REVIEW 4 major objections 5 minor 44 references
Ab Initio Complex Scaling and Similarity Renormalization Group for Continuum Properties of Nuclei
T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash
Pith's one-line read This paper claims that a phase-adjusted renormalization group after complex scaling turns nuclear resonance problems into stable bound-state calculations.
desk verdict A genuinely new CS+SRG combination with honest benchmarking; the broad-resonance numbers rest on an extrapolation the authors themselves say is not robust. 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 central object is the modified SRG generator $\eta(s)=e^{4i\theta}[T_\theta(s),H_\theta(s)]$. Complex scaling by angle $\theta$ multiplies kinetic operators by a phase; the $e^{4i\theta}$ factor compensates that phase so that the flow again suppresses off-diagonal matrix elements, which standard SRG generators stop doing once $\theta$ exceeds roughly 0.2 rad. The other load-bearing element is a momentum-space representation of the NN potential as a sum of $N_f$ functional basis functions, which keeps the complex-rotated integrals stable and pushes the validity of the ABC rotated-spectrum picture to $\theta\approx0.57$ rad in two-body tests. The solver is the translationally invariant No-Core Shell Model, a bound-state method, applied to the evolved CS-SRG Hamiltonian.
What would settle it
Take the same chiral Hamiltonian and compute the four-nucleon scattering problem with explicit continuum boundary conditions, without complex rotation; extract the S-matrix pole of the lowest 0+;0 state and compare its energy and width with the extrapolated values in Table I. A disagreement beyond the quoted uncertainties would locate the failure in the theta and Nmax extrapolation rather than in the Hamiltonian.
Extended reading notes
Core claim
The authors demonstrate that the order of transformations is the key: when the similarity renormalization group is applied after complex scaling with a standard generator, off-diagonal suppression fails beyond rotation angles around 0.2 rad, but with the phase-compensated generator $\eta(s) = e^{4i\theta}[T_\theta(s), H_\theta(s)]$ the flow again decouples low- and high-momentum scales. The resulting CS-SRG Hamiltonian is diagonalized in a translationally invariant No-Core Shell Model basis, and resonance states appear as isolated eigenvalues whose position is stable against changes in the oscillator parameter $\hbar\omega$, while continuum states slide along the rotated cut. Extrapolations in model-space size and rotation angle then give resonance centroids and widths. On $^4$He the method reproduces the narrow evaluated resonances, shifts the broad T=1 states down by 2-3 MeV, and produces only continuum eigenvalues for the tetraneutron, so no 4n resonance near 0.8 MeV is seen.
Load-bearing premise
Everything rests on the adjusted renormalization step preserving the true complex-scaled energy levels, and on the extrapolations from rotation angles no larger than about 0.3 radians and from finite basis sizes recovering the real resonance positions and widths.
Editorial extensions
If this is right
- Resonance energies and widths of light nuclei can be extracted from bound-state NCSM calculations, eliminating the need for explicit continuum basis states in this mass range.
- For helium-4, narrow resonance parameters match R-matrix evaluations, while the broad T=1 states come out 2-3 MeV lower, so the evaluated widths of those states are not reproduced by the chiral NN interactions tested.
- The 4He resonance spectrum depends only weakly on the NN parametrization, on the order of 100 keV, which points to 3N interactions as the likely source of any needed correction.
- The same method sees no tetraneutron resonance near 0.8 MeV, supporting the body of few-body calculations that rule out such a state.
- Because the solver is a bound-state code, the approach extends to A>4 systems and, with heavier ab initio methods, toward nuclei up to about A=16.
Reading between the lines
- The same phase-compensation idea may transfer to other complex-energy many-body approaches, wherever a rotation or Gamow-type basis changes the phase of the kinetic term; the $e^{4i\theta}$ factor is a template rather than a 4He-specific fix.
- The systematic 2-3 MeV downward shift of broad T=1 4He states raises a question the paper leaves open: whether the R-matrix evaluated centroids and widths for these broad states are partly artifacts of the extraction model. A dedicated R-matrix reanalysis that treats the pole positions as free would separate force error from method error.
- A natural next calculation would repeat the 4He spectrum with explicit chiral three-nucleon forces, since the paper includes only the induced 3N contribution, and watch whether the broad T=1 widths rise toward the evaluated values.
- The functional-basis decomposition used here could be adopted as a general device for evaluating complex-rotated matrix elements of non-analytic potentials, independent of the SRG piece.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript proposes a new ab initio method for nuclear continuum properties, combining complex scaling (CS) with similarity renormalization group (SRG) evolution performed after the complex rotation, and using a translationally invariant no-core shell model (NCSM) as the bound-state solver. The authors introduce a functional-basis representation of the chiral NN interaction, a modified SRG generator with an e^{4iθ} phase factor, and extrapolation procedures in Nmax and θ. They benchmark the complex-scaled spectrum on the np system up to θ = 0.57 rad, demonstrate improved Nmax convergence for the triton, and then extract resonance centroids and widths for 4He using three chiral NN interactions plus induced 3N forces, comparing with R-matrix evaluations. They find reasonable agreement for narrow T = 0 and some T = 1 centroids, but report downward shifts of broad T = 1 states and underestimated widths. As an application, they conclude that there is no tetraneutron resonance near 0.8 MeV. The central claim is that this CS-SRG-NCSM combination overcomes previous numerical limitations and makes bound-state NCSM solvers suitable for continuum studies of light nuclei.
Significance. If the extrapolation protocol were validated, this method would represent a practical advance: it avoids explicit continuum basis sets and would allow continuum properties to be studied with standard bound-state NCSM codes for A > 4 systems. The paper has genuine strengths: (i) the np benchmark validates the ABC spectrum up to 72% of the HO basis θmax; (ii) the triton convergence plots clearly show an SRG benefit over the bare CS Hamiltonian; (iii) the comparison with R-matrix data uses chiral Hamiltonians taken from the literature without tuning to 4He data, so the agreement for narrow states is a meaningful external benchmark; (iv) the tetraneutron conclusion is consistent with the majority of few-body calculations. However, the quantitative case for broad resonances and for the abstract's broad claim of 'a reliable representation of the initial Hamiltonian and its continuum properties' rests on an extrapolation in θ and Nmax that the authors themselves state is not yet robust. The paper is honest about these limitations, but the claims in the abstract and conclusion currently outrun the presented evidence.
major comments (4)
- [Table I and Fig. 4] The quantitative claim that the method yields reliable continuum properties is not supported for broad resonances. The caption of Table I states that from the fourth row the quality of extrapolation to higher θ becomes low, the text notes ε_Nmax ≳ 1 MeV and width underestimation of up to 2 MeV for T = 1 states, and the authors write that 'significantly more work is required to establish a robust extrapolation methodology for the CS angle.' Because the broad T = 1 states are exactly the cases where the finite-θ truncation is most severe, the abstract's 'reliable representation of the initial Hamiltonian and its continuum properties' should either be demonstrated against a system with known large width or be restricted to narrow resonances in the central claims.
- [Eq. (6)] The modified generator η(s) = e^{4iθ}[T_θ(s), H_θ(s)] is central to the method, but no derivation of the phase factor is given and no formal argument shows that the resulting flow is a spectrum-preserving similarity transformation. The only evidence is the observation that the standard generator fails beyond 0.2 rad and that 'other impediments appear beyond 0.3 rad.' Since the practical bound θ ≤ 0.3 rad is the direct cause of the broad-width underestimation, the paper should provide a proof or a dedicated numerical demonstration that the modified generator decouples the Hamiltonian without changing its eigenvalue spectrum, together with a detailed account of what limits θ to 0.3 rad.
- [Extrapolation forms in Table I] The functional forms E(Nmax) = E∞ + a exp(−b Nmax) and d/dθ Re(E(θ)) = −2 ΔE_thres sin(2θ) exp(−(θ/a)^b) are introduced without derivation and are fitted to the authors' own finite-basis, finite-θ results; no parameter values or goodness-of-fit statistics are reported. The statement that 'extrapolated results agree with exact calculations already published in the literature' therefore cannot be independently checked. A validation of these forms against known resonance parameters, or against an exactly solvable model with CS-SRG calculations at θ values beyond the current practical limit, is needed before the extrapolated widths in Table I can be used as quantitative results.
- [Tetraneutron section, Fig. 5] The conclusion that there is no tetraneutron resonance near 0.8 MeV is presented as a definitive result, but it rests on a single chiral Hamiltonian (I-N3LO) at θ = 0.3 rad and Nmax = 18. Because the paper has already established that moderate and broad widths can be underestimated at θ = 0.3 rad, the authors should either add a θ- and Nmax-convergence study for the 4n continuum around the hypothesized energy or soften the claim to 'no evidence in this calculation' rather than a definitive exclusion.
minor comments (5)
- [Eq. (2) and following text] The phrase 'c-normalization' is used without definition; please spell out the normalization convention for the complex-scaled wave functions.
- [Fig. 2 caption] The sentence 'For technical, the chiral 3NF is omitted' is ungrammatical; it should read 'For technical reasons, the chiral 3NF is omitted' or similar.
- [Reference [13]] The URL in reference [13] is malformed, containing two concatenated DOIs; it should be corrected to the single Phys. Rev. Lett. DOI.
- [Eq. (4)] The notation V^{ll'S12}_{NN}(k,k') is not defined; clarify whether 'S12' denotes the total spin or a tensor operator and define all indices.
- [Fig. 4 and Table I] The text cites both the TUNL Nuclear Data Project [17] and the older evaluation [16] for the R-matrix data; please clarify which evaluation is used for the comparison in Fig. 4 and Table I.
Circularity Check
No circularity: the central CS-SRG derivation is self-contained, and the acknowledged extrapolation limitations are validation concerns, not input-output reductions.
full rationale
After walking the derivation chain, I find no circular step that reduces the paper's central claim to its own inputs. The chiral EFT Hamiltonians (I-N3LO, N2LOsat, I-N4LO) are external inputs whose low-energy constants are constrained by nucleon-nucleon data, not by the 4He resonance observables; the comparisons to R-matrix evaluations [16] and to exact few-body calculations [18] are external benchmarks, not fits. The CS-SRG construction (Eqs. 1-6) is a stated transformation whose spectrum-preserving property relies on the ABC theorem [8,9], an external mathematical result, and the modified generator Eq. 6 is an explicitly introduced ansatz rather than a quantity defined in terms of the target resonance parameters. The finite-basis and finite-CS-angle extrapolations in Table I and its footnote are fits to the authors' own computed eigenvalues, so the extrapolated E_inf and Gamma_inf carry model uncertainty; the paper itself flags this clearly ('Significantly more work is required to establish a robust extrapolation methodology for the CS angle,' and 'the quality of extrapolation to higher theta becomes low'). That is a correctness/validation limitation, not circularity: the extrapolands are not experimental resonance parameters, and no evaluated R-matrix value is used to select the functional forms. Self-citations [3,13,14] are method comparisons or sources of the input Hamiltonian, and none carries the weight of the continuum claims. The tetraneutron conclusion likewise follows from the same independently benchmarked machinery and agrees with a body of external few-body calculations rather than being defined by them.
Assumptions & free parameters
free parameters (5)
- Functional-basis coefficients (a_i, b_i, c_i, b'_i, c'_i) in Eq. (4) =
Not reported; Nf=6 gives ~2% deuteron binding error, Nf=12 below keV
- Number of basis functions Nf =
12 for production results
- Complex-scaling angle theta =
0.3 rad (0.36 rad for some Table I extrapolations)
- SRG resolution scale lambda =
2.0-2.1 fm^-1
- Extrapolation parameters a and b in E(Nmax) and theta-extrapolation forms =
Not reported per state
assumptions (5)
- standard math ABC theorem for complex scaling applies to the finite-basis many-body Hamiltonians used here (single-channel result extended to multi-channel).
- ad hoc to paper The modified SRG generator eta=e^{4i theta}[T_theta,H_theta] defines a spectrum-preserving similarity transformation and decouples high and low momenta for theta<=0.3 rad.
- domain assumption Chiral NN interactions from Refs. [12,14,15] (with SRG-induced 3N, but without chiral 3NF) are adequate for 4He resonance properties.
- domain assumption The functional-basis fit of the NN potential on the real momentum axis remains accurate after analytic continuation to complex momenta.
- ad hoc to paper The functional forms used for Nmax and theta extrapolation in Table I are valid for the studied resonances.
Cite this review
Pith. "Pith review of Ab Initio Complex Scaling and Similarity Renormalization Group for Continuum Properties of Nuclei." pith.science (2026). https://pith.science/paper/EWCVIBR3
@misc{pith2026250701595,
author = {Pith},
title = {Pith review of: Ab Initio Complex Scaling and Similarity Renormalization Group for Continuum Properties of Nuclei},
year = {2026},
howpublished = {\url{https://pith.science/paper/EWCVIBR3}},
note = {Machine review of arXiv:2507.01595}
}
abstract
We introduce a novel \abinitio many-body method designed to compute the properties of nuclei in the continuum. This approach combines well-established techniques, namely the Complex Scaling (CS) and Similarity Renormalization Group (SRG) methods while employing the translationally invariant No-Core Shell Model (NCSM) as a few-body solver. We demonstrate that this combination effectively overcomes numerical limitations previously encountered in exploring continuum properties of light nuclei with standard many-body techniques, and at the same time makes less imperative the need for a continuous set of basis states for the continuum. To benchmark the method for applications in the many-body sector, we apply it to the \textsuperscript{4}He system, where semi-exact calculations within a finite basis are feasible. Our extrapolated results agree with exact calculations already published in the literature. We argue that different NN parametrizations of chiral EFT Hamiltonians will not permit to reproduce evaluated resonance properties of \textsuperscript{4}He. As an application, we showcase the case of the tetraneutron. This work enables the application of the method to $A>4$-mass systems, providing a reliable representation of the initial Hamiltonian and its continuum properties.
Figures
Reference graph
Works this paper leans on
-
[1]
J. Li, Y. Ma, N. Michel, B. Hu, Z. Sun, W. Zuo, and F. Xu, Physics 3, 977 (2021), ISSN 2624-8174, URL https://www.mdpi.com/2624-8174/3/4/62
work page 2021
- [2]
-
[3]
P. Navr´ atil, S. Quaglioni, G. Hupin, C. Romero- Redondo, and A. Calci, Physica Scripta 91, 053002 (2016), URL https://dx.doi.org/10.1088/0031-8949/ 91/5/053002
-
[5]
A. Idini, C. Barbieri, and P. Navr´ atil, Physical Review Letters 123, 92501 (2019), ISSN 1079-7114, URL https: //doi.org/10.1103/PhysRevLett.123.092501
-
[6]
B. G. Giraud and K. Kato, Annals of Physics 308, 115 (2003), ISSN 00034916
work page 2003
-
[7]
S. Aoyama, T. Myo, K. Kat¯ o, and K. Ikeda, Progress of Theoretical Physics 116, 1 (2006), ISSN 0033-068X, https://academic.oup.com/ptp/article- pdf/116/1/1/19571980/116-1.pdf, URL https://doi. org/10.1143/PTP.116.1
-
[8]
J. Aguilar and J. M. Combes, Communications in Math- ematical Physics 22, 269 (1971), ISSN 0010-3616, URL http://link.springer.com/10.1007/BF01877510
-
[9]
E. Balslev and J. M. Combes, Communications in Math- ematical Physics 22, 280 (1971), ISSN 0010-3616, URL http://link.springer.com/10.1007/BF01877511
Show all 44 references
-
[10]
Papadimitriou and J
G. Papadimitriou and J. Vary, Physics Let- ters B 746, 121 (2015), ISSN 0370-2693, URL https://www.sciencedirect.com/science/article/ pii/S037026931500324X
2015
-
[11]
Papadimitriou and J
G. Papadimitriou and J. P. Vary, Phys. Rev. C 91, 021001 (2015), URL https://link.aps.org/doi/10. 1103/PhysRevC.91.021001
2015
-
[12]
D. R. Entem and R. Machleidt, Physical Review C 68, 041001 (2003), ISSN 0556-2813, URL http://link.aps. org/doi/10.1103/PhysRevC.68.041001
2003 doi
-
[13]
E. D. Jurgenson, P. Navr´ atil, and R. J. Furnstahl, Physical Review Letters 103, 082501 (2009), ISSN 0031-9007, URL https://link.aps.org/doi/10.1103/ PhysRevC.83.034301https://link.aps.org/doi/10. 1103/PhysRevLett.103.082501
2009
-
[14]
Ekstr¨ om, G
A. Ekstr¨ om, G. R. Jansen, K. A. Wendt, G. Hagen, T. Papenbrock, B. D. Carlsson, C. Forss´ en, M. Hjorth- Jensen, P. Navr´ atil, and W. Nazarewicz, Phys. Rev. C 91, 051301 (2015), URL https://link.aps.org/doi/ 10.1103/PhysRevC.91.051301
2015 doi
-
[15]
D. R. Entem, R. Machleidt, and Y. Nosyk, Physical Re- view C 96, 024004 (2017), ISSN 2469-9985
2017
-
[16]
Tilley, H
D. Tilley, H. Weller, and G. Hale, Nuclear Physics A541, 1 (1992), ISSN 03759474, URL https://linkinghub. elsevier.com/retrieve/pii/037594749290635W
1992
-
[17]
T. N. D. Group, Tunl nuclear data evaluation project (2025), URL https://nucldata.tunl.duke. edu/nucldata/,note={Accessed:19-Feb-2025}
2025
-
[18]
Lazauskas, J
R. Lazauskas, J. Carbonell, and E. Hiyama, Progress of Theoretical and Experimental Physics 2017, 073D03 (2017), ISSN 2050- 3911, https://academic.oup.com/ptep/article- pdf/2017/7/073D03/19368904/ptx078.pdf, URL https://doi.org/10.1093/ptep/ptx078
2017 doi
-
[19]
Michel, W
N. Michel, W. Nazarewicz, and M. P loszajczak, Phys. Rev. Lett. 131, 242502 (2023), URL https://link.aps. org/doi/10.1103/PhysRevLett.131.242502
2023 doi
-
[20]
Bacca, N
S. Bacca, N. Barnea, W. Leidemann, and G. Orlandini, Physical Review Letters 110, 042503 (2013), ISSN 0031- 9007
2013
-
[21]
Kegel, P
S. Kegel, P. Achenbach, S. Bacca, N. Barnea, J. Beriˇ ciˇ c, D. Bosnar, L. Correa, M. Distler, A. Esser, H. Fonvieille, et al., Physical Review Letters 130, 152502 (2023), ISSN 0031-9007
2023
-
[22]
F. M. Marqu´ es, M. Labiche, N. A. Orr, J. C. Ang´ elique, L. Axelsson, B. Benoit, U. C. Bergmann, M. J. G. Borge, W. N. Catford, S. P. G. Chappell, et al., Phys. Rev. C 65, 044006 (2002), URL https://link.aps.org/doi/ 10.1103/PhysRevC.65.044006
2002 doi
-
[23]
S. A. Sofianos, S. A. Rakityansky, and G. P. Vermaak, Journal of Physics G: Nuclear and Particle Physics 23, 1619 (1997), URL https://dx.doi.org/10.1088/ 0954-3899/23/11/010
1997
-
[24]
N. K. Timofeyuk, Journal of Physics G: Nuclear and Par- ticle Physics 29, L9 (2003), URL https://dx.doi.org/ 10.1088/0954-3899/29/2/102
2003 doi
-
[25]
S. C. Pieper, Phys. Rev. Lett. 90, 252501 (2003), URL https://link.aps.org/doi/10.1103/PhysRevLett.90. 252501
2003 doi
-
[26]
Kisamori, S
K. Kisamori, S. Shimoura, H. Miya, S. Michi- masa, S. Ota, M. Assie, H. Baba, T. Baba, D. Beaumel, M. Dozono, et al., Phys. Rev. Lett. 116, 052501 (2016), URL https://link.aps.org/doi/10. 1103/PhysRevLett.116.052501
2016
-
[27]
M. Duer, T. Aumann, R. Gernh¨ auser, V. Panin, S. Paschalis, D. M. Rossi, N. L. Achouri, D. Ahn, H. Baba, C. A. Bertulani, et al., Nature 606, 678–682 (2022)
2022
-
[28]
Faestermann, A
T. Faestermann, A. Bergmaier, R. Gernh¨ auser, D. Koll, and M. Mahgoub, Physics Letters B 824, 136799 (2022), ISSN 0370-2693, URL https://www.sciencedirect. com/science/article/pii/S0370269321007395
2022
-
[29]
Faestermann, New results on the tetraneutron, seen in context (2022), 2207.10542, URL https://arxiv.org/ abs/2207.10542
T. Faestermann, New results on the tetraneutron, seen in context (2022), 2207.10542, URL https://arxiv.org/ abs/2207.10542
2022 arXiv
-
[30]
A. M. Shirokov, G. Papadimitriou, A. I. Mazur, I. A. Mazur, R. Roth, and J. P. Vary, Phys. Rev. Lett. 117, 182502 (2016), URL https://link.aps.org/doi/ 10.1103/PhysRevLett.117.182502
2016 doi
-
[31]
J. G. Li, N. Michel, B. S. Hu, W. Zuo, and F. R. Xu, Phys. Rev. C 100, 054313 (2019), URL https://link. aps.org/doi/10.1103/PhysRevC.100.054313
2019 doi
-
[32]
Gandolfi, H.-W
S. Gandolfi, H.-W. Hammer, P. Klos, J. E. Lynn, and A. Schwenk, Phys. Rev. Lett. 118, 232501 (2017), URL https://link.aps.org/doi/10.1103/PhysRevLett. 118.232501
2017 doi
-
[33]
L. V. Grigorenko, N. K. Timofeyuk, and M. V. Zhukov, The European Physical Journal A 19, 187–201 (2004)
2004
-
[34]
Lazauskas and J
R. Lazauskas and J. Carbonell, Phys. Rev. C 72, 034003 (2005), URL https://link.aps.org/doi/10. 1103/PhysRevC.72.034003. 7
2005
-
[35]
Lazauskas and J
R. Lazauskas and J. Carbonell, Phys. Rev. C 71, 044004 (2005), URL https://link.aps.org/doi/10. 1103/PhysRevC.71.044004
2005
-
[37]
Carbonell, R
J. Carbonell, R. Lazauskas, E. Hiyama, and M. Kamimura, Few-Body Systems 58 (2017)
2017
-
[38]
Deltuva, Physics Letters B 782, 238 (2018), ISSN 0370-2693, URL https://www.sciencedirect
A. Deltuva, Physics Letters B 782, 238 (2018), ISSN 0370-2693, URL https://www.sciencedirect. com/science/article/pii/S0370269318304052
2018
-
[39]
Fossez, J
K. Fossez, J. Rotureau, N. Michel, and M. Ploszajczak, Phys. Rev. Lett. 119, 032501 (2017), URL https:// link.aps.org/doi/10.1103/PhysRevLett.119.032501
2017 doi
-
[40]
Deltuva, Phys
A. Deltuva, Phys. Rev. C 97, 034001 (2018), URL https: //link.aps.org/doi/10.1103/PhysRevC.97.034001
2018 doi
-
[41]
Deltuva and R
A. Deltuva and R. Lazauskas, Phys. Rev. Lett. 123, 069201 (2019), URL https://link.aps.org/doi/10. 1103/PhysRevLett.123.069201
2019
-
[42]
Deltuva and R
A. Deltuva and R. Lazauskas, Phys. Rev. C 100, 044002 (2019), URL https://link.aps.org/doi/10. 1103/PhysRevC.100.044002
2019
-
[43]
Ishikawa, Phys
S. Ishikawa, Phys. Rev. C 102, 034002 (2020), URL https://link.aps.org/doi/10.1103/PhysRevC. 102.034002
2020 doi
-
[44]
M. D. Higgins, C. H. Greene, A. Kievsky, and M. Viviani, Phys. Rev. Lett. 125, 052501 (2020), URL https:// link.aps.org/doi/10.1103/PhysRevLett.125.052501
2020 doi
-
[45]
M. D. Higgins, C. H. Greene, A. Kievsky, and M. Viviani, Phys. Rev. C 103, 024004 (2021), URL https://link. aps.org/doi/10.1103/PhysRevC.103.024004
2021 doi
-
[46]
Lazauskas, E
R. Lazauskas, E. Hiyama, and J. Carbonell, Phys. Rev. Lett. 130, 102501 (2023), URL https://link.aps.org/ doi/10.1103/PhysRevLett.130.102501
2023 doi
Reviewed August 6, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.