Pith. sign in

REVIEW 4 major objections 4 minor 25 references

Finite-Temperature Spin-Adapted ROKS and TDDFT

T0 review · 4 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Boltzmann-averaging high-spin components puts temperature into spin-adapted DFT.

desk verdict A solid ground-state extension with a genuine gap in the response derivation; the T1→T6 assignment rests on an unproven averaged-pencil approximation. read the letter →

arxiv 2608.08454 v1 pith:VJ3O6WNO submitted 2026-08-09 physics.chem-ph

Xiaoyu Zhang This is my paper · ORCID
classification physics.chem-ph
keywords finite-temperatureDFTspin-adaptedTDDFTROKSthermalensemblediradicaloidtriplet-triplettransitionsoscillatorstrengthspinpurity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proposes a way to include temperature in spin-adapted density functional theory for open-shell molecules without breaking spin symmetry. It builds a canonical ensemble from integer high-spin restricted open-shell Kohn–Sham (ROKS) components, each weighted by a Boltzmann factor, and then averages the spin-adapted TDDFT response matrix pairs of those components in a common orbital basis. In the zero-temperature single-component limit the theory reduces to ordinary high-spin ROKS and the corresponding spin-adapted TDDFT. Applied to the diradicaloid TTM–TTM, the calculation assigns the thermally activated absorption near 425 nm to T1→T6 and other allowed triplet–triplet transitions, whose intensities grow as the triplet population rises with temperature.

What carries the argument

The central object is the thermal ensemble built from integer high-spin ROKS components $|\Phi^{S,S}_I\rangle$, each assigning hard closed/open/virtual labels to spatial orbitals and each carrying Boltzmann weight $w_I \propto \exp(-\beta E_I)$. Because every component has the same $\hat{S}^2$ eigenvalue, the ensemble density operator is spin-pure. The argument is carried by embedding the zero-temperature spin-adapted TDDFT response matrix pair $(M_I, N_I)$ of every retained component into a common global response basis and weighting them by $w_I$; the averaged matrix pencil then yields thermal excitation energies and, via the derived transition density matrices and oscillator strengths, thermal absorption intensities.

What would settle it

Compute the TTM–TTM triplet absorption spectrum using a formally exact finite-temperature linear-response TDDFT formulation that builds the retarded response function from explicitly Boltzmann-weighted many-body states, and compare the resulting excitation energies and oscillator strengths with the Boltzmann-averaged matrix-pencil roots; if they differ beyond numerical precision, the central averaging identity is false.

Watch

Extended reading notes

Core claim

The central claim is that a finite-temperature spin-adapted description of ground and excited states follows from Boltzmann-averaging ordinary zero-temperature high-spin ROKS determinants and their spin-adapted response matrices. The paper defines the thermal ensemble density operator as a weighted sum of highest-weight $|SS\rangle$ determinants with the same spin quantum number, so the ensemble commutes with $\hat{S}^2$ and remains spin-pure. For excited states, each component supplies an ordinary spin-adapted TDDFT matrix pair $(M_I, N_I)$; these are embedded into one global spatial-orbital basis and averaged to form the thermal matrix pencil $M^\theta Z = \omega N^\theta Z$. The roots are eigenvalues of the averaged pencil, not averages of component roots. In the zero-temperature single-component limit, the method returns to standard high-spin ROKS and the corresponding spin-adapted TDDFT.

Load-bearing premise

The argument assumes that a warm molecule's excited-state behavior is exactly the sum of the behaviors of its cold high-spin components, each weighted by thermal probability, with no additional mixing between components; if that simple averaging is wrong, the predicted 425 nm explanation falls apart.

Editorial extensions

If this is right

  • If the averaging is correct, finite-temperature excitation spectra of open-shell systems can be computed at ordinary zero-temperature SA-TDDFT cost, one component at a time, without enumerating all ensemble states beyond a small active window.
  • Spin purity is preserved by construction, so thermal spectra of diradicals and other open-shell species avoid the spin contamination that plagues finite-temperature broken-symmetry approaches.
  • The theory supplies thermal transition density matrices and oscillator strengths, so temperature-dependent absorption intensities can be interpreted quantitatively through the Boltzmann population of the reference spin state.
  • For TTM–TTM specifically, the paper identifies T1→T6 at 2.983 eV (416 nm), with oscillator strength 0.00995, as the main carrier of the thermally activated 425 nm band, with T1→T3 and T1→T4 contributing in the blue-visible region.
  • The zero-temperature limit recovers ordinary ROKS and spin-adapted TDDFT, making the method a controlled extension rather than a separate formalism.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension would apply the same Boltzmann-averaged matrix-pencil construction to other open-shell chromophores with measured temperature-dependent absorption, comparing predicted thermally activated band positions and intensities with experiment.
  • If the linear-response averaging is not exact, a formal derivation starting from the finite-temperature linear-response function would clarify whether this construction is an approximation or an identity; comparing its roots against an exact thermal response calculation on a small molecule would settle the question.
  • The two-multiplet estimate of triplet population implies that the 425 nm band intensity should track the triplet population with roughly the same activation energy as the 0.194 eV singlet–triplet gap, a scaling that intensity measurements over temperature could test.
  • The active-window restriction (highest-energy closed orbital, all open orbitals, two lowest virtual orbitals) means accuracy depends on the window choice; expanding the window on small systems would probe convergence of the thermal excitation energies.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper proposes finite-temperature extensions of restricted open-shell Kohn-Sham (ROKS) theory and spin-adapted time-dependent density functional theory (TDDFT). For FT-SA-ROKS, a canonical ensemble is built from integer high-spin ROKS Slater determinants that share the same spin quantum numbers, with Boltzmann weights over component energies; the orbital gradient is the weighted sum of component gradients, and the zero-temperature single-component limit recovers ordinary high-spin ROKS. For FT-SA-TDDFT, each integer component contributes a zero-temperature spin-adapted TDDFT matrix pair (M_I, N_I), which is embedded into a common response basis and averaged with the same Boltzmann weights; the resulting matrix pencil is diagonalized to yield finite-temperature excitation energies, transition densities, and oscillator strengths. The method is applied to the diradicaloid TTM-TTM, and the calculated T1->T6 transition at 2.983 eV (about 416 nm, oscillator strength 0.00995) is used to interpret the experimentally observed temperature-activated absorption near 425 nm. The ground-state free-energy construction is clean, but the central response step is asserted rather than derived from ensemble linear-response theory, and the numerical application relies on a separately supplied singlet-triplet gap.

Significance. If the central response equations were properly derived, this would fill a genuine gap: a spin-adapted finite-temperature TDDFT for open-shell systems with a correct zero-temperature limit. The FT-SA-ROKS part of the paper is transparent and internally consistent, and the authors are to be credited for shipping the IQC package, reporting raw data, and providing explicit, falsifiable predictions (e.g., the T1->T6 transition at 2.983 eV and its oscillator strength). However, the ensemble linear-response construction is not established: Eqs. (39)-(41) average component response matrices without deriving the response of the thermal ensemble, and the eigenvalues of an averaged matrix pencil need not coincide with the poles of the exact thermal response. Since the numerical assignment of the 425 nm absorption depends directly on this step, the central claim is currently supported only conditionally. The paper is well written and the zero-temperature reduction is a strength, but the response formalism requires a derivation or an explicit validation before the method can be accepted.

major comments (4)
  1. [§2.2, Eqs. (39)-(41)] The finite-temperature response matrices are asserted to be the weighted averages of component matrices by 'the linearity of the trace (eq. 8)'. That step is not justified, because Eq. (8) concerns only the density operator Gamma_theta, whereas M_I and N_I are response matrices, not expectation values of a common operator on Gamma_theta; they depend on component Fock matrices, component densities, and the exchange-correlation kernel evaluated at each component density. The exact linear response of an incoherent ensemble is the Boltzmann-weighted sum of the component response functions, whose poles are the union of the component excitation energies, whereas the roots of Eq. (41) are neither the component roots nor necessarily the poles of the ensemble response. Please provide a derivation of Eqs. (39)-(40) from ensemble linear-response theory, or explicitly present the averaging as an approximation and benchmark it against the component-resolved response.
  2. [§3, Table 1 and Eq. (72)] The paper reports Delta_EST = 0.194 eV and uses it in the two-multiplet population model of Eq. (72), but it does not state how the singlet energy S0 was obtained. Since the FT-SA-ROKS ensemble is restricted to high-spin components with S = S_i, it cannot by itself produce an open-shell singlet. Please specify the level of theory, functional, basis, and geometry used for S0 and for the lowest triplet, and confirm that both are computed consistently; without this, the temperature dependence of the absorption interpretation is incomplete.
  3. [§2.2, Eqs. (42)-(49) and §3] The numerical results depend on the active-window composition (highest C, all O, and two lowest V orbitals) and on the metric truncation threshold lambda_cut = 10^-10 max(1, max_j |lambda_j|), but no convergence or sensitivity tests are reported. Since the finite-temperature roots in Table 1 are obtained from an averaged pencil after embedding and truncation, the sensitivity of the T1->T6 energy and oscillator strength to these choices should be quantified before the 425 nm assignment can be considered robust.
  4. [§2.1, Eq. (9)] The ensemble is restricted to components with a fixed spin quantum number S, so the theory does not describe a thermal mixture across different spin multiplets. The application compensates with the ad hoc two-multiplet model of Eq. (72), which is outside the presented formalism. This limitation should be stated explicitly in the abstract or introduction, since the current wording suggests a unified finite-temperature framework for both ground and excited states.
minor comments (4)
  1. [Table 1] The caption states that the table reports energies 'at 77 and 290 K', but only one set of energies is displayed; please report the 77 K and 290 K values separately or state explicitly that they are identical to the precision shown.
  2. [§2.2, Eqs. (61)-(63)] The notation for the transition coupling in Eq. (61) and the subsequent vectors in Eqs. (62)-(63) is not self-contained; a reference to the original definitions in SA-TDDFT would improve readability for readers not familiar with the author's prior work.
  3. [§3] The geometry was optimized at U-M06-2X/def2-SVP+D3 while the response calculations use BHHLYP/cc-pVDZ; since the singlet-triplet gap and excited-state energies can be sensitive to geometry, the choice of a fixed geometry from a different functional should be justified.
  4. [Manuscript text] There are minor typographical issues, including the stray 'P APER' at the beginning of the text and missing spacing in 'Keywords:density'; the manuscript should be copyedited.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the finite-temperature construction is definitional and the application is not fitted to the target band.

full rationale

FT-SA-ROKS and FT-SA-TDDFT are defined as Boltzmann-weighted averages of integer-component ROKS energies and SA-TDDFT matrix pairs (Eqs. 12, 39-40); the zero-temperature single-component limit is a consistency check of that definition, not a circular prediction. The TTM-TTM application does not fit the experimental 425 nm band: the reported triplet energy (0.194 eV), excitation energies (e.g., T1->T6 at 2.983 eV), and oscillator strengths (0.00995) are computed, and the Boltzmann population model (Eq. 72) uses the calculated Delta_EST rather than an adjusted value. The most serious derivation gap is the justification of Eq. 39 by 'linearity of the trace (eq. 8)': M_I and N_I are not expectation values of a fixed common operator, so the weighted average of component response matrices is an unproved approximation to thermal linear response rather than a forced consequence. This is a correctness/derivation concern, not circularity, because the averaged matrices are defined, not fitted, and the experimental assignment depends on the computed values. Self-citations (refs. 16, 18, 24, 25) provide the SA-TDDFT method and IQC code; they are prior separate method developments and do not embed the present finite-temperature result as an assumption. No circular step with quoted equation-to-equation reduction was found, so the circularity score is 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on two ad hoc modeling assumptions (Boltzmann ensemble over ROKS determinants; averaged response matrix pencil) plus domain assumptions about DFT accuracy and the prior SA-TDDFT formalism. The method introduces no fitted physics parameters, but the active window and numerical thresholds are hand-chosen.

free parameters (3)
  • Active window composition = highest-energy C orbital, all O orbitals, two lowest V orbitals
    Truncates the ROKS component ensemble in the TTM-TTM application; no convergence check with respect to window size is reported. Affects weights, energies, and spectral assignment.
  • Weight discard threshold = 10^-4
    Components with w_I <= 10^-4 are discarded and remaining weights are renormalized; a numerical approximation not tested for convergence.
  • Metric eigenvalue truncation threshold lambda_cut = 10^-10 * max(1, max|lambda_j|)
    Used to project out zero or near-zero modes of the averaged N matrix for Sf = Si - 1 sectors; a numerical choice.
assumptions (4)
  • ad hoc to paper The thermal state of the real system is represented by a canonical ensemble of integer high-spin ROKS Slater determinants with Boltzmann weights over ROKS energies E_I.
    Section 2.1, Eqs. 8 and 12-13. The weights use Kohn-Sham ROKS energies, not exact many-body energies; this is a modeling choice without proof.
  • ad hoc to paper The finite-temperature linear response of the ensemble is the weighted average of zero-temperature component SA-TDDFT matrix pencils embedded in a common orbital basis.
    Section 2.2, Eqs. 39-41. Justified by 'linearity of the trace', but response matrix pencils are not trace-linear objects; no derivation from ensemble linear-response theory is provided.
  • domain assumption The spin-adapted TDDFT matrix pair (M_I, N_I) from prior work is correct.
    Eq. 33 references prior SA-TDDFT and spin-adapted RPA formulations (refs. 16-18); the present paper inherits their validity.
  • domain assumption BHHLYP/cc-pVDZ on a fixed U-M06-2X/def2-SVP+D3 geometry gives accurate state ordering and transition energies for TTM-TTM.
    Section 3. No benchmarks against higher-level methods or experimental state energies are provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Finite-Temperature Spin-Adapted ROKS and TDDFT." pith.science (2026). https://pith.science/paper/VJ3O6WNO

@misc{pith2026260808454,
  author       = {Pith},
  title        = {Pith review of: Finite-Temperature Spin-Adapted ROKS and TDDFT},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VJ3O6WNO}},
  note         = {Machine review of arXiv:2608.08454}
}
abstract

Finite-temperature conditions constitute a central regime of interest in quantum chemistry. Nonetheless, a consistent incorporation of finite-temperature effects into density functional theory (DFT) for both ground and excited states, while rigorously preserving the spin-symmetry associated with the $\hat{S}^2$ operator, has not yet been achieved. In this work, we develop a finite-temperature extension of restricted open-shell Kohn-Sham (ROKS) theory and spin-adapted time-dependent density functional theory (TDDFT), providing unified frameworks for the description of ground and excited states, respectively. For finite-temperature ROKS, we construct a canonical ensemble by using integer high-spin ROKS components, where each component $I$ has the same spin number $|SS \rangle$ and its weight $w_I$ follows the Boltzmann distribution. For finite-temperature spin-adapted TDDFT, every integer component $I$ supplies an ordinary zero-temperature spin-adapted TDDFT matrix pair $(\Mmat_I,\Nmat_I)$. These matrices are arranged in one unified spatial-orbital order and then averaged. In the zero-temperature single-component limit, the theory reduces to ordinary high-spin ROKS and the corresponding spin-adapted TDDFT. As a numerical application, we apply the theory to a diradicaloid and use the calculated excitations to interpret the thermally activated absorption observed in variable-temperature UV/Vis spectroscopy.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 17 canonical work pages

  1. [1]

    Pribram-Jones A, Pittalis S, Gross E K U and Burke K 2014 Thermal density functional theory in contextFrontiers and Challenges in Warm Dense Mattered Graziani F, Desjarlais M P, Redmer R and Trickey S B (Cham: Springer International Publishing) pp 25–60 ISBN 978-3-319-04912-0

    work page 2014

  2. [2]

    Liu Y, Cho M and Rubenstein B 2018 Ab initio finite temperature auxiliary field quantum monte carloJournal of Chemical Theory and Computation144722–4732 pMID: 30102856 (Preprint https://doi.org/10.1021/acs.jctc.8b00569) URL https://doi.org/10.1021/acs.jctc.8b00569

    work page doi:10.1021/acs.jctc.8b00569 2018

  3. [3]

    Chai J D 2012 Density functional theory with fractional orbital occupationsThe Journal of Chemical Physics136154104 ISSN 0021-9606 URL https://doi.org/ 10.1063/1.3703894

    work page doi:10.1063/1.3703894 2012

  4. [4]

    Yeh S H, Manjanath A, Cheng Y C, Chai J D and Hsu C P 2020 Excitation energies from thermally assisted-occupation density functional theory: Theory and computational implementationThe Journal of Chemical Physics153084120 ISSN 0021-9606 Electronic Structure 13

    work page 2020

  5. [5]

    Hirata S and Jha P K 2020 Finite-temperature many-body perturbation theory in the grand canonical ensembleThe Journal of Chemical Physics153014103 ISSN 0021-9606 (Preprint https://pubs.aip.org/aip/jcp/article-pdf/doi/ 10.1063/5.0009679/15575660/014103_1_online.pdf) URL https://doi.org/ 10.1063/5.0009679

    work page doi:10.1063/5.0009679/15575660/014103_1_online.pdf 2020

  6. [6]

    Rev.140(4A) A1133–A1138 URL https://link.aps.org/ doi/10.1103/PhysRev.140.A1133

    Kohn W and Sham L J 1965 Self-consistent equations including exchange and correlation effectsPhys. Rev.140(4A) A1133–A1138 URL https://link.aps.org/ doi/10.1103/PhysRev.140.A1133

    doi:10.1103/physrev.140.a1133 1965

  7. [7]

    1142/9789812830586_0005) URL https://www.worldscientific.com/doi/abs/ 10.1142/9789812830586_0005

    CASIDA M ETime-Dependent Density Functional Response Theory for Moleculespp 155–192 (Preprint https://www.worldscientific.com/doi/pdf/10. 1142/9789812830586_0005) URL https://www.worldscientific.com/doi/abs/ 10.1142/9789812830586_0005

    doi:10.1142/9789812830586_0005

  8. [8]

    Mermin N D 1965 Thermal properties of the inhomogeneous electron gasPhys. Rev. 137(5A) A1441–A1443 URL https://link.aps.org/doi/10.1103/PhysRev.137. A1441

    doi:10.1103/physrev.137 1965

Show all 25 references
  1. [9]

    Janak J F 1978 Proof that ∂e ∂ni = ϵ in density-functional theoryPhys. Rev. B18(12) 7165–7168 URLhttps://link.aps.org/doi/10.1103/PhysRevB.18.7165

    1978 doi

  2. [10]

    Gross E K U, Oliveira L N and Kohn W 1988 Density-functional theory for ensembles of fractionally occupied states. i. basic formalismPhys. Rev. A37(8) 2809–2820 URL https://link.aps.org/doi/10.1103/PhysRevA.37.2809

    1988 doi

  3. [11]

    Pribram-Jones A, Grabowski P E and Burke K 2016 Thermal density func- tional theory: Time-dependent linear response and approximate functionals from the fluctuation-dissipation theoremPhys. Rev. Lett.116(23) 233001 URL https: //link.aps.org/doi/10.1103/PhysRevLett.116.233001

    2016 doi

  4. [12]

    5144527/13895434/244111_1_online.pdf) URL https://doi.org/10.1063/1

    Yoshikawa T, Doi T and Nakai H 2020 Finite-temperature-based time- dependent density-functional theory method for static electron correla- tion systemsThe Journal of Chemical Physics152244111 ISSN 0021- 9606 (Preprint https://pubs.aip.org/aip/jcp/article-pdf/doi/10.1063/1. 514...

    2020 doi

  5. [13]

    Niehaus T A 2026 Rigorous excited-state entropy in finite-temperature time-dependent density functional theoryThe Journal of Physical Chemistry Letters174546– 4550 pMID: 41937753 (Preprint https://doi.org/10.1021/acs.jpclett.6c00501) URLhttps://doi.org/10.1021/acs.jpclett.6c00501

    2026 doi

  6. [14]

    Zhang X 2026 A unified formulation for〈 ˆS2〉in two-component tddftJournal of Chemical Theory and Computation224429–4438 pMID: 42047467 (Preprint https://doi.org/10.1021/acs.jctc.6c00314) URL https://doi.org/10.1021/ acs.jctc.6c00314 Electronic Structure 14

    2026 doi

  7. [15]

    Roothaan C C J 1960 Self-consistent field theory for open shells of electronic sys- temsRev. Mod. Phys.32(2) 179–185 URL https://link.aps.org/doi/10.1103/ RevModPhys.32.179

    1960

  8. [16]

    Zhang X and Wang T 2026 Translating spin-adapted RPA to spin-adapted TDDFT arXiv:2605.19361v3; accepted for publication inJournal of Chemical Theory and Computation(Preprint2605.19361)

    2026 arXiv

  9. [17]

    Li Z and Liu W 2010 Spin-adapted open-shell random phase approximation and time-dependent density functional theory. i. theoryThe Journal of Chemical Physics 133064106 URLhttps://doi.org/10.1063/1.3463799

    2010 doi

  10. [18]

    Zhang X and Herbert J M 2015 Spin-flip, tensor equation-of-motion configuration interaction with a density-functional correction: A spin-complete method for exploring excited-state potential energy surfacesThe Journal of Chemical Physics143234107 URLhttps://doi.org/10.1063/1.4937571

    2015 doi

  11. [19]

    Rowe D J and Ngo-Trong C 1975 Tensor equations of motion for the excitations of rotationally invariant or charge-independent systemsRev. Mod. Phys.47(2) 471–485 URLhttps://link.aps.org/doi/10.1103/RevModPhys.47.471

    1975 doi

  12. [20]

    sciencedirect.com/science/article/pii/S0009261403004226

    Head-Gordon M 2003 Characterizing unpaired electrons from the one-particle density matrixChemical Physics Letters372508–511 ISSN 0009-2614 URL https://www. sciencedirect.com/science/article/pii/S0009261403004226

    2003

  13. [21]

    Orms N, Rehn D R, Dreuw A and Krylov A I 2018 Characterizing bonding patterns in diradicals and triradicals by density-based wave function analysis: A uniform approachJournal of Chemical Theory and Computation14638–648

    2018

  14. [22]

    Chang X, Arnold M E, Blinder R, Zolg J, Wischnat J, van Slageren J, Jelezko F, Kuehne A J C and von Delius M 2024 A stable chichibabin di- radicaloid with near-infrared emissionAngewandte Chemie International Edition 63e202404853 (Preprint https://onlinelibrary.wiley.com/doi/p...

    2024

  15. [23]

    Mesto D, Orza M, Bardi B, Punzi A, Ratera I, Veciana J, Farinola G, Painelli A, Terenziani F, Blasi D and Negri F 2025 Luminescent trityl-based diradicaloids: A theoretical and experimental assessment of charge-resonance in low-lying excited statesChemistry—A European Journal3...

    2025

  16. [24]

    Zhang X 2026 End-to-end differentiable learning of a single functional for DFT and linear-response TDDFTJournal of Chemical Theory and ComputationAccepted for publication (Preprint2602.05345)

    2026 arXiv

  17. [25]

    Xiaoyu Zhang 2026 IQC v1.0.2 Binary User Distribution https://github.com/ YuleZhang936/iqc-user/releases/tag/v1.0.2 version v1.0.2, GitHub release, commit 97b970d

    2026

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.