REVIEW 3 major objections 5 minor 2 references
Valence photoelectron spectra of thiouracils in the gas phase
T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash
Pith's one-line read Valence photoelectron spectra of all three thiouracils show a first ionization potential of 8.4-8.7 eV, about 1 eV below uracil, with bands up to 12 eV reproduced by EOM-IP-CCSD calculations.
desk verdict New reference spectra for two thiouracils with a sound S/O assignment, but the abstract overstates theory–experiment agreement up to 12 eV. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The argument is carried by pairing EOM-IP-CCSD (equation-of-motion coupled cluster for ionization potentials) with the time-independent double-harmonic adiabatic Hessian approach. EOM-IP-CCSD computes the ionization energies and cationic states directly, while the adiabatic Hessian approach computes Franck-Condon factors from the harmonic normal modes of the neutral and cation states, including Duschinsky rotations and temperature effects. The resulting stick spectra are broadened and compared with the experimental spectra, and Dyson orbital norms estimate relative band intensities. Normal-mode analysis then identifies the specific vibrations that generate the long progressions, such as mode 26, an NH-bending plus C=O-stretching plus C-C-stretching mode near 1800-2100 cm-1, which dominates the oxygen-ionization band in 2- and 4-thiouracil.
What would settle it
Measure the high-resolution valence photoelectron spectrum of 2,4-dithiouracil in the 10-11 eV binding-energy range: the paper's assignment predicts no oxygen-localized band there, so seeing a structured band would contradict the central orbital assignment. Alternatively, comparing the measured vibrational spacing in 4-thiouracil's second band with the calculated ~1836 cm-1 mode would test whether the harmonic model holds.
Extended reading notes
Core claim
The paper's central claim is that every thiouracil ionizes at 8.4-8.7 eV—measured as 8.74 ± 0.03 eV for 2-thiouracil, 8.42 ± 0.01 eV for 4-thiouracil, and 8.45 ± 0.01 eV for 2,4-dithiouracil—roughly 1 eV below uracil, and that this lowering is a general consequence of thionation. The spectra of 4-thiouracil and 2,4-dithiouracil are reported for the first time. EOM-IP-CCSD/cc-pVTZ calculations reproduce the vibrational progressions of the first two bands and show that the lowest band comes from ionizing sulfur-dominated orbitals, while the band near 10-11 eV in the two singly thionated molecules comes from oxygen-dominated orbitals; 2,4-dithiouracil lacks this band because it has no oxygen. The same calculations show that Koopmans' orbital-energy estimates overshoot the ionization energies by up to about 2.3 eV and that the Dyson orbitals for the low-lying transitions are close to canonical Hartree-Fock orbitals, with norms above 0.8. For binding energies above about 12 eV, the paper argues that nonadiabatic couplings between cationic states must be included, since geometry optimizations of higher states repeatedly land in crossing regions.
Load-bearing premise
The calculations assume that the low-lying cation states responsible for the bands below 12 eV are each a single harmonic well, well separated from other electronic states; if that assumption is false, the calculated vibrational progressions and band assignments lose their footing.
Editorial extensions
If this is right
- The first ionization potentials of all three thiouracils fall between 8.4 and 8.7 eV, so thionation lowers uracil's ionization energy by roughly 1 eV regardless of which oxygen is replaced.
- The structured band near 10-11 eV in 2- and 4-thiouracil comes from ionizing an oxygen-localized orbital, and 2,4-dithiouracil shows no such band, confirming the assignment.
- For binding energies up to about 12 eV, EOM-IP-CCSD with the double-harmonic adiabatic Hessian approximation reproduces the measured vibrational progressions, enabling mode assignments such as the C=O-stretch-dominated progression in the oxygen band.
- Simple Koopmans' theorem estimates overestimate the measured ionization energies by up to about 2.3 eV, so a correlated treatment is necessary even for assigning the lowest bands.
- Above about 12 eV, the single-state harmonic approximation breaks down because cationic states cross, so interpreting those bands requires a nonadiabatic or multistate treatment.
Reading between the lines
- The measured ~1 eV drop in ionization energy, which the paper does not discuss in redox terms, implies thiouracils are markedly easier to oxidize than uracil, a property that could matter in charge-transfer contexts.
- The oxygen-localized band near 10-11 eV could serve as a spectroscopic fingerprint for distinguishing singly from doubly thionated uracils, an application the paper does not propose.
- A natural next step, not taken here, would be to build a two-state vibronic-coupling model for the crossing regions and test whether it reproduces the broad 12-16 eV envelope in the experimental spectra.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports gas-phase valence photoelectron spectra of 2-thiouracil, 4-thiouracil, and 2,4-dithiouracil measured at 100 eV photon energy, including what the authors state are the first spectra of 4-TU and 2,4-dTU. The experimental bands are assigned with EOM-IP-CCSD vertical ionization energies, Dyson orbitals, and time-independent double-harmonic adiabatic Hessian vibronic simulations. The central claims are that the first ionization potentials lie near 8.4-8.7 eV, about 1 eV below uracil; that the bands up to 12 eV show strong vibrational progressions and are well reproduced by the calculations; and that these bands arise from sulfur- and oxygen-localized orbitals, with nonadiabatic couplings important at higher binding energies.
Significance. If the results stand, the paper provides valuable high-resolution reference spectra for three thionated nucleobases, with consistent experimental-theoretical orbital assignments across the series. The experimental first IPs have small uncertainties, the S/O assignment is corroborated by the systematic absence of the oxygen band in 2,4-dTU, and the data are openly deposited. The EOM-IP-CCSD calculations and Dyson orbital analysis are a strength, as is the explicit identification of where the double-harmonic adiabatic Hessian approximation breaks down. The main weakness is that the abstract overstates the level of agreement with theory, and one numerical claim in the abstract is not consistent with the paper's own Table 1.
major comments (3)
- [Abstract vs. Table 1] The abstract states that the first ionization potential is between 8.4 and 8.7 eV for all three thiouracils, but Table 1 reports 8.74 ± 0.03 eV for 2-TU. This is outside the stated range. Please revise the range or the wording (e.g., '8.4-8.8 eV' or 'around 8.4-8.7 eV') so the abstract matches the reported data.
- [Abstract vs. §Vibronic transitions and Fig. 2] The abstract's claim that 'ionization bands up to 12 eV show strong vibrational progressions and are well reproduced by the calculations' is internally inconsistent with the manuscript's own results. In the 'Vibronic transitions' section the authors state that the D3 intensity for 2,4-dTU is largely overestimated compared with experiment, and D3 has a vertical IP of 9.137 eV (Table 4), well inside the 12 eV range. The D7 intensity is also described as strongly overestimated. The abstract should be qualified, for example by restricting the 'well reproduced' statement to the first two bands of 2-TU and 4-TU, and by noting the overestimation in 2,4-dTU.
- [Computational Details and Fig. 1 caption] The reported agreement is achieved after applying per-molecule energy shifts of 0.05, 0.10, -0.02, and -0.07 eV to the theoretical spectra (Fig. 1 caption), in addition to a hand-chosen Lorentzian width of 0.02 eV. Because the 'well reproduced up to 12 eV' claim depends on these alignment choices, the abstract and the results discussion should explicitly disclose that the theoretical spectra were shifted to match experiment. Without this disclosure, the claim overstates the parameter-free character of the comparison.
minor comments (5)
- [Methods, Lorentzian broadening equation] In the equation for I(E), the linewidth parameter appears to be missing in the rendered text; it should read γ = 0.02 eV rather than ' = 0.02 eV'.
- [§Vibronic transitions] The sentence 'This was also observed in the D2 optimizations for 2,4-TU' should read '2,4-dTU' for consistency with the rest of the manuscript.
- [§Vibronic transitions] The phrase 'the undertaken geometry optimizations for higher energy cationic excited states usually hit intersection regions' is awkward; consider rephrasing to 'the geometry optimizations of higher cationic states usually reached intersection regions'.
- [Throughout] The manuscript alternates between '2,4-TU' and '2,4-dTU'; please standardize on '2,4-dTU'.
- [Conclusion] The conclusion says the approach 'fails to reproduce the bands for higher cationic states'; this is appropriately hedged, but the abstract should be brought in line with this statement.
Circularity Check
No significant circularity: the theoretical spectra are ab initio, the experimental spectra are new measurements, and the disclosed rigid energy shifts do not force the vibrational or relative-band content.
full rationale
The paper's central claims—first ionization potentials of 8.4–8.7 eV for the three thiouracils, vibrational progressions up to 12 eV, sulfur/oxygen orbital assignments, and first gas-phase spectra for 4-TU and 2,4-dTU—rest on new synchrotron measurements and EOM-IP-CCSD/adiabatic-Hessian calculations, not on the fitted inputs. The only adjustable parameters are per-molecule rigid energy shifts (0.05, 0.1, –0.02, –0.07 eV) and a Lorentzian width (0.02 eV). The paper explicitly states that the shifts were applied 'to match the experimental spectra'; a rigid shift cannot determine relative band spacings, Franck–Condon intensities, or the vibrational mode assignments, which are the substantive theoretical content. The paper also reports a genuine failure within the claimed 12 eV range: 'The D3 intensity for the 2,4-dTU is largely overestimated compared to the experimental spectrum.' This shows that the agreement was not forced by construction. Self-citations (refs. 13–17) provide prior experimental context and are not load-bearing for the derivation. No equation is defined in terms of the target result, and no prediction reduces to a fitted parameter. Therefore no circular step is exhibited.
Assumptions & free parameters
free parameters (4)
- Theoretical energy shifts =
Ura +0.05, 2-TU +0.10, 4-TU -0.02, 2,4-dTU -0.07 eV
- Lorentzian broadening width =
0.02 eV
- Vibrational quantum cutoffs =
Up to 2 quanta in S0, up to 5 in Di
- Boltzmann temperature =
393.15 K (120 C)
assumptions (5)
- domain assumption EOM-IP-CCSD with the cc-pVTZ basis provides reliable ionization energies and Dyson orbitals for the low-lying states of thiouracils.
- domain assumption The double-harmonic adiabatic Hessian approximation describes D0-D3 vibronic spectra well enough to assign the observed progressions.
- domain assumption The heated gas-phase sample consists of intact neutral thiouracils in their thione tautomeric form.
- domain assumption Born-Oppenheimer adiabatic separation holds for the low-lying cation states considered; nonadiabatic couplings only affect higher states.
- domain assumption The electron analyzer transmission is flat over the measured 80 to 95 eV kinetic energy window.
Cite this review
Pith. "Pith review of Valence photoelectron spectra of thiouracils in the gas phase." pith.science (2026). https://pith.science/paper/DNETU7UE
@misc{pith2026250507361,
author = {Pith},
title = {Pith review of: Valence photoelectron spectra of thiouracils in the gas phase},
year = {2026},
howpublished = {\url{https://pith.science/paper/DNETU7UE}},
note = {Machine review of arXiv:2505.07361}
}
read the original abstract
We present a combined experimental and theoretical study of the vibrationally resolved valence photoelectron spectra of the complete series of thiouracils (2-thiouracil, 4-thiouracil and 2,4-dithiouracil) for binding energies between 8 and 17 eV. The theoretical spectra were calculated using equation-of-motion coupled cluster theory for ionization potential (EOM-IP-CCSD) combined with the time-independent double-harmonic adiabatic Hessian approach. For all three thiouracils, the first ionization potential is found between 8.4 and 8.7 eV, which is 1 eV lower than for the canonical nucleobase uracil. Ionization bands up to 12 eV show strong vibrational progressions and are well reproduced by the calculations. These bands are attributed to the ionization of (primarily) sulfur- and oxygen-localized valence molecular orbitals. For higher binding energies, the calculations indicate that nonadiabatic couplings are important for the interpretation of the photoelectron spectra.
Figures
Reference graph
Works this paper leans on
-
[1]
1 S. Arslancan, L. Martínez-Fernández, and I. Corral, “Photophysics and Photochemistry of Canonical Nucleobases’ Thioanalogs: From Quantum Mechanical Studies to Time Resolved Experiments,” Molecules 22(6), 998 (2017). 2 B. Ashwood, M. Pollum, and C.E. Crespo-Hernández, “Photochemical and Photodynamical Properties of Sulfur-Substituted Nucleic Acid Bases,”...
work page 2017
-
[3]
A study of the valence photoelectron spectrum of uracil and mixed water–uracil clusters,
Ultraviolet photoelectron spectra of 2- thiouracil and its methyl derivatives,” Journal of the Chemical Society, Perkin Transactions 2 (6), 871–876 (1990). 36 G. Mattioli, L. Avaldi, P. Bolognesi, A. Casavola, F. Morini, T. Van Caekenberghe, J.D. Bozek, M.C. Castrovilli, J. Chiarinelli, A. Domaracka, S. Indrajith, S. Maclot, A.R. Milosavljević, C. Nicolaf...
1990
Reviewed August 15, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.