Now let’s look at the interference of matter. Electrons passing through a double-slit apparatus create the same interference patterns as photons. The pattern is built up electron by electron, but there is no way to predict where any particular electron will be detected. Even so, we can establish the probability of an electron landing in a narrow strip of width by measuring the positions of many individual electrons.
Electrons create interference fringes.
For light, we were able to relate the photon probability density to the amplitude of an electromagnetic wave. But there is no wave for electrons like electromagnetic waves for light. So how do we find the probability density for electrons? We have reached the point where we must make an inspired leap beyond classical physics. Let us assume that there is some kind of continuous, wave-like function for matter that plays a role analogous to the electromagnetic amplitude function for light. We will call this function the wave function , where is a lowercase Greek psi. The wave function is a function of position, which is why we write it as .
To connect the wave function to the real world of experimental measurements, we will interpret in terms of the probability of detecting a particle at position . If a matter particle, such as an electron, is described by the wave function , then the probability of finding the particle within a narrow region of width at position is
That is, the probability density for finding the particle is
With Equations 39.14 and 39.15, we are defining the wave function to play the same role for material particles that the amplitude function does for photons. The only difference is that is for particles, whereas Equation 39.13 for photons is . The difference is that the electromagnetic field amplitude had previously been defined through the laws of electricity and magnetism. is proportional to the probability density for finding a photon, but it is not directly the probability density. In contrast, we do not have any preexisting definition for the wave function . Thus we are free to define so that is exactly the probability density. That is why we used rather than in Equation 39.15.
FIGURE 39.5 shows the double-slit experiment with electrons. This time we will work backward. From the observed distribution of electrons, which represents the probabilities of their landing in any particular location, we can deduce that has alternating maxima and zeros. The oscillatory wave function is the square root at each point of . Notice the very close analogy with the amplitude function in Figure 39.1.
NOTE is uniquely determined by the data, but the wave function is not unique. The alternative wave function —an upside-down version of the graph in Figure 39.5—would be equally acceptable.
FIGURE 39.6 is a different example of a wave function. After squaring it at each point, as shown in the bottom half of the figure, we see that this wave function represents a particle most likely to be detected very near or . These are the points where is a maximum. There is zero likelihood of finding the particle right in the center. The particle is more likely to be detected at some positions than at others, but we cannot predict what its exact location will be at any given time.
NOTE One of the difficulties in learning to use the concept of a wave function is coming to grips with the fact that there is no “thing” that is waving. There is no disturbance associated with a physical medium. The wave function is simply a wave-like function (i.e., it oscillates between positive and negative values) that can be used to make probabilistic predictions about atomic particles.