Professor Feynman (Lect. on Phys., v. 3, 1965, ps. 3-7 to 3-9) maintained in his neutron scattering off a crystal experiment that which-way information can exist even if one does not perform a measurement. This interaction involves the possibility of a spin flip for both the neutron and nucleus that the neutron scatters off. With the flip, the spin of the nucleus that the neutron scattered off becomes different than the spin direction of the other nuclei in the crystal that the neutron could have scattered off. The spins of the other nuclei are the same. It may be possible to eliminate the which-way information as long as particle detections have not been made, either through spin-lattice relaxation after the neutron-nucleus interaction occurs or through shutting off the uniform, strong, external magnetic field B that initially aligns all of the spins of the crystal nuclei along the same axis. A delayed choice allows for obtaining either a probability distribution showing interference or instead one showing which-way information.
Which-Way Information vs No Which-Way Information (Interference)
editProfessor Feynman summarized the difference between which-way information and the absence of which-way information (interference)in a distribution of particles in what he called "First principles of quantum mechanics". Vol. 3, Sect.1-7,The Feynman Lectures on Physics - Vol. 3 - Quantum Mechanics. These principles are:
1) The probability of an event in an ideal experiment is given by the square of the absolute value of a complex number ϕ which is called the probability amplitude.
2) When an event can occur in several alternative ways, the probability amplitude for the event is the sum of the probability amplitudes for each way considered separately. There is interference.
3) If an experiment is performed which is capable of determining whether one or another alternative is actually taken, the probability of the event is the sum of the probabilities for each alternative. The interference is lost.
Given the same experimental setup, the probabilities of the same events in a number of experimental runs is determined by 1). If these probabilities are determined subject to 2) there is interference in the distribution. If these probabilities are determined subject to 3) there is no interference and instead there is which-way information in the distribution.
 |
 |
Experiments
editProfessor Feynman in vol. 3 of his lectures (quantum mechanics) discussed the following experiments, the overall structure of which follows: Neutrons are sent one at a time against a crystal. The concern is with the nuclei of the atoms comprising the crystal that are lined up basically in a periodic array. The neutrons scatter off the nuclei in the crystal into a detector (Fig. 1). The nuclei in the crystal either do not have spin or instead spin 1/2. The neutrons have spin 1/2. Professor Feynman wrote: “We have here a large number of apparently indistinguishable routes. They are indistinguishable because a low-energy neutron is scattered from a nucleus without knocking the atom out of its place in the crystal—no ‘record’ is left of the scattering [except if there is spin flip in the nucleus the neutron scattered off].”
Experiment 1 - All of the Nuclei Don't Have Spin
editIf all of the nuclei of the crystal do not have spin, the distribution of the neutrons at the detector exhibit interference since it is not known which atomic nucleus the neutron scattered off. The distribution of the neutrons is given by Eqn. 1:
- [1]
![{\displaystyle P=|\psi _{total}|^{2}=|[{\frac {1}{\sqrt[{2}]{a}}}[\psi _{neutron_{1}}\psi _{crystal_{1}}+\psi _{neutron_{2}}\psi _{crystal_{2}}+...+\psi _{neutron_{a}}\psi _{crystal_{a}}]]|^{2}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/11c63d859662adb9e4d2ba21ea9436bc6f548a30)
The distribution of the neutrons is depicted in Fig. 2. Table 1 indicates the initial states of the neutrons and nuclei in Experiment 1, and Table 2 indicates the final states of the neutrons and nuclei in Experiment 1.
| Neutrons | Nuclei |
|---|---|
| Spin 1/2 | No Spin |
| Neutrons | Nuclei |
|---|---|
| Spin 1/2 | No Spin |
Experiment 2 - Nuclei Have Spin and Their Spin Direction is Same as the Neutron
editIf the nuclei have spin one half and if the spin of both the neutron and the nuclei are in the same direction, interference is exhibited in the distribution of the neutrons since there is no change in spin of the neutron or the nucleus that it impacts. The nucleus that the neutron impacted cannot be distinguished from the other nuclei in the crystal that the nuclei could have impacted, but did not, impact. The distribution of the neutrons is given by:
- [1]
The distribution of the neutrons is depicted in Fig. 2. Table 3 indicates the initial states of the neutrons and nuclei in Experiment 2, and Table 4 indicates the final states of the neutrons and nuclei in Experiment 2.
| Neutrons | Nuclei |
|---|---|
| Spin ½ of all neutrons in same direction |
Spin ½ in same direction as neutrons |
| Neutrons | Nuclei |
|---|---|
| Spin ½ of all neutrons in same direction |
Spin ½ in same direction as neutrons |
Experiment 3 - Nuclei Have Spin and Their Spin Direction in Opposite Direction as the Neutron
editIf the spin of the neutron and the spins of the nuclei are initially in opposite directions, one up and the other down, two outcomes are possible. The distribution of neutrons in this Experiment, as in all of the Experiments, is developed over a number of runs. In Experiment 3 we do not know whether or not we will have a spin flip in the neutron and the nucleus it impacts.
Possibility 1 - Spin Flip in Interaction between Neutron and Specific Nucleus
editIf there is spin flip when the neutron interacts with one of the nuclei, we obtain a which-way pattern since it can be known which nucleus the neutron scattered off (possibility 1). In possibility 1 there is which way information since we have spin flip. The neutron can scatter off any one of the nuclei in the crystal. The probability distribution of the neutrons is depicted in Fig. 3. Professor Feynman wrote: “Well, if we can tell which atom did the scattering, what have the other atoms got to do with it? Nothing, of course. The scattering is exactly the same as that from a single atom.” The probability distribution of the neutrons is given by:
- [2]
![{\displaystyle P_{possibility1\psi }=[1/a[|(\psi _{neutron_{1}}|crystal_{1}\rangle )|^{2}+|(\psi _{neutron_{2}}|crystal_{2}\rangle )|^{2}+...+|(\psi _{neutron_{a}}|crystal_{a}\rangle )|^{2}]}](https://wikimedia.org/api/rest_v1/media/math/render/svg/9d2fd8bdc6d75049801a91294261c9347117d686)
where
and
The probability distribution of the neutrons is depicted in Fig. 3. Table 5 indicates the initial states of the neutrons and nuclei in Experiment 3 – possibility 1 or possibility 2, and Table 6 indicates the final states of the neutrons and nuclei in a particular run in this scenario for possibility 1. The probability distribution of neutrons in this Experiment, as in all of the Experiments, is developed over a number of runs. Now in Experiment 3 we do not know whether or not we will have a spin flip in the neutron and the nucleus it impacts. Figure 3 and the Table 6 are concerned with those runs in Experiment 3 where we have a spin flip (possibility 1).
| Neutrons | Nuclei |
|---|---|
| Spin ½ of all neutrons in same direction |
Spin ½ of all the nuclei in opposite direction to neutrons |
| Neutrons | Nuclei |
|---|---|
| Spin ½ of the neutron in a run of the experiment flips to the opposite direction as result of interaction with the crystal. |
In a run of the experiment, spin ½ of the nucleus that impacted by the incoming neutron flips to opposite direction. Thus there is which-way information as to the specific path of the neutron since the other nuclei do not flip. |
In possibility 1, Professor Feynman concludes we don’t have to observe “which atom [which atomic nucleus] is up" (that is flipped) to get a which-way distribution from these neutrons (where there is spin flip) over a series of runs. The information is in the nuclei and that is enough to provide which-way information to the neutron. This which-way information is found in the distribution of the neutrons over a series of runs where they are detected in the case where there is spin flip in the interaction between a neutron and the nucleus it impacts. Professor Feynman wrote: “You may argue, ‘I don’t care which atom is up.’ Perhaps you don’t, but nature knows; and the probability is, in fact, what we gave above—there isn’t any interference.” Section 3–3 Scattering from a crystal.
The implication appears to be that once this interaction that produces the which-way information occurs, this which-way information appears irrevocable since it does not matter whether the spin of the nucleus that the neutron scattered off is measured. Contrary to this implication, it may be possible to eliminate the which-way information as long as particle detections have not been made. Eliminating the which-way information before any particle detections are made is found in Experiment 4 below.
Possibility 2 - No Spin Flip in Interaction between Neutron and Specific Nucleus
editIf there is no spin flip when the neutron interacts with one of the nuclei, we obtain the interference pattern since we still cannot tell which nucleus the neutron scattered off (possibility 2). Since we do not have spin flip, the interference pattern is still exhibited since we cannot tell which nucleus the neutron scattered off. The relevant eqn. is:
- [1]
The probability distribution of the neutrons is depicted in Fig. 2. Table 5 indicates the initial states of the neutrons and nuclei in Experiment 3 – possibility 1 or possibility 2, and Table 7 indicates the final states of the neutrons and nuclei in a particular run in this scenario for possibility 2. Figure 2 depicts data for a series of runs in Experiment 3 where we do not have a spin flip (possibility 2).
| Neutrons | Nuclei |
|---|---|
| Spin ½ of the neutron in a run of the experiment does not flip to the opposite direction as result of interaction with the crystal |
In a run of the experiment, spin ½ of the nucleus that impacted by the incoming neutron does not flip to opposite direction as all the other nuclei. Thus there is no which-way information as to the specific path of the neutron. Instead, there is interference. |
The Difference Between Possibilities 1 and 2
editProfessor Feynman wrote that the difference between possibilities 1 and 2 is the following. In possibility, 1 the phases disappear since we first take the absolute square of each possible neutron-nucleus interaction and then sum the probabilities for each possible interaction with the result that the phases disappear when the absolute squares are calculated. In possibility 2, we first sum the amplitudes for each possible interaction, which now means summing the phases for each amplitude, and then take the absolute square of the sum yielding interference since the phase for each amplitude is preserved.
Spin Flip and No Spin Flip in Interaction between Neutrons and Specific Nuclei
editIn Experiment 3, since we don’t know whether or not there is going to be spin flip (since the neutron and the atomic nuclei have opposite spins to begin with), the resulting overall probability distribution from combining the results for possibilities 1 and 2 over a number of runs with neutrons shows a which-way pattern punctuated with interference (Fig. 4).
- [3]
the sum of Eqns. 1 and 2:
![{\displaystyle P_{total\psi }=[1/a[|(\psi _{neutron_{1}}|crystal_{1}\rangle )|^{2}+|(\psi _{neutron_{2}}|crystal_{2}\rangle )|^{2}+...+|(\psi _{neutron_{a}}|crystal_{a}\rangle )|^{2}]+}](https://wikimedia.org/api/rest_v1/media/math/render/svg/7600024683bf9a3c82b68820706ab91baf1f9ecf)
- [4]
![{\displaystyle |[{\frac {1}{\sqrt[{2}]{a}}}[\psi _{neutron_{1}}\psi _{crystal_{1}}+\psi _{neutron_{2}}\psi _{crystal_{2}}+...+\psi _{neutron_{a}}\psi _{crystal_{a}}]]|^{2}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/76009b4f8c504f44a5464acb6086d3d22b917f3e)
 |
Experiment 4
editThrough the use of a delayed choice, it may be possible to eliminate the which-way information as long as particle detections have not been made. A delayed choice occurs where the choice of whether an experiment will yield which-way information or instead no which-way information (interference) in a series of experimental runs is made after a particle is already in route to a detection device (i.e., after the experiment has begun). Perhaps through spin-lattice relaxation or a similar process after the neutron-nucleus interaction occurs, a spin flip, if it occurred in a nucleus as a result of its interaction with the incoming neutron, would reverse before any detection is made. It would no longer be possible to determine which nucleus in the crystal the neutron scattered off (possibility 1). The result is interference in the distribution of the neutrons where there would have been which-way information exhibited if the spin flip was not reversed before any detections are made. The distribution of the neutrons would be affected by a change in information regarding the nuclei in the crystal since there is no physical process whereby the change in the nuclei as a result of the delayed choice can affect the distribution of the neutrons. The phonon created in the relaxation of the spin of the impacted nucleus would be for all practical purposes not possible to detect with the thermal activity in the crystal. (G. Jeschke, Relaxation and slow motion in EPR spectroscopy)
If spin-lattice relaxation occurs before detections are made, even if we have a spin flip in the interaction between a neutron and a nucleus, all the nuclei in the crystal that can interact with the neutron will be found with their spin in the same direction (possibility 1). Which-way information held by the nucleus off of which the neutron is initially to have thought to have scattered is lost, since it can no longer be distinguished from the other nuclei in the crystal. The motion of the neutron after impacting the crystal itself is certainly capable of producing interference since when no spin flip occurs, interference in the distribution of the neutrons in a series of experimental runs occurs. The equation for the situation after spin lattice relaxation and all the nuclei again have their spin in the same original direction is Eqn 1. All the nuclei that the neutron could have scattered off now have the same orientation before any detections are made. The crystal possesses no which-way information before detections are made and the distribution of the neutrons in possibility 1 should show interference.
Possibility 2 is making a neutron detection before spin-lattice relaxation occurs for the nucleus that scattered the neutron so that there is which-way information when the neutron is detected and the distribution of these neutrons where there is spin flip in interaction with specific nuclei in the crystal is the wide round hump characteristic of which way information. Possibility 2 could occur through lengthening the spin relaxation time for the impacted nucleus over the spin relaxation time possibility 1.
There exists a delayed choice for possibilities 1 or 2 whether or not to make a detection of the neutron before spin-lattice relaxation occurs. The two possible distributions of neutrons over a series of runs either exhibit only interference (Figure 2) OR instead a combination of interference and which-way information (Figure 4).
| |
|
Possibility 1
editThe equation is:
- [1]
![{\displaystyle P_{total\psi }=|\psi _{total}|^{2}=|[{\frac {1}{\sqrt[{2}]{a}}}[\psi _{neutron_{1}}\psi _{crystal_{1}}+\psi _{neutron_{2}}\psi _{crystal_{2}}+...+\psi _{neutron_{a}}\psi _{crystal_{a}}]]|^{2}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/2227580a5e66f12b8c847a76b776f40f5c6db9a5)
The relevant figure for the neutron distribution is Figure 2.
Possibility 2
editThe relevant equation is Eqn. 4 for Experiment 3,
- [3]
where
- [2]
![{\displaystyle P_{\psi \;possibility1\;in\;Expt.3}=[1/a[|(\psi _{neutron_{1}}|crystal_{1}\rangle )|^{2}+|(\psi _{neutron_{2}}|crystal_{2}\rangle )|^{2}+...+|(\psi _{neutron_{a}}|crystal_{a}\rangle )|^{2}]}](https://wikimedia.org/api/rest_v1/media/math/render/svg/7414a85b8e76e7034ea0c5ab02a684036986cc27)
and
- [1]
![{\displaystyle P_{\psi \;possibility2\;in\;Expt.3}=|[{\frac {1}{\sqrt[{2}]{a}}}[\psi _{neutron_{1}}\psi _{crystal_{1}}+\psi _{neutron_{2}}\psi _{crystal_{2}}+...+\psi _{neutron_{a}}\psi _{crystal_{a}}]]|^{2}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/acbdaf58b1c45edddb0bc21f98a44e5ae009bdf5)
The relevant figure for the neutron distribution is Figure 4.
Experiment 5
editAnother experiment would be to shut off the uniform, strong external magnetic field that initially lines up all of the spins of the nuclei in the same direction after the spin flip and before the neutron is detected. An ultrafast switch would be needed in this circumstance that would shut off the external magnetic field after the neutron was scattered off the crystal but before the neutron was detected (and leaving enough time for the spins of the nuclei to randomize before the neutron is detected). Which-way information would be eliminated since the spin directions of all the nuclei would quickly become essentially random. This is very similar to possibility 1 in Experiment 4where which-way information is lost.
Possibility 1 is turning off the external magnetic field before a neutron detection is made and with enough time for the nuclei spins to randomize. In this case we obtain interference only as in Equation 1 and Figure 2. Figure 5 shows the experimental design. Possibility 2 is not altering the external magnetic field. In this case we obtain the mix of the which-way and interference distributions discussed in Experiment 4 in possibility 2: Equation 4 and Figure 4. Figure 6 shows the experimental design.
Neutralizing Any Effect of the Neutron Spin after It Interacts with the Crystal on the Neutron Intensity Distribution
editIt is useful to take into account of the possibility that the spin flip in the neutron, when it occurs, could be a factor. In the intensity distributions of the neutrons. To eliminate this possibility one could “mix up” the spin orientations along the axis of the B field in the following way: 1) extend the B field past the crystal almost to the neutron detector. Apply pulsed oscillating magnetic radiation (B’) to the neutrons in the area between the crystal and the neutron detector at a 90O angle to the axis of the external homogenous magnetic field B (where the intensity of the oscillating is much less than that of the external homogenous magnetic field B. This frequency of the oscillating magnetic field would be at the Larmor precession frequency of the neutrons. It can flip neutrons from a low energy state to a high energy state or from a high energy state to a low energy state. If one used an appropriate intensity for B’ where only a certain percentage of the neutron spins were flipped by the oscillating magnetic field (either higher to lower or lower to higher in the same proportion) one could eliminate the neutron spin state itself as a factor in determining whether there is or is not which-way information in the neutron intensity distribution as determined at the neutron detector.