Pith. sign in

REVIEW 3 major objections 4 minor 40 references

Taylor-mode automatic differentiation for constructing molecular rovibrational Hamiltonian operators

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A single user-defined coordinate mapping is enough for the framework to produce Taylor-series kinetic and potential operators that yield converged vibrational energies for formaldehyde and ammonia.

desk verdict Solid, well-documented tool paper that converts clever AD machinery into a usable Python package, but it oversells 'arbitrary molecules' and needs an external energy benchmark before the convergence claims carry full weight. read the letter →

arxiv 2506.20129 v3 pith:WZJ52BTB submitted 2025-06-25 physics.atm-clus physics.chem-ph

Andrey Yachmenev , Emil Vogt , Álvaro Fernández Corral , Yahya Saleh This is my paper · ORCID
classification physics.atm-clusphysics.chem-ph
keywords automaticdifferentiationTaylorserieskineticenergyoperatorrovibrationalHamiltoniansum-of-productsrepresentationEckartframevibrationallevelslarge-amplitudemotion
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 presents an automated framework for constructing the kinetic and potential energy operators that appear in a molecule's rovibrational Hamiltonian. The framework takes a user-supplied function that maps internal coordinates (bond lengths, angles, and similar) to the Cartesian positions of the atoms, plus an optional condition for orienting the molecular frame, and returns the operators as truncated Taylor series in a sum-of-products form. That form makes subsequent quantum calculations efficient because matrix elements factor into products of one-dimensional integrals. The authors demonstrate the approach on formaldehyde and ammonia, showing that computed vibrational energy levels converge as the Taylor order and the N-mode coupling order are increased, with fourth-order coupling giving sub-wavenumber accuracy for the lowest 100 states.

What carries the argument

The load-bearing mechanism is Taylor-mode automatic differentiation: arithmetic operations, intrinsic functions, and linear algebra routines are redefined to act on truncated Taylor polynomials, so derivatives of any composition follow from the chain rule (Faà di Bruno) and the Leibniz product rule. For a molecule, the input is the map $r_i = f_i(\xi)$ from internal coordinates $\xi$ to Cartesian atomic positions, and the output is the contravariant metric tensor $G$ and pseudopotential $U$ defining the kinetic operator. Frame embeddings are handled by parametrizing the orientation rotation as $d = e^{-\kappa}$ with skew-symmetric $\kappa$; differentiating the Eckart conditions $l$ times yields a linear system for the derivative $\kappa_l$ whose matrix is the same at every order, so its inverse is computed once and reused. Multivariate partial derivatives are obtained by propagating families of univariate Taylor series and interpolating, which avoids building full high-dimensional derivative tensors. The final expansions are organized in an N-mode representation that keeps only terms coupling up to $N$ coordinates at a time.

What would settle it

Compute the ammonia inversion tunneling splitting with an eighth-order Taylor expansion of the kinetic and potential operators in the transformed coordinates of Eq. (10) and compare with a converged grid calculation; if the splitting or the lowest levels deviate beyond the sub-wavenumber convergence claimed in the paper, the single-expansion strategy for large-amplitude coordinates is falsified. A stronger test would be to run the framework on a second double-well molecule using only raw internal coordinates—no Morse or trigonometric transformations—and check whether the energy differences continue to shrink with increasing Taylor order.

Watch

Extended reading notes

Core claim

The central discovery is that Taylor-mode automatic differentiation—propagating truncated Taylor polynomials through every operation in the coordinate transformation—can replace hand derivation of kinetic energy operators. The implementation extends the Taylor-mode differentiation module of a Python automatic-differentiation library with the missing linear algebra primitives (matrix determinant, inverse, exponential, and eigendecomposition) and with routines for differentiating the Eckart-frame rotation matrix, parametrized as $d = e^{-\kappa}$ with skew-symmetric $\kappa$. Given a user-supplied internal-to-Cartesian function, the framework computes the kinetic metric tensor, the pseudopotential, and the potential energy surface at an expansion point and returns Taylor coefficients for any requested multi-indices. For ammonia, the paper shows that expanding in specially transformed coordinates—Morse functions for bond stretches and $\sin \rho$ for the inversion angle—lets a single Taylor series reproduce the double-well potential and converge much faster than expansion in raw internal coordinates. Variational calculations for both molecules show the resulting vibrational energies converging with Taylor order and with N-mode coupling order.

Load-bearing premise

The framework presumes the user can supply a smooth, invertible mapping from internal coordinates to Cartesian positions, and presumes that for large-amplitude vibrations a coordinate transformation exists that makes a low-order Taylor series faithfully represent the operators; the paper gives no general recipe for finding that transformation.

Editorial extensions

If this is right

  • For any molecule with a smooth, invertible internal-coordinate mapping, the same code produces a sum-of-products kinetic and potential operator without system-specific symbolic derivation.
  • Because matrix elements of the resulting operators factor into one-dimensional integrals, variational vibrational calculations can be driven directly by the Taylor coefficients.
  • Large-amplitude motions such as ammonia inversion can be described by a single Taylor expansion, without a grid over the large-amplitude coordinate, provided suitable transformed coordinates are chosen.
  • The Eckart frame, which is usually too algebraically complex to hand-derive, becomes routine because its rotation matrix is differentiated numerically through the exponential parametrization.
  • The same expansion machinery applies to any multivariate function, so potential energy surfaces and dipole moment surfaces can be expanded alongside the kinetic operator.

Reading between the lines

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

  • As an extension beyond the paper, the framework could automatically produce dipole moment surface expansions and therefore vibrational transition intensities, since the paper notes its derivative routine is not specific to the kinetic operator.
  • A natural stress test is to apply the framework to a molecule with two coupled large-amplitude coordinates without hand-picked transformations; the convergence rate would show how much of the ammonia success comes from the framework itself and how much from the chosen coordinates.
  • The paper's recursive Eckart differentiation sidesteps eigenvalue-based frame solvers, which the authors note fail for highly symmetric molecules such as methane; testing on such symmetric cases would clarify the practical range of the 'arbitrary molecules' claim.
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

3 major / 4 minor

Summary. The paper introduces Vibrojet, a Python/JAX framework that uses Taylor-mode automatic differentiation to construct truncated Taylor series expansions of molecular rovibrational kinetic energy operators (KEOs) and potential energy surfaces (PESs) in sum-of-products form. The KEO formalism follows standard Wilson-type treatment with vibrational, rotational, and translational coordinates, and the frame embedding (Eckart or principal axes) is incorporated into the internal-to-Cartesian mapping. The implementation extends JAX's jet module with additional linear algebra primitives and custom routines for Eckart-frame differentiation. The method is demonstrated on H2CO and NH3: one-dimensional G-matrix and PES slices are compared with exact values, and vibrational energy levels are computed with varying N-mode truncation and Taylor expansion orders to show convergence. The code is publicly available on GitHub.

Significance. If substantiated, this is a useful methodological contribution: it replaces tedious symbolic derivation and hand-coding of rovibrational operators with a general numerical pipeline, and it extends earlier Fortran-based AD work by leveraging JAX's Taylor-mode jet and the Griewank-Utke-Walther directional-derivative interpolation for efficient N-mode expansions. The direct validation of fourth-order Taylor expansions against exact G-matrix and PES slices in Figs. 2-3 is a genuine strength, as is the open-source implementation. The paper also addresses a real technical issue by providing a derivative-friendly iterative solution of the Eckart equations. However, the advertised generality to 'arbitrary molecules' rests on unstated user expertise in choosing coordinate transformations, and the vibrational-energy validation is only self-consistency between truncation orders rather than a comparison with an unexpanded Hamiltonian or reference data.

major comments (3)
  1. [§V, Eq. (10) and §VI] The central claim of an automated framework for 'arbitrary molecules, internal coordinate systems, and molecular frame embedding conditions' is not fully supported. The only nontrivial example, NH3, relies on hand-selected transformations y6^(G)=cos ρ, y6^(V)=sin ρ, and Morse functions for the stretches in Eq. (10). No algorithm, criterion, or heuristic is provided for discovering such transformations for other large-amplitude coordinates, yet the convergence properties of the Taylor expansion depend critically on these choices. The paper should either supply a systematic procedure for selecting/constructing suitable coordinates or explicitly qualify the claim to 'given a suitable user-supplied coordinate transformation.'
  2. [§V, Figs. 6-7] The vibrational-energy convergence study compares energies computed at Taylor orders D and D−2 (and N-mode orders N and N−1) within the same expanded-Hamiltonian family. This demonstrates internal consistency but not absolute accuracy: a systematically wrong but convergent Hamiltonian would not be detected. A benchmark against a calculation using the exact, unexpanded G-matrix and PES, or against established reference vibrational energies, is needed to validate both the KEO and PES expansions. This is especially important for NH3 because the PES expansion is performed around one inversion minimum in transformed coordinates, and the tunneling splitting is not checked.
  3. [§V, Figs. 4-5] The N-mode convergence analysis fixes the PES at 8th order when testing the KEO, and vice versa, but the combined truncation error is not assessed. The statement that 'N=4 is sufficient to achieve sub-wavenumber accuracy' refers only to differences between N and N−1 truncations, not to accuracy relative to the exact N=6 result. Reporting the maximum absolute deviation from the N=6 calculation (rather than pairwise differences) would make the convergence claim more quantitative and directly interpretable.
minor comments (4)
  1. [§IV] The argument order in the deriv_list example is inconsistent: the interface is defined as deriv_list(func, x0, deriv_ind, if_taylor), but the example call is deriv_list(func, deriv_ind, q0, if_taylor=True).
  2. [Fig. 7 caption] Panel (a) of Fig. 7 is labeled 'N = 4, N = 6, N = 8' but the panel shows Taylor expansion order convergence and should read 'D = 4, D = 6, D = 8' to match the text and panel (b).
  3. [§IV] There is a typo: 'For example, in can be used to expand PESs' should read 'it can be used to expand PESs.'
  4. [§II, Eq. (6)] The linear system for κ in Eq. (6) would benefit from an explicit definition of the matrix entries and the right-hand side in terms of u and λ; currently the reader must infer the arrangement from the preceding text and Eq. (5).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: Taylor-mode AD derives operator coefficients from the user-supplied coordinate map; convergence tests are truncation consistency checks, not fitted predictions.

full rationale

No circular step reduces a claimed result to an input by construction. The central derivation is Eqs. (1)-(9): given internal_to_cartesian, the G-matrix and pseudopotential are evaluated from exact derivatives of the coordinate map, and Taylor coefficients are obtained by automatic differentiation rather than by fitting. Figures 2 and 3 validate the Taylor slices against exact analytical G-matrix and PES values, which is a check of the approximation against its exact input, not a fit of the input. The transformed coordinates in Eq. (10), including Morse functions and sin/cos of the umbrella angle, are re-parametrizations chosen to improve convergence; the Morse exponent a_m = 2.0 Å^-1 is fixed, and no parameter is optimized against the computed vibrational energies. The convergence studies in Figs. 4-7 compare successive truncation orders D and D-2; this is an internal consistency check, not an external prediction, but lack of external benchmarking is not circularity. References to prior work, including the authors' own [17] for an Eckart-frame numerical solution and [38]-[40] for potentials and line lists, are transparent, and the essential equations are re-derived in the paper; none is invoked as an unverified uniqueness theorem that forces the result. The abstract's 'arbitrary molecules' wording exceeds what is demonstrated because no automatic rule is given for finding good large-amplitude coordinate transformations, and the GitHub issue on methane documents a limitation, but that is a completeness gap, not a circular argument.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The free parameters are coordinate-transformation and expansion-point choices, common in Taylor-expansion methods. The main assumptions are the standard KEO formalism, smooth invertible coordinate mappings, Taylor convergence, and the reliability of the iterative Eckart solver. These are reasonable for the demonstrated molecules but unproven for the full 'arbitrary molecules' claim.

free parameters (2)
  • Morse exponent a_m for NH3 stretching coordinates = 2.0 A^-1
    Chosen by hand in Eq. (10) to define y1..3^(V) = 1 - exp(-a_m(r_i - r_i(0))); affects Taylor convergence of the PES but is not fitted to target energies.
  • NH3 PES expansion point rho = 112.1 degrees = 112.1 degrees
    Selected as one of the two equivalent minima of the NH3 PES (Section V); the expansion point is a user choice that materially affects convergence of the potential expansion.
assumptions (4)
  • domain assumption Wilson-Sorensen KEO formalism: the rovibrational kinetic operator is exactly given by the mass-weighted contravariant metric G and pseudopotential U (Section II, Eqs. 1-5).
    The paper relies on the standard derivation of the KEO in terms of the G matrix; this is a well-established physical assumption but is not proved in the paper.
  • domain assumption The internal-to-Cartesian mapping f_i in Eq. (1) is smooth and invertible over the domain of interest, so AD can compute derivatives and define a valid metric.
    Stated in Section II: 'coordinate transformation must be invertible across the domain of the vibrational coordinates.' The entire derivative machinery depends on this.
  • standard math Taylor series of all composite functions converge over the sampled coordinate range; Faa di Bruno and Leibniz rules are valid for the differentiation (Section III).
    The paper assumes the elementary functions and linear algebra routines used in the KEO construction have convergent Taylor series; this is standard but unverified for arbitrary embeddings.
  • domain assumption The iterative Eckart solver (Eqs. 6-9) converges to the unique rotation matrix d(xi) and is differentiable to the required order, including for symmetric molecules such as methane.
    The paper states convergence in 6-10 iterations but gives no proof of convergence or differentiability for all geometries; the claim of avoiding degenerate-eigenvalue problems rests on this.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Taylor-mode automatic differentiation for constructing molecular rovibrational Hamiltonian operators." pith.science (2026). https://pith.science/paper/WZJ52BTB

@misc{pith2026250620129,
  author       = {Pith},
  title        = {Pith review of: Taylor-mode automatic differentiation for constructing molecular rovibrational Hamiltonian operators},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WZJ52BTB}},
  note         = {Machine review of arXiv:2506.20129}
}
read the original abstract

We present an automated framework for constructing Taylor series expansions of rovibrational kinetic and potential energy operators for arbitrary molecules, internal coordinate systems, and molecular frame embedding conditions. Expressing operators in a sum-of-products form allows for computationally efficient evaluations of matrix elements in product basis sets. Our approach uses automatic differentiation tools from the Python machine learning ecosystem, particularly the JAX library, to efficiently and accurately generate high-order Taylor expansions of rovibrational operators. The implementation is available at https://github.com/robochimps/vibrojet.

Figures

Figures reproduced from arXiv: 2506.20129 by the authors.

Figure 1
Figure 1. FIG. 1. Definition of the internal angular coordinates used [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. One-dimensional slices of selected elements of the [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. One-dimensional slices of the PES (in cm [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Convergence of the KEO and PES [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Convergence of the KEO and PES [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Convergence of the KEO and PES Taylor series expan [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 38 canonical work pages

  1. [1]

    Oenen, D

    K. Oenen, D. F. Dinu, and K. R. Liedl, Determining inter- nal coordinate sets for optimal representation of molecular vibration, J. Chem. Phys.160, 014104 (2024)

    work page 2024

  2. [2]

    Schneider and G

    M. Schneider and G. Rauhut, Comparison of curvilin- ear coordinates within vibrational structure calculations based on automatically generated potential energy sur- faces, J. Chem. Phys.161, 094102 (2024)

    work page 2024

  3. [3]

    Yachmenev, Computing excited states of molecules using normalizing flows, J

    Y.Saleh, Á.FernándezCorral, E.Vogt, A.Iske, J.Küpper, and A. Yachmenev, Computing excited states of molecules using normalizing flows, J. Chem. Theory Comput.21, 5221 (2025)

    work page 2025

  4. [4]

    E. Vogt, Á. F. Corral, Y. Saleh, and A. Yachmenev, Trans- ferability and interpretability of vibrational normalizing- flow coordinates (2025), arXiv:2502.15750 [physics]

    work page arXiv 2025

  5. [5]

    S. N. Yurchenko, W. Thiel, and P. Jensen, Theoretical ROVibrational Energies (TROVE): A robust numerical approach to the calculation of rovibrational energies for polyatomic molecules, J. Mol. Spectrosc.245, 126 (2007)

    work page 2007

  6. [6]

    Mátyus, G

    E. Mátyus, G. Czakó, and A. G. Császár, Toward black-box-type full- and reduced-dimensional variational (ro)vibrational computations, J. Chem. Phys.130, 134112 (2009)

    work page 2009

  7. [7]

    Wang and T

    X.-G. Wang and T. Carrington, A discrete variable repre- sentation method for studying the rovibrational quantum dynamics of molecules with more than three atoms, J. Chem. Phys. 130, 094101 (2009)

    work page 2009

  8. [8]

    Zhang, R.-S

    Q. Zhang, R.-S. Wang, and L. Wang, Neural canonical transformations for vibrational spectra of molecules, J. Chem. Phys. 161, 024103 (2024)

    2024

Show all 40 references
  1. [9]

    Avila and T

    G. Avila and T. Carrington, Solving the Schrödinger equa- tion using Smolyak interpolants, J. Chem. Phys.139, 134114 (2013)

    2013

  2. [10]

    J. C. Light and T. Carrington, Discrete-variable represen- tations and their utilization (2000)

    2000

  3. [11]

    Tennyson, M

    J. Tennyson, M. A. Kostin, P. Barletta, G. J. Harris, O. L. Polyansky, J. Ramanlal, and N. F. Zobov, DVR3D: a program suite for the calculation of rotation-vibration spectra of triatomic molecules, Computer Physics Com- munications 163, 85 (2004)

    2004

  4. [12]

    S. N. Yurchenko and T. M. Mellor, Treating linear molecules in calculations of rotation-vibration spectra, J. Chem. Phys.153, 154106 (2020)

    2020

  5. [13]

    Mladenović, Rovibrational Hamiltonians for general polyatomic molecules in spherical polar parametrization

    M. Mladenović, Rovibrational Hamiltonians for general polyatomic molecules in spherical polar parametrization. I. Orthogonal representations, J. Chem. Phys.112, 1070 (2000)

    2000

  6. [14]

    Wang and T

    X.-G. Wang and T. Carrington, A simple method for deriving kinetic energy operators, J. Chem. Phys.113, 7097 (2000)

    2000

  7. [15]

    Gatti, C

    F. Gatti, C. Muñoz, and C. Iung, A general expression of the exact kinetic energy operator in polyspherical coordi- nates, J. Chem. Phys.114, 8275 (2001)

    2001

  8. [16]

    D. W. Schwenke, New rovibrational kinetic energy op- erators using polyspherical coordinates for polyatomic molecules, J. Chem. Phys.118, 10431 (2003)

    2003

  9. [17]

    Yachmenev and S

    A. Yachmenev and S. N. Yurchenko, Automatic differenti- ation method for numerical construction of the rotational- vibrational Hamiltonian as a power series in the curvilin- ear internal coordinates using the Eckart frame, J. Chem. Phys. 143, 014105 (2015)

    2015

  10. [18]

    Peláez and H.-D

    D. Peláez and H.-D. Meyer, The multigrid POTFIT (MGPF) method: Grid representations of potentials for quantum dynamics of large systems, J. Chem. Phys.138, 014108 (2013)

    2013

  11. [19]

    Ziegler and G

    B. Ziegler and G. Rauhut, Efficient generation of sum- of-products representations of high-dimensional potential energy surfaces based on multimode expansions, J. Chem. Phys. 144, 114114 (2016)

    2016

  12. [20]

    R. I. Ovsyannikov, P. Jensen, M. Y. Tretyakov, and S. N. Yurchenko, On the use of the finite difference method in a calculation of vibration-rotation energies, Opt. Spectrosc. 107, 221–227 (2009)

    2009

  13. [21]

    Griewank, J

    A. Griewank, J. Utke, and A. Walther, Evaluating higher derivative tensors by forward propagation of univariate Taylor series, Math. Comput.69, 1117 (2000)

    2000

  14. [22]

    Walter, Structured higher-order algorithmic differ- entiation in the forward and reverse mode with ap- plication in optimum experimental design, Ph.D

    S. Walter, Structured higher-order algorithmic differ- entiation in the forward and reverse mode with ap- plication in optimum experimental design, Ph.D. the- sis, Humboldt-Universität zu Berlin, Mathematisch- Naturwissenschaftliche Fakultät II (2012)

    2012

  15. [23]

    G.O.Sørensen,Anewapproachtothehamiltonianofnon- rigid molecules, inLarge Amplitude Motion in Molecules II (1979) pp. 97–175

    1979

  16. [24]

    V. V. Rybkin, U. Ekström, and T. Helgaker, Internal- to-Cartesian back transformation of molecular geometry steps using high-order geometric derivatives, J. Comput. Chem. 34, 1842 (2013)

    2013

  17. [25]

    A. Y. Dymarsky and K. N. Kudin, Computation of the pseudorotation matrix to satisfy the Eckart axis condi- tions, J. Chem. Phys.122, 124103 (2005)

    2005

  18. [26]

    S. V. Krasnoshchekov, E. V. Isayeva, and N. F. Stepanov, Determination of the Eckart molecule-fixed frame by use of the apparatus of quaternion algebra, J. Chem. Phys. 140, 154104 (2014)

    2014

  19. [27]

    Yachmenev, Poblem with Eckart frame solution for methane using quaternion method,https://github.com/ robochimps/vibrojet/issues/4 (2025), GitHub issue #4

    A. Yachmenev, Poblem with Eckart frame solution for methane using quaternion method,https://github.com/ robochimps/vibrojet/issues/4 (2025), GitHub issue #4

    2025

  20. [28]

    Moler and C

    C. Moler and C. Van Loan, Nineteen dubious ways to compute the exponential of a matrix, twenty-five years later, SIAM Review45, 3 (2003)

    2003

  21. [29]

    Griewank and A

    A. Griewank and A. Walther,Evaluating Derivatives: Principles and Techniques of Algorithmic Differentiation, Second Edition(Society for Industrial and Applied Math- ematics, 2008)

    2008

  22. [30]

    Griewank, D

    A. Griewank, D. Juedes, and J. Utke, Algorithm 755: ADOL-C: a package for the automatic differentiation of algorithms written in C/C++, ACM Transactions on 10 Mathematical Software 22, 131 (1996)

    1996

  23. [31]

    Benet and D

    L. Benet and D. Sanders, Taylorseries.jl: Taylor expan- sions in one and several variables in Julia, Journal of Open Source Software4, 1043 (2019)

    2019

  24. [32]

    Bettencourt, M

    J. Bettencourt, M. J. Johnson, and D. Duvenaud, Taylor- mode automatic differentiation for higher-order deriva- tives in JAX, inProgram Transformations for ML Work- shop at NeurIPS 2019(2019)

    2019

  25. [33]

    Bradbury, R

    J. Bradbury, R. Frostig, P. Hawkins, M. J. Johnson, C. Leary, D. Maclaurin, G. Necula, A. Paszke, J. Van- derPlas, S. Wanderman-Milne, and Q. Zhang, JAX: com- posable transformations of Python+NumPy programs (2018)

    2018

  26. [34]

    Mach and M

    T. Mach and M. A. Freitag, Solving the parametric eigen- value problem by Taylor series and Chebyshev expansion (2023), arXiv:2302.03661 [math]

    2023 arXiv

  27. [35]

    J. M. Bowman, T. Carrington, and H.-D. Meyer, Vari- ational quantum approaches for computing vibrational energies of polyatomic molecules, Mol. Phys.106, 2145 (2008)

    2008

  28. [36]

    Hougen, P

    J. Hougen, P. Bunker, and J. Johns, The vibration- rotation problem in triatomic molecules allowing for a large-amplitude bending vibration, J. Mol. Spectrosc.34, 136 (1970)

    1970

  29. [37]

    Neff and G

    M. Neff and G. Rauhut, Towards black-box calculations of tunneling splittings obtained from vibrational structure methods based on normal coordinates, Spectrochim. Acta A Mol. Biomol. Spectrosc.119, 100 (2014)

    2014

  30. [38]

    A. F. Al-Refaie, A. Yachmenev, J. Tennyson, and S. N. Yurchenko, ExoMol line lists - VIII. A variationally com- puted line list for hot formaldehyde, Mon. Not. R. Astron. Soc. 448, 1704 (2015)

    2015

  31. [39]

    O. L. Polyansky, R. I. Ovsyannikov, A. A. Kyuberis, L. Lodi, J. Tennyson, A. Yachmenev, S. N. Yurchenko, and N. F. Zobov, Calculation of rotation-vibration energy levels of the ammonia molecule based on anab initio po- tential energy surface, J. Mol. Spectrosc.327, 21 (2016)

    2016

  32. [40]

    S. N. Yurchenko, R. J. Barber, A. Yachmenev, W. Thiel, P. Jensen, and J. Tennyson, A variationally computed T = 300K line list for NH3, J. Phys. Chem. A113, 11845 (2009)

    2009

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.