Experiment of Marlan Scully

The experiment of Marlan Scully is an experiment of quantum Mécanique which constitutes an extension of that of Alain Aspect and Fentes of Young while introducing there what seems to be a implicit feedback in time .

Schematically, two devices similar to the Fentes of Young are installed in cascade.

It is known that the quantum uncertainty concerning the passage (possible!) of particle S by one or the other slit

is levable only by one process of detection,
and remains in the absence of this one not only as a knowledge of the experimenter , but well as a state of the system .

The idea of Marlan Scully is to decide the intervention of this observant only at the last time, whereas the particle has already crossed the first series of slits.

The equations of the quantum Mécanique or not force the particle to have checked at the time of the first passage of the conditions which are however stipulated only subsequently , by later intervention of the detector. In other words, this intervention of the detector seems to modify the past of the particle.

The observation confirms for the moment this result envisaged, but Marlan Scully does not come to a conclusion for the moment about the lesson which one can or not draw. John Wheeler is shown less held and holds on this subject of the remarks currently discussed on the modification of passed by processes of observation (with less, according to another interpretation of the same phenomenon, that it is not about a definition of the present by the result of the observation of last phenomena - to see the Théorie of Everett).

Description of the experiment

The experiment is simpler than it does not appear to with it. The device will be described gradually, in order to reveal well the ideas behind each element of the experiment.

Let us describe initially the first part of the device: if we replace the apparatuses B and C by simple mirrors, we find ourselves with an alternative of the experiment of the Fentes of Young: the semi-reflective mirror has causes an interference “of the photon with itself” and causes a figure of interference out of I. It is important to include/understand the experiment of Young well before trying to include/understand this one.

In fact, out of B and C, “low converters are placed”. A “converter low” is an apparatus which, starting from a photon in entry, creates two photons at exit, correlated, and double wavelength compared to the photon in entry. Being correlated, any measurement taken on one of the two photons of exit informs us about the state of the other photon. By definition, one of the two photons at exit will be called “photon signal” and the other “pilot photon”. It is important also to stress that the “low converter” does not destroy the quantum state of the photon: there is not “measurement” and the state of the two photons at exit respects the state of superposition of the photon in entry.

Now, that there is no semi-reflective mirror in D and E. couldn't one imagines detect by which way (“by B” or “C”) passed the initially emitted photon? If the detector J starts, it that the photon passed by B, if it is K, it is that shouldn't the photon passed by C. the “photons signals” behaving same manner that if there were mirrors out of B or C, the figure of interference is appear, while informing us about the way taken by the photon? (it would be in contradiction with the experiment of Young)

In fact, not. The “measurement” taken by one of the detector J or K destroys the quantum state of the photons “signal” and “witness” (those quantiquement being quantiquement correlated, to see Paradoxe EPR), and no figure of interference does not appear out of I. We find well the results of the experiment of Young.

Now, let us consider the complete device, represented by the figure. The pilot photon has a chance on two to be considered by the mirror D or E. In this case it arrives out of F and there is not then more average to know if the photon passed by B or C. Indeed, that the photon comes from E or D, it has in both cases a chance on two to be detected out of H or in G. Thus detection out of H or G does not make it possible to know from which the photon comes. This mirror F is the “quantum Gomme” imagined by Sculley: it destroys information making it possible to know by which way the photon passed.

However, so instead of to be reflected by D or E, the pilot photon was detected by J or K, then it is possible to know the way borrowed by the photon, and the photon signal corresponding recorded out of I does not contribute to make a figure of interference. The mirrors D and E " draw with the sort" - to some extent - the destiny of the pilot photon: a chance on two to become a photon which one knows the way, a chance on two to become a photon whose way is unspecified.

However, distance data base (and a fortiori BF) can be much higher than the distance BI, and of the same for respectively CE/CF and Ci. And it is the case in this experiment. Therefore, when the photon signal comes to impress the photographic plate out of I, the pilot photon did not reach D or E yet, and even less F. It is the " retardé" choice; it is question in the experiment. the result recorded out of I is thus fixed avant that the pilot photon was detected in J/K, or G/H .

At the time when the photon signal impresses I, the way of the pilot photon is still unspecified. The figure out of I should thus be organized systematically in figure of interference. However, a pilot photon on two on average will be detected in J/K, and the photons signals corresponding should not be organized in figure of interference (since one knows the borrowed way). How the photon signal " it knows " what the pilot photon will be detected in J/K or not? Such is the fundamental question of this experiment.

In experiments it is noted that it there forever of error: the photons signals whose pilot photons are detected in J/K do not organize in figure of interference, the photons signals whose pilot photons are detected in G/H are organized in figure of interference (see following section).

Figures of interferences

When the impacts of photons are materialized out of I, one sees only one fog without significance. It is the correlation between each impact and the sensor which makes it possible to reveal the interference rings. Here, the photons are coloured according to the sensor which received them. When the image into four is separated, one clearly sees appearing the figures of interference for the photons detected in G and H.

To note that two figures 3. and 4. are complementary, in opposition of phase. The addition of these two figures gives a figure similar to figure 2. or 1. That explains the need for receiving correlations to reveal the figures of interference. Without correlations, it is possible no to distinguish case 3.+4 from case 5.+6.

Random links: