[Jarek Duda (OP)]

Time/CPT symmetry is at heart of many models of physics, like unitary evolution in quantum mechanics, or Lagrangian formalism we use from classical mechanics, electromagnetism, up to general relativity and quantum field theories.
In theory we should be able to decompose any scenario (history of the Universe?) into ensemble of Feynman diagrams, apply CPT symmetry to all of them, getting CPT analogue of entire scenario (?)

There are many QM-based experiments which kind of use time symmetry (?), for example (slides with links):
Wheeler experiment, delayed choice quantum eraser (DCQE), “asking photons where they have been”, “photonic quantum routers”, Shor algorithm as more sophisticated DCQE.

However, this symmetry is quite nonintuitive, very difficult to really accept – mainly due to irreversibly, thermodynamical counterarguments (are there other reasons?)
Can e.g. this conflict with 2nd law of thermodynamics be resolved by just saying that symmetry of fundamental theories can be broken on the level of solution, like throwing a rock into symmetric lake surface?
Are all processes reversible? (e.g. wavefunction collapse, measurement)

So is our world time/CPT symmetric?
What does it mean?

Personally I interpret it that we live in 4D spacetime, (Einstein's) block universe/eternalism: only travel through some solution (history of the universe) already found in time/CPT symmetric way, like the least action principle or Feynman path/diagram ensemble - is it the proper way to understand this symmetry?
Are there other ways to interpret it?

[yor_on]

You should define what CPT symmetry is Jarek as that is from where this question seem to arise. And CPT is a broken symmetry as I understands it. Using a Feynman diagram it do seems as if time is a symmetry though. You could put it as a question of the logic holding the universe together would hold for it running backwards, if it is time you speculate over?

https://www2.lbl.gov/nsd/education/ABC/wallchart/chapters/05/2.html

[Jarek Duda (OP)]

CPT symmetry is required for QFT like the Standard Model - from https://en.wikipedia.org/wiki/CPT_symmetry

The CPT theorem says that CPT symmetry holds for all physical phenomena, or more precisely, that any Lorentz invariant local quantum field theory with a Hermitian Hamiltonian must have CPT symmetry.

The question is how should we understand it?

The basic understanding is that any Feynman diagram can be transformed this way: applying spatial, temporal symmetry and replacing particles with antiparticles ...
... antiparticles "traveling back in time": https://en.wikipedia.org/wiki/Antiparticle#Feynman%E2%80%93Stueckelberg_interpretation

But in theory we should be able to decompose any scenario into Feynaman diagrams, apply CPT, getting CPT analogue of entire scenario ...
Cannot we imagine History of the Universe this way?

For me the only way to understand it is through living in spacetime/block universe which was originally solved in CPT-symmetric way like the least action principle, Feynman path/diagram ensembles.
It is also called eternalism: https://en.wikipedia.org/wiki/Eternalism_(philosophy_of_time)

That e.g. thermodynamical (effective) breaking of this symmetry is only a property of the solution - like symmetry of lake surface broken by throwing a rock.
This "rock" for our Universe seems the Big Bang - where everything was localized, so entropy should be low (?)

The question to discuss here is if it is the proper understanding?
Are there other ways to see it?

ps. Slides with lots of links and diagrams: https://www.dropbox.com/s/0zl18yttgnpc52w/causality.pdf

[yor_on]

I don't know Jarek. What I do know is that your mind is a living thing, It search and it question. Keep it that way, it would be a pleasure to meet you.

[Jarek Duda (OP)]

We have QM Foundations & Nature of Time seminar, e.g. Kenneth Wharton has recently talked about time symmetry as the source of quantum nonintuitiveness: http://th.if.uj.edu.pl/~dudaj/QMFNoT
You can join if wanting to discuss, also talks are welcomed.

[yor_on]

You know one thing Jarek. The impression I've gained from Einstein isn't that he was dogmatic. He just used his mind., then he spent a lot of time defending his ideas, trying them to become mathematically explainable. Nobody's perfect, nor me or you. But your mind is a different thing, when you use it, it shines.

[Jarek Duda (OP)]

Thanks, seeing time-symmetry as the source of nonintuitiveness of quantum mechanics is an old concept:
https://en.wikipedia.org/wiki/Two-state_vector_formalism

I got to this conclusion more than a decade ago by studying MERW - which explains why standard diffusion disagrees with physics: only approximates the maximal entropy principle crucial for statistical physics models.
Repairing this approximation, using time-symmetric random walk, this disagreement disappears:
https://en.wikipedia.org/wiki/Maximal_entropy_random_walk

I recommend looking closer at quantum experiments directly exploiting time symmetry, like:
https://en.wikipedia.org/wiki/Wheeler%27s_delayed-choice_experiment
https://en.wikipedia.org/wiki/Delayed-choice_quantum_eraser

[Jaaanosik]

Jarek,
Here is a quote:

Electromagnetism, gravity and the strong interaction all obey C-symmetry, but weak interactions violate C-symmetry.

From here https://en.wikipedia.org/wiki/C-symmetry
Please, can you elaborate on the cause why weak interactions violate C-symmetry?
Thanks,
Jano

[Jarek Duda (OP)]

Time symmetry is satisfied e.g. in electromagnetism, gravity and quantum mechanics.
In standard model it is more subtle - for which it is believed that applying this symmetry also requires to apply C and P - personally I am not so certain of "kaons decay in two ways, hence CP is violated" types of conclusions, but this is for a completely different discussion.

Let us focus on time symmetry here, if needed combined with CP.

For example: are there possible CPT analogues of any phenomena?
E.g. of laser - which stimulated emission causes excitation of target later.
Is CPT analogue of laser possible? If so, shouldn't its stimulated absorption cause deexcitation of (constantly excited like lamp) target earlier?
For free electron laser building its CPT analogue seems quite straightforward ...

[evan_au]

Is CPT analogue of laser possible?

I don't think so.

The clue is in the metastable state that lases.
- The laser medium has a ground state
- You have to pump the laser medium up to an energy level which is too unstable to lase (it doesn't hang around long enough)
- But this unstable state decays into a metastable state which doesn't readily decay by itself. This decay is dissipated as heat.
- Most of the laser medium has to be in this metastable state before lasing can occur (a population inversion)
      - If too many atoms are in the ground state, they will just absorb any laser output before it has time to build up a coherent resonance.
      - This part of the process is certainly reversible!
- The metastable state stays around long enough to be triggered by a passing photon of the right frequency into emitting a photon of identical frequency and phase (stimulated emission of radiation).

The reason that illuminating a laser medium with light does not produce electricity + a cooling effect is:
- The incoming laser photons have the right energy to pump electrons from the ground state to the metastable state
- But they lack the energy to pump the electrons to the higher energy state - this would involve turning heat into orderly energy, which violates entropy
- Different lasers have different mechanisms for producing the metastable state- some laser diodes inject holes and electrons into the laser medium; to reverse this, thermal energy would have to cause random holes and electrons to undergo an orderly diffusion in opposite directions, again violating entropy.
- In general, turning a source of heat into useful energy has to battle the Carnot equation, which was the original basis of thermodynamics.
See: https://en.wikipedia.org/wiki/Laser_diode#Direct_and_indirect_bandgap_semiconductors

In general, I think that interactions on a Feynman diagram with 2 particles on each side might be reversible.
- But by the time you get 3 or more particles on one side of the Feynman diagram, entropy holds the trump card.
- Because it is statistically unlikely that all 3 components would appear in the right place at the right time with the right energy to reverse the process.

[Jarek Duda (OP)]

evan_au,
while I also don't know how to realize CPT analogue of laser based on population inversion for 3/4 states ... for free electron laser it seems quite straightforward:

A free-electron laser (FEL) is a (fourth generation) synchrotron light source producing extremely brilliant and short pulses of synchrotron radiation. An FEL functions and behaves in many ways like a laser, but instead of using stimulated emission from atomic or molecular excitations, it employs relativistic electrons as a gain medium.[1][2] Synchrotron radiation is generated as a bunch of electrons passes through a magnetic structure (called undulator or wiggler). In an FEL, this radiation is further amplified as the synchrotron radiation re-interacts with the electron bunch such that the electrons start to emit coherently, thus allowing an exponential increase in overall radiation intensity.

It is just electrons traveling in alternating magnetic field.
Isn't its CPT analogue just positrons traveling in opposite direction?

ps. There is also "antilaser" concept, but it is a bit different: “coherent perfect absorber” https://www.wired.com/2010/07/antilaser/
Here we would like to get CPT analogue of excitation of target caused by laser: cause deexcitation of target ... which comes naturally from thermodynamics, but maybe we could make it faster with CPT analogue of laser?

[CPT ArkAngel]

The problem of CPT symmetry in particle physics arises mainly due to an error in thinking that all electric charges are the same with the exception of their sign. The charge of the electron and the charge of the proton are not exactly the same. The minimal distances of interaction are different so the potential are the same at long distance but not at the minimum. But when you consider this, it is at the scale of the universe that there is an asymmetry. There is the same quantity of protons and electrons, but almost no positron  and no constituent of the antiproton. The total potential doesn't seem to be equal to zero...

CPT symmetry is about conservation. If you suppose no creation, but conservation. There must be a complement somewhere.
What is dark matter? Why does the universe expand... Why gravity is so weak? The complement does not interact with the electric charge of matter, only at the beginning and at the end of the process, where CPT is conserved.

Time CPT symmetry is either space or time.

[Jarek Duda (OP)]

CPT ArkAngel,
As for throwing a rock into symmetric lake surface, or entropy growth, there is no problem with violating fundamental symmetries in the level of solution.

For entropy growth I would say the reason is that in Big Bang everything was localized, hence entropy was low - leading to observed entropy gradient/growth.

For baryon asymmetry, I would say it is statistical symmetry breaking: that symmetric situation is statistically repelling.
Like for e.g. L-amino acids in our life: while there is symmetry, the more one type of them, the easier development of life based on the same - making 1/2-1/2 statistically repelling, leading to nearly complete domination of one of them - originally chosen e.g. in a random way.

For baryon asymmetry we don't need CPT violation, it is sufficient to have this kind of statistical repelling.
For example that in baryogenesis:
"creation of proton is a bit more likely in presence of electron" plus symmetric
"creation of antiproton is a bit more likely in presence of positron"
such two symmetric rules would make that 1/2-1/2 matter-antimatter state is statistically repelling, leading to nearly complete domination of one of them, e.g. initially chosen in a random way.

[CPT ArkAngel]

Thank you Jarek, you helped me to understand my point further.

Yes, if you take a neutral plasma, statistically you have a repulsion maintaining the density of the plasma according to the temperature and pressure. But at high densities, the experiment defies the theory. The calculations used are mostly empirical rather than completely theoretical. Why a neutron star has a magnetic field that strong? What is dark energy? When the density is low, both charges are equal because it respects the minimal distance but when you compress further, an asymmetry of the charge is produced. I say add electrons or muons in your plasma to help your fusion in fusion reactors. If you can...

[Jarek Duda (OP)]

I don't think I follow?

The only topic-related question I have seen here is the dominance of matter, for which CPT violation is not necessary - other options are e.g.:
- fixed baryon number of the Universe - in Big Crunch preceding our Big Bang,
- statistical symmetry breaking - some positive feedback mechanism like above, making "the more of one type, the easier to get even more" - finally leading to domination of one type, which initially could be chosen at random, analogously to chirality of life.

[CPT ArkAngel]

I say fundamentally, it is not statistical and I have a possible example.