector should continue to show alternating light and dark bands.

FIGURE 38.10 shows the actual outcome at four different times. At early times, contrary to our prediction, the detector shows not dim interference fringes but discrete, bright dots. If we didn’t know that light is a wave, we would interpret the dots as evidence that light is a stream of some type of particle-like objects. They arrive one by one, seemingly randomly, and each is localized at a specific point on the detector. (Waves, you will recall, are not localized at a specific point in space.)

FIGURE 38.10 A double-slit experiment with light of very low intensity.

As the detector builds up the image for a longer period of time, we see that these dots are not entirely random. They are grouped into bands at exactly the positions where we expected to see bright constructive-interference fringes. No dot ever appears at points of destructive interference. After a long time, the individual dots overlap and the image looks like the photographs of interference fringes in Chapter 33.

We’re detecting individual photons! Most light sources—even very dim sources—emit such vast numbers of photons that you are aware of only their wave-like superposition, just as you notice only the roar of a heavy rain on your roof and not the individual raindrops. But at extremely low intensities the light begins to appear as a stream of individual photons, like the random patter of raindrops when it is barely sprinkling. Each dot on the detector in Figure 38.10 signifies a point where one particle-like photon delivered its energy and caused a measurable signal.

But photons are certainly not classical particles. Classical particles, such as Newton’s corpuscles of light, would travel in straight lines through the two slits of a double-slit experiment and make just two bright areas on the detector. Instead, as Figure 38.10 shows, the particle-like photons seem to be landing at places where a wave undergoes constructive interference, thus forming the bands of dots.

Today, it is quite feasible to do this experiment with a light intensity so low that only one photon at a time is passing through the double-slit apparatus. But if one photon at a time can build up a wave-like interference pattern, what is the photon interfering with? The only possible answer is that the photon is interfering with itself. Nothing else is present. But if each photon interferes with itself, rather than with other photons, then each photon, despite the fact that it is a particle-like object, must somehow go through both slits! Photons seem to be both wave-like and particle-like at the same time.

This all seems pretty crazy, but it’s the way light actually behaves. Sometimes light exhibits particle-like behavior and sometimes it exhibits wave-like behavior. The thing we call light is stranger and more complex than it first appeared. Furthermore, as we will see, this half-wave/half-particle behavior is not restricted to light.