Based on the framework developed in the preceding sections, this section examines how the photoelectric and tunneling effects arise. The double-slit experiment and the delayed-choice quantum eraser experiment, both conventionally cited as evidence that photons and electrons possess wave-like and particle-like properties, are also analyzed.
4.3. Double-Slit Experiment
The double-slit experiment is among the most widely cited demonstrations that photons and electrons exhibit both wave-like and particle-like properties.[
20] When an electron passes through two slits, it interferes as a wave, producing an interference pattern on the screen behind (
Figure 5(a)). Even when electrons are sent through one at a time, an interference pattern gradually builds up on the screen, indicating that a single electron behaves as a wave. When one observes which slit the electron traversed, however, the interference pattern vanishes and the electron appears to have passed through the slits as a particle (
Figure 5(b)).
If electrons are treated as waves interacting at a single spatial point, the double-slit experiment admits the following explanation. Because the electron is a wave, even a single electron passes through both slits simultaneously and undergoes interference. The resulting wave remains spatially extended until it reaches the screen and interacts with the atoms composing it. At the screen, the wave interacts electromagnetically with atomic electrons at a single point, with a probability proportional to the squared absolute value of the wave function (|ψ|
2). An interference pattern therefore emerges on the screen. Observing which slit the electron traversed, on the other hand, necessarily involves interacting with the electron, for example by directing a photon at it. As illustrated in
Figure 5(b), when a photon strikes the wave packet that has passed through the left slit, the electron materializes at that location and the two split wave packets vanish instantaneously. The electron, having materialized through observation, propagates from that position as a new, single wave packet. Because only one wave packet now exists, no interference occurs.
The wave function formalism describes this situation as follows. After passing through both slits, the electron’s wave function is given by equation (4).
where
and
denote the wave functions corresponding to passage through the right and left slits, respectively. The probability distribution of the electron after passing through both slits is given by equation (5).
The third and fourth terms on the right-hand side are interference terms. If the electron materializes at the left slit as a result of observation, the wave function is given by equation (6).
and the post observation probability distribution is given by equation (7).
No interference terms appear in this expression.
Under the standard interpretation, particle-like behavior is taken to emerge when the electron’s path is observed. What actually occurs, however, is the disappearance of the interference terms from the probability distribution as a consequence of the observation; no genuine “particle-like” nature manifests in this process.
4.4. Delayed Choice Quantum Eraser Experiment
The delayed-choice quantum eraser experiment is a striking experiment that appears to permit a retroactive choice of whether a photon behaves as a wave or as a particle. A detailed description of the experimental setup is provided in Reference.[
21]
The experiment can be summarized as follows. Positioned behind the two slits is a nonlinear optical crystal (BBO), which emits a pair of entangled photons from each of the two paths, A and B (
Figure 6). One member of each entangled pair, termed the signal photon, is registered by detector D
0, while the other, the idler photon, is registered by one of the detectors D
1, D
2, D
3, or D
4 (detector D
4 is omitted in
Figure 6). Detector D
0 is arranged to scan along the x-direction so as to record any interference pattern. The idler photon from path A is divided by beam splitter BSA, with 50% directed toward detector D
4 and the remaining 50% toward detectors D
1 and D
2. The idler photon from path B is likewise divided by a beam splitter, with 50% directed toward detector D
3 and the remaining 50% toward detectors D
1 and D
2. Detectors D
1 and D
2 are arranged so that beam splitter BS causes the two idler photons from paths A and B to interfere; D
1 registers only in-phase photons, whereas D
2 registers only out-of-phase photons. Because detectors D
1 and D
2 can receive photons from both paths, a detection event at either D
1 or D
2 does not reveal whether the photon originated from path A or path B. Detectors D
3 and D
4, by contrast, receive photons exclusively from paths B and A, respectively, so a detection event at D
3 or D
4 does identify the photon’s origin. The optical path lengths are further adjusted so that detector D
0 registers photons 8 ns before the monitoring detectors D
1, D
2, D
3, and D
4. By the time an idler photon is detected, the corresponding signal photon has therefore already been registered. A coincidence circuit connecting the interference and monitoring detectors enables the identification of correlated signal–idler pairs.
The experimental results can be summarized as follows.
Figure 7 shows the joint detection rate R
01 and R
02 against the x coordinates of detector D
0. In both cases, a clear Young’s double-slit interference pattern is observed. There is a phase difference of π between the two interference patterns of R
01 and R
02.
Figure 8 shows the joint detection rate R
03 between D
0 and D
3; here, no interference pattern is discernible. These results are conventionally interpreted as follows. The interference patterns in R
01 and R
02 arise because detectors D
1 and D
2 cannot determine whether the idler photons traveled along path A or path B. The absence of an interference pattern in R
03, by contrast, is attributed to the fact that detector D
3 can identify the path taken by the idler photons. In this view, one can select whether an interference pattern appears, and hence whether the photon behaves as a wave or as a particle, depending on whether the photon’s path is determined. It is further interpreted that one can retroactively choose the quantum’s wave-like or particle-like behavior by selecting the photon’s path even after the detection process is complete.
If photons are regarded as waves that interact at a single spatial point, the results of this experiment admit a different interpretation. Two entangled photons are emitted from paths A and B of the BBO. Although the signal and idler photons are emitted simultaneously, the optical path lengths are adjusted so that the signal photon reaches detector D0 8 ns before the idler photon reaches detectors D1, D2, D3, or D4. At the moment a signal photon is detected by D0, the signal photon’s wave packet collapses; the idler photon, however, has not yet arrived at its detector and remains in its wave packet state. The idler wave packet from path A is divided into two components by beam splitter BSA: one directed toward detector D3 and the other toward mirror MA. The component reflected by mirror MA is further divided by beam splitter BS, with one portion reflected toward detector D1 and the other transmitted toward D2. The component of wave packet A directed toward D1 combines with the component of wave packet B transmitted through beam splitter BS. Meanwhile, the component of wave packet A directed toward D2 combines with the component of wave packet B reflected by beam splitter BS. The system is configured so that wave packets A and B arriving at detector D1 reinforce each other when in phase and cancel when out of phase; for detector D2, the converse holds, with reinforcement occurring for out-of-phase wave packets and cancellation for in-phase ones.
The phase relationships of the photons detected at D0, D1, D2, and D3 are summarized below for two cases:
1) Case where signal photons A and B are in phase:
The wave function of the wave packets entering detector D
0 is a superposition of in-phase signal photons:
where
and
denote the wave packets of signal photons originating from paths A and B, respectively, and
θ denotes the photon phase.
Because the idler photons are entangled with the signal photons, they are in phase with the signal photons immediately after emission from the BBO crystal. The idler photon from path A entering detector D
1 undergoes two reflections, so its phase is inverted twice and returns to its original value. The idler photon from path B entering D
1 undergoes a single reflection, resulting in a phase inversion. The wave function of the wave packets entering detector D
1 is therefore
where
and
are the wave packets of the idler photons originating from paths A and B, respectively.
Similarly, the idler photon from path A entering detector D
2 undergoes a single reflection at mirror M
A, acquiring a phase inversion. The idler photon from path B undergoes a total of two reflections, at mirror M
B and beam splitter BS, and its phase returns to its original value. The wave function of the wave packets entering detector D
2 is:
As Equations (9) and (10) indicate, the waves entering detectors D1 and D2 consist of out-of-phase combinations. When signal photons A and B are in phase, they are consequently detected only at detector D2. Although individual signal photons are detected randomly at single spatial points by detector D0, the cumulative detection results exhibit the interference pattern given by Equation (8).
2) Case where signal photons A and B are out of phase:
The wave function of the wave packets entering detector D
0 is a superposition of out-of-phase photons:
Following the same reasoning as before, the idler photon from path A entering detector D
1 undergoes two phase inversions and thereby returns to its original phase. The idler photon from path B, which begins in an out-of-phase state, undergoes a single inversion and consequently becomes in phase with the original reference. The wave function of the wave packets entering detector D
1 is
For detector D
2, the idler photon from path A undergoes a single reflection, acquiring a phase inversion. The idler photon from path B undergoes two reflections and returns to its original out-of-phase state. The wave function of the wave packets entering detector D
2 is
As indicated in Equations (12) and (13), the waves entering detectors D1 and D2 consist of in-phase combinations. When signal photons A and B are out of phase, they are therefore detected only at detector D1. The cumulative detection results at D0 exhibit the interference pattern given by Equation (11).
From these results, it is clear that the coincidence rate R01 between detectors D0 and D1 displays the interference pattern of Equation (11), while the coincidence rate R02 between D0 and D2 displays the interference pattern of Equation (8), shifted in phase by π. Because detector D3 receives photons exclusively from path B, however, the phase relationship relative to photons from path A remains undetermined. The coincidence rate R03 between detectors D0 and D3 therefore consists of a superposition of both the in-phase and out-of-phase interference patterns, which are mutually indistinguishable. The absence of a visible interference pattern in R03 does not arise because the photon’s path was identified; it arises because two opposing interference patterns are superimposed and cannot be individually resolved. Nor does this result imply that one can retroactively select whether the quantum behaves as a wave or as a particle; rather, interference patterns are inherently present in all detection results of the signal photons.