Discusses Mach-Zehnder interferometers in terms of a prototype unified model for light.  Part of a series arguing that a unified model for light should be possible and setting out a protocol for achieving this. The emerging unitary model, code named ‘phots’, was created for heuristic purposes but is turning out to have some promise. 

Introduction 

Mach-Zehnder, interferometers were invented by Ludwig Mach (son of Ernst Mach) and Ludwig Zehnder in 1891-92. 

Fig 1:  Mach-Zehnder Interferometer.  Coherent monochromatic light from a laser encounters a half silvered plate glass beam splitter. A phot is reflected upwards onto a full mirror where it reflected a second time across to another half silvered plate to a detector at D1.  Each reflection causes a 180° change in phase.  The phot travelling the lower path experiences a 180° phase change when reflected at the first mirror (lower right) and again when reflected from the silver coating on half of the plate beam splitter 2 (top right). It is in phase with the phot travelling the top path and also ends up in Detector 1.  The top detector D2 receives no light when the two path lengths are optically equal.  

The wave model explanation is that a wavelike object is divided in two by the first beam splitter and these waves combine constructively and destructively at the second beam splitter.  However, if the upper or lower bean is sampled it is found to be exactly the same color (and hence frequency and energy) as the incident light, so this explanation is inadequate.

Attempts to describe what is going on in terms of photons also run into problems, especially in dim light conditions where photons are considered to enter the apparatus in small numbers.  Start by taking away the second beam splitter.  At the first beam splitter there are equal probabilities for transmission or reflection. It is observed that photons show up in equal numbers at the two detectors, but not both at the same time.  

Now replace the second beam splitter and make sure the two paths are exactly the same optical length (i.e. after allowing for slightly lower speeds in the glass plates etc).  If the top path passes straight through beam splitter 2 then it will have endured two reflections and one transmission.  Total phase change of 360°.  If the top path passes straight through beam splitter 2 then it will have endured one reflection and two transmissions.  Total phase change of 180°.  But if it is reflected at beam splitter 2 then it will have endured two reflections and one transmission.  Total phase change of 360°. 

It is observed that light reaches D1 but not D2.  No light reaches the top detector D2.  This is interpreted as the upper beam interfering with the lower beam in a wave like interaction.  But how can a photon be a discrete particle from the time of its emission to the time of its detection, and travel by one path or the other when there is no second beam splitter, and yet interfere with itself as if it had travelled by both paths simultaneously when there is a second beam splitter?

This is where the conventional explanations start to unravel.  Proponents say that each photon “knows” whether there is a second beam splitter or not.  John Wheeler then developed a thought experiment that inserts the second beam splitter when the photon is already in flight and has already passed the first beam splitter.  This is called a delayed choice experiment.  Experimenters put this idea into practice with some ultra fast switching equipment.  They start with the second beam splitter removed until the photon could be inferred to have already passed the first beam splitter and then quickly switched it in.  All the light went to the right hand detector and none to the top detector. 

Then they started with the second beam splitter in place until the photon could be inferred to have already passed the first beam splitter and then quickly switched it out.  All the light went to the right hand detector and none to the top detector. 

It was as if the photon knew what was going to happen to it and decided to change its nature into a wave and interfere with itself before reaching the detectors.  Some scientists accept this as an explanation, some do not.  Some even believe that choices about detection methods can retrospectively alter what happened prior to that choice being made!  Some just conclude it is just another weird and paradoxical behavior that occurs at quantum levels.  Einstein thought this was all a lot of nonsense.

Phots

Pervious essays (Van de Vusse, 2024) have described a protocol for developing a unified model for light by abandoning inappropriate analogies and accumulating experimental evidence in a logical framework that eventually allows a unified model to emerge by logic and the sheer weight of evidence.  To illustrate the approach the essays present a prototype example of light called phots.  

The model that starts to emerge suggests properties as follows.  (This is a summary only and all aspects are provisional until a lot more evidence is collected and analysed by whoever can contribute).  Phots are discrete parcels of energy.  They reveal properties when they interact and then only one property at a time.  In keeping with special relativity phots in flight have no length and are essentially unchanging.  They have effective width but no length so should be thought of as being two dimensional in a plane orthogonal to the line of travel.  Phots cannot be observed “in flight”.  To observe a phot is to destroy it.  However, lot can be inferred from how they interact and from how their ‘fellow travelers’ interact.  The basic properties of a phot are energy, phase, orientation, spin, and line of travel.  They convey energy, momentum, angular momentum and spin.  Their effects upon interaction can be modeled as coming from two sine wave drivers of electromagnetic energy.  Phots of the same energy level travelling close together in time and space may have a tendency to pair up.

Mach-Zehnder interferometers are reasonably simple, but this is deceptive.  It seems that Nature is presenting a paradox but it is probably just a conjuring trick involving misdirection and false assumptions.  One of the false assumptions in the particle explanation is that photons behave like little rubber bullets.

In the phot approach a lot more consideration is given to what happens at the beam splitters and mirrors.  A phot doesn’t just ‘bounce off’ a mirror.  It is absorbed into or onto the surface of a mirror and is reborn as a ‘child phot’.  The child phot is not exactly the same as its parent.  Its phase is changed by 180°.  But what about all its other properties?  Energy, orientation, momentum, spin and line of travel?  Energy stays the same, momentum and line of travel of travel clearly change, and orientation probably changes.  Spin might change sign as well – the investigation has not reached that question yet, but the answer is probably yes.

Here is a conjecture for what is going on in terms of the provisional prototype phot model.

Conjectured explanation in terms of phots

When a phot encounters a half silvered plate beam splitter it does not behave capriciously.  It is reflected upwards or it passes straight through for some physical reasons.  And it leaves traces of what it has done.  Momentum, angular momentum and spin all have to be conserved.  This affects what happens to the next phot that comes along.  The child phot that it creates is affected by what happened to the previous phot/child phot event or events.  One of the things that seems likely in this scenario is that if a child phot leaves the beam splitter via one path then the next phot that comes along will likely produce a child phot that leaves via the other path.

The timing of events at the first beam splitter is open to question.  It is possible in principle that the first phot gets held up at the beam splitter because it does not know what to do.  Half of it is trying to be reflected and the other half is trying to be refracted.  It sits there as a disturbance in the beam splitter, probably in the silver.  When another phot comes along the combined disturbance results in the birth of two child phots, more or less at the same time, one in either direction, both with the same energy and phase. Note that the child phot on the lower path has to travel through the glass plate.

At the angled mirrors the child phots give rise to new child phots.  The main change is a 180° change in phase.  The orientation is likely to change as well.  

Let us not bother to call the next generation grandchild phots etc. The point has been made and the typing becomes cumbersome.  All subsequent phots will just be called phots.

Things become more interesting when these new phots interact with the second beam splitter.  Note that this is not a repeat of what happened at the first beam splitter.  There are three differences worth noting.  The first is that the top phot is entering the beam splitter from the back, i.e. through the glass plate.  Secondly the two phots are reaching the edge of the silver coating at the same time, plus a time difference for any difference in their optical path lengths.  Thirdly they approach the second beams splitter not in parallel paths but in paths that are at right angles to each other.

Each phot delivers sinsusoidal disturbances to the second beam splitter, probably concentrated in the highly conductive silver layer.  These disturbances reinforce each other constructively or destructively just like electromagnetic waves and the resultant child phots go off in the appropriate directions as predicted by wave mechanics.

Discussion

The conjectured explanation in terms of phots may have merit or not.  But at least it suggests that a simple explanation might be possible.  

This particular conjecture has not problems dealing with Wheeler’s delayed choice experiments.  The conjecture is also consistent with Conjecture #2 offered up as an explanation for Young’s Double Slit experiment.  Maybe the evidence is trying to tell us something.

References

Van de Vusse, Sjoerd B.A., 2024, Some ideas and experiments for issues affecting modern physics,   https://hereticalphysics.com.au

John Archibald Wheeler, 1978, “‘The Past’ and the ‘Delayed Choice’ Double-Slit experiment,” which appeared in 1978 and has been reprinted in several locations, e.g. Lisa M. Dolling, Arthur F. Gianelli, Glenn N. Statilem, Readings in the Development of Physical Theory, p. 486ff.

Author contact:  SBAvan@utas.edu.au
Author’s location:  Hobart, Australia