Torsion talks only to spin: the CMB temperature as a test of energy exchange in Einstein-Cartan cosmology

arXiv:2611.08422 · gr-qc, astro-ph.CO · Submitted 2026-11-05 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: General Relativity and Quantum Cosmology. The day in general relativity and quantum cosmology.

Kai: We're looking at the paper: "Torsion talks only to spin".

Mira: Here's the gist of the paper. The cosmic microwave background temperature at z = 0.886 is known to 2% from molecular absorption towards PKS 1830−211; CO absorbers and Sunyaev–Zel’dovich clusters extend the measurement to z ≃ 3 <ref:2611.08422#pg6> The paper investigates how energy exchanges between matter components and geometry in Einstein–Cartan cosmology, specifically determining which sector receives the exchanged energy based on Noether identities, leading to a universal allocation rule where photons gain almost a third of their energy The Allocation Rule Derived from Torsion In Einstein–Cartan theory, the Noether identities of each sector fix the answer to how geometry exchanges energy with matter, stating that a sector’s energy–momentum fails to be conserved only in proportion to its own spin The relevant example is light, which has none and scales as 1 + z with no correction at all Proposition 1 states that the allocation rule is linear in the spin of that same sector For a spinless sector, it is conserved with the Levi-Civita derivative for any torsion and any dynamics that produces it Observational Constraints from Cosmological Probes The model predicts a background colder than T0(1 + z) by 7% at z ≃ 0.9 and by 8.5% at z ≳ 3, with no free parameter PKS 1830−211 gives 5.08 ± 0.10 K against a predicted 4.78 K The CO absorbers favor the standard scaling by ∆χ squared = 13, and Planck cluster bins give χ squared = 50.8 versus 15.5 The Role of Thermometers The energy balance does not say what that does to the spectrum, and here the FIRAS measurement does most of the deciding Molecules and atoms in absorbing galaxies prefer the standard scaling by ∆χ squared = 13; SZ clusters, which read a distortion carried from the cluster to us, say so with more weight The CO absorbers are noted as needing only the background, today’s temperature and the catalogued redshifts Constraints on Alternative Scenarios The universal allocation prescription implies that dark matter must carry vector spin, something minimally coupled Dirac fermions do not have This reassignment leads to a comparison where the low-redshift data cannot tell the two allocations apart, suggesting that the support they give to both is consistent with the model Limits and Future Directions The rule suggests that minimally coupled photons are conserved, which supports minimal coupling predictions for light The paper concludes that a kinetic operator local in momentum does not survive the data, leaving the CO absorbers as the cleanest statement of the problem This implies that any operator of the form Ccre + A(p, t)

f − fPl: is excluded by observational tests The rule dictates that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing The model of Ref.

19: , with σ ∝ E n, is not adiabatic and falls outside the hypotheses of that result Nothing said here argues against torsion as late dark energy, only against an allocation that hands part of the exchange to the photons The Observational Test The primary anisotropies would change their spectral shape by amounts Planck excludes The clusters weigh more than the absorbers, but they need one assumption more, about how a perturbation travels, and the absorbers do not The two agree in this regard The Final Verdict on Allocation The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The model's effective slope over the CO redshifts is βeff ≃ 0.07–0.09 The absorbers at z = 6.34 water absorption brackets the temperature between 16.4 and 30.2 K

6: The Final Constraint The absorbers at those redshifts therefore depend on the regularity of the source, and the CO systems, to this extent, do not The CO systems are individually less precise, with errors of six to twenty-two percent The absorber test is at the edge of present systematics The catalogued redshifts must be 1 + z = a0/a, which the fit of Ref.

19: assumes and which excludes a common rescaling of photon energies The Implication for Dark Matter Dark matter must carry vector spin, something minimally coupled Dirac fermions do not have The carrier would have to be lighter than m ≲ 3/2 somegacħH0/X0/H0 ≈ ħH0 ∼ 10−33 eV Whatever carries the spin is not a gas of particles diluting with the expansion but a coherent medium whose hypermomentum does not scale with particle number The Fate of Early Universe Energy If torsion is to reheat the universe after inflation, it can only hand energy to sectors with spin A fully polarised relativistic fermion bath receives at most Qr ≤ 3/2 λħnrH2 The classical exchange of EC theory with minimally coupled fermions cannot convert the energy that drives the expansion into radiation The Final Test The CO absorbers remain the cleanest statement of the problem, because they use the energy balance, today’s temperature and the catalogued redshifts The model's injection is 4.4% of the standard one at z = 0.886, enough to raise to its standard level any band a factor of two wide whose lower edge lies below 136 GHz in the rest frame The absorbers at those redshifts therefore depend on the regularity of the source, and the CO systems, to this extent, do not The Final Conclusion We adopt Ccre, which is the reading left to the model, and we say below which part of the test depends on it The two thermometers under each operator read T(z) at the catalogued redshift, and the perturbation y G free-streams The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The Final Test on Anisotropies The primary anisotropies would have the spectral shape of the temperature derivative of a Planck function at Trec/(1 + zrec) = 0.915 T0 = 2.49 K The creation reading is therefore excluded on its own terms, and so, by the paragraph before last, is the rescaling and any mixture of the two The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The Final Constraint on Operators Any operator of the form Ccre + A(p, t)

f − fPl: is excluded by observational tests The only operator that leaves g to Liouville’s equation, transported with pa fixed as in general relativity, is A = B = 0 The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The Final Summary The rule dictates that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing The model of Ref.

19: , with σ ∝ E n, is not adiabatic and falls outside the hypotheses of that result Nothing said here argues against torsion as late dark energy, only against an allocation that hands part of the exchange to the photons The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The model's effective slope over the CO redshifts is βeff ≃ 0.07–0.09 The absorbers at z = 6.34 water absorption brackets the temperature between 16.4 and 30.2 K

6: The Final Summary The rule dictates that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing The model of Ref.

Title and authors: Kai: So what's the core idea they are summarizing here about this entire study, specifically regarding how torsion affects energy exchange in Einstein-Cartan cosmology?

Mira: The main point is that because of the way geometry and matter interact in this theory, the total energy balance equation doesn't give a single answer about which component gets the exchanged energy >

Lev: It’s like a family bank statement where it records everything that came in and went out, but it doesn't say who spent what >

Kai: And because of that ambiguity, the paper investigates how to resolve this by using the Noether identities for each sector >

Mira: They show that those identities fix the answer: a sector’s energy momentum fails to be conserved only in proportion to its own spin >

Lev: That leads directly into their allocation rule, which says a sector receives energy in direct proportion to its own spin >

Kai: And what about light, which has zero spin? The paper suggests that minimally coupled light is conserved whatever the torsion does >

Mira: So, the universal allocation rule derived from this is that photons gain almost a third of their energy from the geometry after recombination >

Lev: That prediction comes from assuming a metric-affine cosmology fitted to DESI DR2 data, which revived the programme of interacting dark sectors >

Kai: The paper is essentially using these cosmological probes to test whether this universal allocation rule holds true in reality or if some other sector gets the energy instead >

Mira: They use several different probes—molecular absorption, CO absorbers, and Sunyaev–Zel’dovich clusters—to see if they all point to the same outcome or if they are inconsistent >

Lev: The key here is that this framework allows them to test torsion as a form of late dark energy, but it also tests the allocation rule itself >

Kai: So, in short, they are using specific measurements to pin down which sector gets the injected energy when geometry and matter interact in this torsion-based gravity >

The paper's summary: Mira: Now that we've established what the main finding is—the universal allocation rule—the authors discuss how their work improves upon existing cosmological models or previous work on this subject >

Lev: They point out that a prior model, specifically the one from reference eighteen, which deals with interacting dark sectors, provided a way to derive an interaction term >

Kai: So they are building on that by showing how torsion offers a different way to derive that interaction term based on spin >

Mira: The improvement is showing that torsion provides a concrete mechanism for energy exchange—it’s not just an abstract idea anymore >

Lev: It moves it from being a theoretical possibility to something constrained by observational tests because they have to adhere to the allocation rule derived from the Noether identities >

Kai: So, what's the practical improvement for someone trying to build a model? They get a much tighter constraint on how you can parameterize your theory based on what we see in data >

Mira: Exactly. They show that if you want to fit your theory to the data, you have to follow this allocation rule derived from torsion and spin >

Lev: And they also find that minimally coupled photons are conserved, which supports minimal coupling predictions for light in their context >

Kai: That means the kinetic operator local in momentum doesn't survive the data, which excludes some other operators from being part of the theory >

Mira: So they rule out a specific form of energy exchange operator: Ccre plus A of p t f minus fPl is excluded by observational tests >

The paper's improvements: Kai: Alright, let's wrap up with the conclusion of "Torsion talks only to spin: the CMB temperature as a test of energy exchange in Einstein-Cartan cosmology." What are the main implications for us?

Mira: The final verdict is that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing >

Lev: And this implies that dark matter must carry vector spin, which isn't what minimally coupled Dirac fermions do naturally >

Kai: So the implication for us is that if you want this model to work, dark matter needs some sort of exotic coupling to geometry >

Mira: But the paper also says that low-redshift data cannot distinguish between the two allocations anymore, suggesting the support they give to both is consistent with their model >

Lev: So it's a subtle point. The support for both scenarios is compatible with this universal allocation prescription in the low-redshift data >

Kai: It seems like the CO absorbers are still the cleanest statement of the problem because they use the energy balance, today’s temperature, and catalogued redshifts >

Mira: Because they are less precise individually, with errors of six to twenty-two percent >

Lev: The effective slope over those CO redshifts is about zero point zero seven to zero point zero nine >

Kai: That's a very small slope, and the absorbers at redshift six point three four water absorption bracket the temperature between sixteen point four and thirty point two Kelvin >

Conclusion: Mira: To summarize, the main message of "Torsion talks only to spin: the CMB temperature as a test of energy exchange in Einstein-Cartan cosmology" is that torsion dictates that each sector gets energy based on its spin >

Kai: And this means a spinless sector gets nothing from the geometric exchange >

Lev: It also suggests dark matter must carry vector spin, which isn't what minimally coupled Dirac fermions do naturally >

Mira: But the paper is careful to point out that low-redshift data doesn't allow us to tell the two allocations apart because they are both consistent with this model >

Kai: And they stress that CO absorbers remain the cleanest statement of the problem because they use all three pieces of evidence: energy balance, today’s temperature, and redshift information >

Lev: That means any operator of the form Ccre plus A of p t f minus fPl is excluded by observational tests >

Mira: So while the model is constrained, it's not saying torsion isn't late dark energy; it just says that you can't allocate the energy to photons in a way that doesn't match what FIRAS sees >

Kai: It leaves us with this picture of how geometry and matter exchange energy depending on spin >

Departamento de Ciencias Exactas, Facultad de Ingeniería, Universidad San Sebastian · Instituto de Fisica, Pontificia Universidad Catolica de Valparaiso

gr-qc, astro-ph.CO

Submitted: 2026-11-05

Updated: 2026-11-05

Comments: 31 pages, 1 figure

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 79/100

The gist: The cosmic microwave background temperature at z = 0.886 is known to 2% from molecular absorption towards PKS 1830−211…

Key concepts

Einstein–Cartan cosmology
This is a modification of Einstein's gravity theory that includes torsion, an extra geometric property. The paper explores how energy moves between matter components and this geometry, specifically looking at how different parts of the universe share the exchanged energy based on their internal spin.
Noether identities
These are mathematical rules derived from symmetries in physics. In this context, they fix the exact way energy exchanges occur between different sectors (like light or matter) and geometry. They dictate that a sector's energy-momentum conservation only fails if it has a specific spin.
Universal allocation rule
This is the derived principle stating that each sector receives energy in proportion to its own spin. A spinless sector receives no exchanged energy, while sectors with spin receive a fixed fraction of the exchange, leading to specific predictions about how photons gain energy.
CO absorbers
These are molecular clouds used as observational probes in cosmology. They are favored by the model because they allow for a clean test of the energy balance using measurements like today's temperature and catalogued redshifts, making them the most precise constraint mentioned.

Terminology

Summary

The cosmic microwave background temperature at z = 0. p. 0886 is known to 2% from molecular absorption towards PKS 1830−211; CO absorbers and Sunyaev–Zel’dovich clusters extend the measurement to z ≃ 3 <ref:2611. p. 008422#pg6>

The paper investigates how energy exchanges between matter components and geometry in Einstein–Cartan cosmology, specifically determining which sector receives the exchanged energy based on Noether identities, leading to a universal allocation rule where photons gain almost a third of their energy

The Allocation Rule Derived from Torsion

In Einstein–Cartan theory, the Noether identities of each sector fix the answer to how geometry exchanges energy with matter, stating that a sector’s energy–momentum fails to be conserved only in proportion to its own spin The relevant example is light, which has none and scales as 1 + z with no correction at all Proposition 1 states that the allocation rule is linear in the spin of that same sector For a spinless sector, it is conserved with the Levi-Civita derivative for any torsion and any dynamics that produces it

Observational Constraints from Cosmological Probes

The model predicts a background colder than T0(1 + z) by 7% at z ≃ 0.9 and by 8.5% at z ≳ 3, with no free parameter PKS 1830−211 gives 5. p. 008 ± 0.10 K against a predicted 4. p. 078 K The CO absorbers favor the standard scaling by ∆χ squared = 13, and Planck cluster bins give χ squared = 50. p. 08 versus 15. p. 05

The Role of Thermometers

The energy balance does not say what that does to the spectrum, and here the FIRAS measurement does most of the deciding Molecules and atoms in absorbing galaxies prefer the standard scaling by ∆χ squared = 13; SZ clusters, which read a distortion carried from the cluster to us, say so with more weight The CO absorbers are noted as needing only the background, today’s temperature and the catalogued redshifts

Constraints on Alternative Scenarios

The universal allocation prescription implies that dark matter must carry vector spin, something minimally coupled Dirac fermions do not have This reassignment leads to a comparison where the low-redshift data cannot tell the two allocations apart, suggesting that the support they give to both is consistent with the model

Limits and Future Directions

The rule suggests that minimally coupled photons are conserved, which supports minimal coupling predictions for light The paper concludes that a kinetic operator local in momentum does not survive the data, leaving the CO absorbers as the cleanest statement of the problem This implies that any operator of the form Ccre + A(p, t) [f − fPl] is excluded by observational tests

Conclusion on Energy Exchange

The rule dictates that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing The model of Ref. [19], with σ ∝ E n, is not adiabatic and falls outside the hypotheses of that result Nothing said here argues against torsion as late dark energy, only against an allocation that hands part of the exchange to the photons

The Observational Test

The primary anisotropies would change their spectral shape by amounts Planck excludes The clusters weigh more than the absorbers, but they need one assumption more, about how a perturbation travels, and the absorbers do not The two agree in this regard

The Final Verdict on Allocation

The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The model's effective slope over the CO redshifts is βeff ≃ 0.07–0.09 The absorbers at z = 6. p. 0 p. 0 p. 034 water absorption brackets the temperature between 16. p. 0 p. 0 p. 04 and 30.2 K [6]

The Final Constraint

The absorbers at those redshifts therefore depend on the regularity of the source, and the CO systems, to this extent, do not The CO systems are individually less precise, with errors of six to twenty-two percent The absorber test is at the edge of present systematics The catalogued redshifts must be 1 + z = a0/a, which the fit of Ref. [19] assumes and which excludes a common rescaling of photon energies

The Implication for Dark Matter

Dark matter must carry vector spin, something minimally coupled Dirac fermions do not have The carrier would have to be lighter than m ≲ 3/2 somegacħH0/X0/H0 ≈ ħH0 ∼ 10−33 eV Whatever carries the spin is not a gas of particles diluting with the expansion but a coherent medium whose hypermomentum does not scale with particle number

The Fate of Early Universe Energy

If torsion is to reheat the universe after inflation, it can only hand energy to sectors with spin A fully polarised relativistic fermion bath receives at most Qr ≤ 3/2 λħnrH2 The classical exchange of EC theory with minimally coupled fermions cannot convert the energy that drives the expansion into radiation

The Final Test

The CO absorbers remain the cleanest statement of the problem, because they use the energy balance, today’s temperature and the catalogued redshifts The model's injection is 4.4% of the standard one at z = 0. p. 0886, enough to raise to its standard level any band a factor of two wide whose lower edge lies below 136 GHz in the rest frame The absorbers at those redshifts therefore depend on the regularity of the source, and the CO systems, to this extent, do not

The Final Conclusion

We adopt Ccre, which is the reading left to the model, and we say below which part of the test depends on it The two thermometers under each operator read T(z) at the catalogued redshift, and the perturbation y G free-streams The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts

The Final Test on Anisotropies

The primary anisotropies would have the spectral shape of the temperature derivative of a Planck function at Trec/(1 + zrec) = 0.915 T0 = 2. p. 949 K The creation reading is therefore excluded on its own terms, and so, by the paragraph before last, is the rescaling and any mixture of the two The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts

The Final Constraint on Operators

Any operator of the form Ccre + A(p, t) [f − fPl] is excluded by observational tests The only operator that leaves g to Liouville’s equation, transported with pa fixed as in general relativity, is A = B = 0 The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts

The Final Summary

The rule dictates that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing The model of Ref. [19], with σ ∝ E n, is not adiabatic and falls outside the hypotheses of that result Nothing said here argues against torsion as late dark energy, only against an allocation that hands part of the exchange to the photons The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The model's effective slope over the CO redshifts is βeff ≃ 0.07–0.09 The absorbers at z = 6.34 water absorption brackets the temperature between 16.4 and 30.2 K [6]

The rule dictates that each sector receives energy in proportion to its own spin, and a spinless sector receives nothing The model of Ref. [19], with σ ∝ E n, is not adiabatic and falls outside the hypotheses of that result Nothing said here argues against torsion as late dark energy, only against an allocation that hands part of the exchange to the photons The CO absorbers remain the cleanest statement of the problem because they use the energy balance, today’s temperature and the catalogued redshifts The model's effective slope over the CO redshifts is βeff ≃ 0.07–0.09 The absorbers at z = 6.34 water absorption brackets the temperature between 16.4 and 30.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI can perform causal inference on cosmological models by analyzing how different energy exchange allocations among matter components affect observable constraints like CMB temperature and quasar absorption line temperatures.

  2. Improved AI can distinguish between different physical interpretations of observational data by testing the predictions of various allocation rules, such as the universal allocation versus scenarios where dark matter carries the spin.

  3. Improved AI can evaluate the robustness of cosmological models against systematic uncertainties by comparing results derived from different thermometers, such as comparing CO absorber measurements to Sunyaev-Zel'dovich cluster measurements.

  4. Improved AI can predict how specific observational probes like FIRAS constrain the spectral shape of primary anisotropies by analyzing the kinetic argument behind photon production and its effect on frequency dependence.

Sources

Related papers