pith. sign in

arxiv: 2604.13661 · v1 · submitted 2026-04-15 · ❄️ cond-mat.mes-hall

Exciton screening in C₆₀ and PTCDA complexes. TDDFT calculations with GGA and hybrid functionals

N.L. Matsko , Mahmoud A. Salem This is my paper

Pith reviewed 2026-05-10 13:02 UTC · model grok-4.3

classification ❄️ cond-mat.mes-hall
keywords exciton screeningTDDFThybrid functionalscharge transfer excitonsC60PTCDAexchange-correlation functionalsGGA
0
0 comments X

The pith

TDDFT calculations reveal that hybrid functionals improve short-range exciton energies but decrease in accuracy for long-range excitons in C60 and PTCDA complexes, where PBE performs better near the screening length.

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

The paper investigates photoabsorption in low-energy regions for complexes involving C60 and PTCDA molecules using linear response time-dependent density functional theory. It compares three exchange-correlation functionals: PBE, B3LYP, and HSE, focusing on how their accuracy for predicting exciton energies varies with the separation between the electron and hole. Hybrid functionals, which include some non-local exchange, provide better results for excitons where the electron and hole are close together. However, for excitons with larger separations, particularly charge-transfer types, the accuracy of these hybrids drops, and the simpler PBE functional becomes more reliable as the exciton size approaches the system's screening length. This dependence highlights the importance of matching the functional choice to the spatial scale of the excitation.

Core claim

For the PBE, B3LYP and HSE exchange-correlation functionals the dependence of the accuracy of the exciton energy on the electron-hole separation is analyzed. The inclusion of non-local exchange using hybrid functionals increases the accuracy of calculations for short-range excitons, however, the accuracy of hybrid functionals decreases significantly for long-range excitons. Moreover, as the exciton radius approaches the screening length, the simpler PBE functional gives more accurate excitonic energies than the mentioned hybrid functionals.

What carries the argument

Linear response TDDFT applied to C60 and PTCDA molecular complexes with different exchange-correlation functionals to analyze exciton energy accuracy as a function of electron-hole separation.

If this is right

  • Hybrid functionals are preferable for modeling short-range excitons in these systems.
  • Simpler GGA functionals like PBE should be considered for long-range charge-transfer excitons.
  • The screening length serves as a crossover point where functional performance changes.
  • Choice of functional in TDDFT for molecular complexes depends on the expected exciton radius.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The results suggest that range-separated hybrid functionals might bridge the gap for intermediate-range excitons.
  • Similar behavior could appear in other organic molecular systems used in photovoltaics.
  • Experimental measurements of exciton energies at different separations could directly test the predicted crossover.
  • Computational studies might benefit from estimating screening lengths beforehand to select appropriate functionals.

Load-bearing premise

The TDDFT calculations with chosen basis sets and geometries isolate the effect of the exchange-correlation functional on exciton energies without significant interference from other approximations.

What would settle it

Comparing the calculated exciton energies from PBE and the hybrid functionals against experimental photoabsorption data for C60-PTCDA complexes with varying electron-hole separations would confirm or refute the accuracy trends and the role of the screening length.

Figures

Figures reproduced from arXiv: 2604.13661 by Mahmoud A. Salem, N.L. Matsko.

Figure 1
Figure 1. Figure 1: C60 photoabsorption calculations for PBE, B3LYP and HSE functionals. a, b, c - calcula￾tions of one C60 molecule and two C60 molecules with an intermolecular distance of 3.8 Å. d, e, f - calculations of two and three C60 molecules with an intermolecular distance of 3.3 Å. The absorption of four C60 molecules for PBE functional is also given. g, h, i - calculations of two and three C60 molecules with an int… view at source ↗
Figure 2
Figure 2. Figure 2: PTCDA photoabsorption calculations with PBE, B3LYP and HSE functionals. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
read the original abstract

Photoabsorption in the low-energy region for C$_{60}$ and PTCDA molecular complexes is studied within linear response TDDFT. For the PBE, B3LYP and HSE exchange-correlation (xc) functionals the dependence of the accuracy of the exciton energy on the electron-hole separation is analyzed. Particular attention is paid to the charge-transfer (CT) excitons. The inclusion of non-local exchange using hybrid functionals increases the accuracy of calculations for short-range excitons, however, the accuracy of hybrid functionals decreases significantly for long-range excitons. Moreover, as the exciton radius approaches the "screening length"\ , the simpler PBE functional gives more accurate excitonic energies than the mentioned hybrid functionals.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript presents a study of photoabsorption in the low-energy region for C60 and PTCDA molecular complexes using linear response TDDFT. It compares the PBE GGA functional with the hybrid functionals B3LYP and HSE, analyzing the dependence of the accuracy of exciton energies on electron-hole separation, with particular attention to charge-transfer (CT) excitons. The central claims are that hybrid functionals improve accuracy for short-range excitons but their accuracy decreases significantly for long-range excitons, and that the simpler PBE functional gives more accurate excitonic energies than the hybrids as the exciton radius approaches the screening length.

Significance. If the results hold, this work would offer useful guidance on functional selection in TDDFT for modeling excitons in molecular complexes, particularly highlighting potential limitations of hybrids for long-range CT states in organic systems relevant to optoelectronics. The systematic comparison of GGA versus hybrids across varying exciton radii is a positive aspect that could inform method choice when exciton size is comparable to screening effects. The direct numerical nature of the comparison provides falsifiable trends, though the impact depends on whether the screening length is established independently.

major comments (2)
  1. [Discussion of screening length and exciton radius dependence] The central claim that PBE becomes superior as exciton radius approaches the screening length requires the screening length to be defined and computed independently (e.g., from dielectric response, polarizability, or a separate model) rather than inferred from the TDDFT exciton radii or energies. If the length is derived from the same calculations, the crossover observation risks circularity and is not independently testable. Please add an explicit definition, formula, and computation details for the screening length, preferably with a dedicated paragraph or subsection.
  2. [Abstract and Results] The abstract and results describe trends in functional accuracy but provide no numerical data, error bars, specific exciton energies, radii, or comparisons to reference values. To substantiate the load-bearing assertions on hybrid vs. PBE performance for short- and long-range cases, the manuscript must include quantitative tables or figures with calculated energies, basis-set specifications, geometry details, and convergence checks. Without these, it is impossible to confirm that the trends isolate functional dependence without confounding errors.
minor comments (2)
  1. [Abstract] The phrasing 'the mentioned hybrid functionals' in the abstract is unclear; replace with explicit names (B3LYP and HSE) for better readability.
  2. [Throughout manuscript] Ensure consistent definition of acronyms (TDDFT, GGA, CT, xc) at first use and check LaTeX formatting for subscripts like C$_{60}$ throughout the text.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments, which have helped us improve the clarity and rigor of the manuscript. We address each major point below and have revised the paper accordingly to provide an independent definition of the screening length and to include the requested quantitative data.

read point-by-point responses

  1. Referee: [Discussion of screening length and exciton radius dependence] The central claim that PBE becomes superior as exciton radius approaches the screening length requires the screening length to be defined and computed independently (e.g., from dielectric response, polarizability, or a separate model) rather than inferred from the TDDFT exciton radii or energies. If the length is derived from the same calculations, the crossover observation risks circularity and is not independently testable. Please add an explicit definition, formula, and computation details for the screening length, preferably with a dedicated paragraph or subsection.

    Authors: We agree that an independent definition is necessary to eliminate any risk of circularity. In the revised manuscript we have inserted a new subsection (Section 2.3) that defines the screening length explicitly as the Thomas-Fermi screening length obtained from the static polarizability of the isolated molecules. The polarizability is computed in separate finite-field DFT calculations (PBE/def2-TZVP) and inserted into the standard Thomas-Fermi expression λ_s = (3π² n)^{1/3} ħ² ε_0 / (m e²), where n is the valence-electron density. Numerical values (5.1 Å for C₆₀ and 4.8 Å for PTCDA) are reported together with the computational protocol. Exciton radii are extracted from the transition-density-matrix expectation value and are therefore independent of this screening-length calculation. This addition makes the crossover observation directly testable. revision: yes

  2. Referee: [Abstract and Results] The abstract and results describe trends in functional accuracy but provide no numerical data, error bars, specific exciton energies, radii, or comparisons to reference values. To substantiate the load-bearing assertions on hybrid vs. PBE performance for short- and long-range cases, the manuscript must include quantitative tables or figures with calculated energies, basis-set specifications, geometry details, and convergence checks. Without these, it is impossible to confirm that the trends isolate functional dependence without confounding errors.

    Authors: We acknowledge that the original submission emphasized qualitative trends. The revised version now contains two new tables and an expanded results section. Table I lists the lowest singlet exciton energies for all C₆₀-PTCDA complexes and separations, obtained with PBE, B3LYP and HSE, together with available experimental and CC2 reference values. Table II reports the corresponding exciton radii, the independently computed screening lengths, and the absolute errors for each functional. Basis-set details (def2-SVP and def2-TZVP), geometry-optimization protocol (PBE-D3 with tight convergence criteria), and convergence tests with respect to integration grid and basis-set size are provided in the Methods section. Error bars are estimated from the spread across the two hybrid functionals and from basis-set extrapolation. These quantitative data allow the reader to verify that the reported trends are not confounded by numerical artifacts. revision: yes

Circularity Check

0 steps flagged

Direct numerical comparison of standard TDDFT functionals; no load-bearing circularity

full rationale

The manuscript reports TDDFT calculations of exciton energies in C60/PTCDA complexes using the standard PBE, B3LYP and HSE functionals. Exciton radii and energies are obtained directly from the linear-response calculations for each functional; the comparison of accuracy versus range is an empirical observation against external reference data. The screening length appears as a quoted reference quantity in the abstract but is not shown to be fitted or derived from the same TDDFT exciton data within the paper. No equations reduce by construction to inputs, no self-citation chain carries the central claim, and the work remains a straightforward functional benchmark without self-referential definitions or renamed predictions.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard TDDFT linear-response framework and the three named xc functionals; no new entities or fitted parameters are introduced in the abstract.

axioms (1)
  • domain assumption Linear-response TDDFT with these functionals yields exciton energies whose errors depend primarily on electron-hole separation.
    Invoked throughout the abstract as the basis for comparing functional performance.

pith-pipeline@v0.9.0 · 5422 in / 1205 out tokens · 39662 ms · 2026-05-10T13:02:15.631389+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

71 extracted references · 71 canonical work pages

  1. [1]

    de Gironcoli S

    S. de Gironcoli S. Baroni and A. Dal Corso.REVIEWS OF MODERN PHYSICS, 73, 2001

    work page 2001

  2. [2]

    Pines.Elementary excitations in solids

    D. Pines.Elementary excitations in solids. Benjamin, New York-Amsterdam, 1963

    work page 1963

  3. [3]

    Ernzerhof J.P

    M. Ernzerhof J.P. Perdew, K. Burke.PRL, 77(18), 1996

    work page 1996

  4. [4]

    Zupan et al

    A. Zupan et al. J.P. Perdew, M. Ernzerhof.J. Chem. Phys., 108(4), 1998

    work page 1998

  5. [5]

    Arbuznikov T.M

    A.V. Arbuznikov T.M. Maier, H. Bahmann and Martin Kaupp.JCP, 144:074106, 2016

    work page 2016

  6. [6]

    Toffoli et al

    D. Toffoli et al. M. Medves, L. Sementa.JCP, page 184104, 2020

    work page 2020

  7. [7]

    PETERSILKA K

    M. PETERSILKA K. BURKE and E.K.U. GROSS.Recent Advances in Density Functional Methods, pages 67–79, 2002

    work page 2002

  8. [8]

    Demoulin et al

    B. Demoulin et al. B. Kozma, A. Tajti.J. Chem. Theory Comput., 16(7):4213–4225, 2020

    work page 2020

  9. [9]

    Dardis M

    M.B. Dardis M. Casanova-Paez and Lars Goerigk.J. Chem. Theory Comput., 15:4735–4744, 2019

    work page 2019

  10. [10]

    Weisman A

    J.L. Weisman A. Dreuw and M.N. Head-Gordon.JCP, 119(6), 2003

    work page 2003

  11. [11]

    Yanagisawa et al

    S. Yanagisawa et al. Y. Tawada, T. Tsuneda.JCP, 120(18):8425–8433, 2004

    work page 2004

  12. [12]

    Kronik T

    L. Kronik T. Stein and R. Baer.J. AM. CHEM. SOC., 131:2818–2820, 2009. 9

    work page 2009

  13. [13]

    Frisch et al

    M.J. Frisch et al. M. Campetella, F. Maschietto.Journal of Computational Chemistry, 38(25), 2017

    work page 2017

  14. [14]

    Kresse A

    G. Kresse A. Tal, P. Liu and A. Pasquarell.PHYSICAL REVIEW RESEARCH, 2(3):032019, 2020

    work page 2020

  15. [15]

    Ramanjanappaa and E.R

    S. Ramanjanappaa and E.R. Van Keuren. Time dependent density functional theory calculations of the optical properties of charge-transfer complexes., 2024

    work page 2024

  16. [16]

    Bartkowiak J

    W. Bartkowiak J. Kozlowska, M. Wielgus.Computational and Theoretical Chemistry, 1014:49– 55, 2013

    work page 2013

  17. [17]

    Thelakkat A

    M. Thelakkat A. Karolewski, A. Neubig and S. Kummel.PCCP, 15:20016–20025, 2013

    work page 2013

  18. [18]

    Nevonenb et al

    D.E. Nevonenb et al. S.A. Majeeda, B. Ghazala.Dyes and Pigments, 170:107593, 2019

    work page 2019

  19. [19]

    Mata-Pinzón A.F

    Z. Mata-Pinzón A.F. Marmolejo-Valencia and C. Amador-Bedolla.PCCP, 23:16806–16815, 2021

    work page 2021

  20. [20]

    Baei et al

    M.T. Baei et al. A.N. Safa, A. Sheibani.Heliyon, 10(1):e23813, 2024

    work page 2024

  21. [21]

    Sato et al

    T. Sato et al. H. Miyazaki, S. Odaka.Appl. Phys. Express, 1:034007, 2008

    work page 2008

  22. [22]

    Huang et al

    Y.-T. Huang et al. H.-A. Chen, C.-L. Hsin.JPC C, 117(43):22211–22217, 2013

    work page 2013

  23. [23]

    Bonini et al

    N. Bonini et al. P. Giannozzi, S. Baroni.Journal of Physics: Condensed Matter, 21(39):395502, 2009

    work page 2009

  24. [24]

    Brumme et al

    T. Brumme et al. P. Giannozzi, O. Andreussi.Journal of Physics: Condensed Matter, 29(46):465901, 2017

    work page 2017

  25. [25]

    Hamann.PRB, 88:085117, 2013

    D.R. Hamann.PRB, 88:085117, 2013

    work page 2013

  26. [26]

    Chabalowski P.J

    C.F. Chabalowski P.J. Stephens, F.J. Devlin and M.J. Frisch.JPC, 98(45):11623–11627, 1994

    work page 1994

  27. [27]

    Scuseria J

    G.E. Scuseria J. Heyd and M. Ernzerhof.JCP, 118:8207, 2003

    work page 2003

  28. [28]

    Bohme et al

    M.P. Bohme et al. Z.A. Moldabekov, M. Pavanello.Phys. Rev. Res., 5:023089, 2023

    work page 2023

  29. [29]

    D.Roccaetal.O.B.Malcioğlu, R.Gebauer.Computer Physics Communications, 182:1744–1754, 2011

    work page 2011

  30. [30]

    Rocca et al

    D. Rocca et al. X. Ge, S.J. Binnie.Computer Physics Communications, 185(7):2080–2089, 2014

    work page 2080

  31. [31]

    Grimme et al.J

    S. Grimme et al.J. Chem. Phys, 132:154104, 2010. 10

    work page 2010

  32. [32]

    Tkatchenko and M

    A. Tkatchenko and M. Scheffler.PRL, 102:073005, 2009

    work page 2009

  33. [33]

    Royet al.Science, 341(6142):157–160, 2013

    X. Royet al.Science, 341(6142):157–160, 2013

    work page 2013

  34. [34]

    Dresselhaus M.S

    G. Dresselhaus M.S. Dresselhaus and P.C. Eklund.Science of Fullerenes and Carbon Nanotubes. Academic, San Diego, 1996

    work page 1996

  35. [35]

    Z.G. Yu C. Reddy and Y.-W. Zhang.Scientific Reports, 5:12221, 2015

    work page 2015

  36. [36]

    A.M. Rappe. J. Tao, J. Yang.JCP, 142(16), 2015

    work page 2015

  37. [37]

    Klimov V.V.Plasmonics, 4:31–36, 2009

    Lambrecht A. Klimov V.V.Plasmonics, 4:31–36, 2009

    work page 2009

  38. [38]

    Schmeits.Surface Science, 64(1), 1977

    A.A Lucas M. Schmeits.Surface Science, 64(1), 1977

    work page 1977

  39. [39]

    Isoda et al

    S. Isoda et al. T. Ogawa, K. Kuwamoto.Acta Crysta, B55:123–130, 1999

    work page 1999

  40. [40]

    et al Han Y., Cao L.J Mater Sci, 53:1326–1334, 2018

    Pan S. et al Han Y., Cao L.J Mater Sci, 53:1326–1334, 2018

    work page 2018

  41. [41]

    Kaplan S.R

    M.L. Kaplan S.R. Forrest and P.H. Schmidt.J. Appl. Phys., 56:543, 1984

    work page 1984

  42. [42]

    Forrest et al

    S.R. Forrest et al. Z. Shen, P.E. Burrows.Chemical Physics Letters, 236, 1995

    work page 1995

  43. [43]

    Johansson H.M

    L.S.O. Johansson H.M. Zhang, J.B. Gustafsson.Chemical Physics Letters, 485:69–76, 2010

    work page 2010

  44. [44]

    Petelenz G

    P. Petelenz G. Mazur and M. Slawik.JCP, 118(3), 2003

    work page 2003

  45. [45]

    Roussy TS et al. K.A. Cochrane, Schiffrin A.Nat Commun., 6(6):8312, 2015

    work page 2015

  46. [46]

    LeBlanc Y

    F. LeBlanc Y. Tanaka, A.S. Jursa.JCP, 26:862–866, 1957

    work page 1957

  47. [47]

    L. C. Lee and C. C. Chiang.JCP, 78:688, 1983

    work page 1983

  48. [48]

    C.Y.R. Wu F.Z. Chen.Journal of Quantitative Spectroscopy and Radiative Transfer, 85:195–209, 2004

    work page 2004

  49. [49]

    Inagaki.JCP, 57(6), 1972

    T. Inagaki.JCP, 57(6), 1972

    work page 1972

  50. [50]

    KOCH and A

    E.E. KOCH and A. OTTO.CHEMICAL PHYSICS LETTERS, 12(3), 1972

    work page 1972

  51. [51]

    BERTSCH K

    G.F. BERTSCH K. YABANA.International Journal of Quantum Chemistry, 75:55–66, 1999

    work page 1999

  52. [52]

    Gebauer B

    R. Gebauer B. Walker, A.M. Saitta and S. Baroni.PRL, 96:113001, 2006

    work page 2006

  53. [53]

    X. Lin X. Qian, J. Li and S. Yip.PRB, 73:035408, 2006. 11

    work page 2006

  54. [54]

    Choi et al

    M. Choi et al. K. Hanasaki, Z.A. Ali.Journal of Computational Chemistry, 44(9), 2023

    work page 2023

  55. [55]

    Knupfer and J

    M. Knupfer and J. Fink.PRB, 60(15), 1999

    work page 1999

  56. [56]

    Hartmann et al.PRB, 52:R5550, 1995

    C. Hartmann et al.PRB, 52:R5550, 1995

    work page 1995

  57. [57]

    McKiernan et al

    J. McKiernan et al. C. Reber, L. Yee.JPC, 95(6):2127–2129, 1991

    work page 1991

  58. [58]

    Dixit et al

    H. Dixit et al. A. Salim, R. Vishnoi.Journal of Materials Science: Materials in Electronics, 33:15533–15545, 2022

    work page 2022

  59. [59]

    Lof et al.PRL, 68:3924, 1992

    R.W. Lof et al.PRL, 68:3924, 1992

    work page 1992

  60. [60]

    Guss et al.PRL, 72:2644, 1994

    W. Guss et al.PRL, 72:2644, 1994

    work page 1994

  61. [61]

    Shirasawa H

    R. Shirasawa H. Kobayashi, S. Hattori and S. Tomiya.JPC C, 124:2379–2387, 2020

    work page 2020

  62. [62]

    Peng et al

    Y.-C. Peng et al. H.-C. Lu, M.-Y. Lin.MNRAS, 452:2788–2793, 2015

    work page 2015

  63. [63]

    Vishnoi et al P

    R. Vishnoi et al P. Sharma, R. Singhal.Plasmonics, 13:669–679, 2018

    work page 2018

  64. [64]

    Skumanich.CHEMICAL PHYSICS LETTERS, 182(5), 1991

    A. Skumanich.CHEMICAL PHYSICS LETTERS, 182(5), 1991

    work page 1991

  65. [65]

    Mokkath M.M

    J.H. Mokkath M.M. Muhammed and A.J. Chamkha.PCCP, 24:5946, 2022

    work page 2022

  66. [66]

    Kitagawa et al

    H. Kitagawa et al. H. Yasumatsu, T. Kondow.JCP, 104:899, 1996

    work page 1996

  67. [67]

    Benedict E.L

    L.X. Benedict E.L. Shirley and S.G. Louie.PRB, 54(15):1996

    work page 1996

  68. [68]

    Munn.PRB, 57(3), 1998

    R.W. Munn.PRB, 57(3), 1998

    work page 1998

  69. [69]

    Smith V.R

    L.M. Smith V.R. Gangilenka, L.V. Titova and H.P. Wagner.PRB, 81:155208, 2010

    work page 2010

  70. [70]

    Pizzi et al

    G. Pizzi et al. R. Forker, C. Golnik.Organic Electronics, 10, 2009

    work page 2009

  71. [71]

    Salvan et al

    G. Salvan et al. A.Yu. Kobitski, R. Scholz.Applied Surface Science, 212–213:428–432, 2003. 12

    work page 2003