pith. sign in

arxiv: 2604.03019 · v1 · submitted 2026-04-03 · ❄️ cond-mat.mtrl-sci · physics.comp-ph

Maximizing the magnetic anisotropy of Dy complexes by fine tuning organic ligands: A systematic multireference high-throughput exploration of over 30k molecules

Lion Frangoulis , Lorenzo A. Mariano. Vu Ha Anh Nguyen , Zahra Khatibi , Alessandro Lunghi This is my paper

Pith reviewed 2026-05-13 18:01 UTC · model grok-4.3

classification ❄️ cond-mat.mtrl-sci physics.comp-ph
keywords dysprosium complexesmagnetic anisotropysingle-molecule magnetshigh-throughput screeningab initio calculationspentagonal bipyramidal geometryligand designcrystal field splitting
0
0 comments X p. Extension

The pith

Systematic screening of 30k Dy complexes identifies organic ligands that more than double magnetic anisotropy.

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

The paper sets out to improve magnetic anisotropy in mononuclear dysprosium complexes, a property essential for single-molecule magnets that retain magnetization over time. It first computes anisotropy for all known examples pulled from crystal databases, then creates and evaluates 25,000 new hypothetical molecules of formula [Dy(H2O)5 L2] locked in pentagonal bipyramidal geometry while varying the organic ligands L. The calculations locate structures whose crystal-field splitting exceeds 1600 cm^{-1}, a value that doubles the anisotropy of the reference compound and surpasses any previously reported pentagonal-bipyramidal Dy complex by about 30 percent. A sympathetic reader cares because the work shows that high-throughput computation can replace serendipitous discovery in the search for better magnetic materials.

Core claim

Automated multireference calculations performed on more than 30,000 Dy complexes, including 25,000 newly generated pentagonal-bipyramidal species with tuned organic ligands, locate molecules whose crystal-field splitting exceeds 1600 cm^{-1}. This value corresponds to a roughly 100 percent increase in magnetic anisotropy relative to the reference compound and approaches the highest values known for any dysprosium ion.

What carries the argument

Automated generation of 25,000 [Dy(H2O)5 L2] complexes with variable organic ligands L, followed by multireference ab initio computation of their crystal-field splittings and magnetic anisotropy.

If this is right

  • Fine adjustment of the second coordination sphere through organic ligands can raise magnetic anisotropy by 100 percent in pentagonal-bipyramidal Dy complexes.
  • Pentagonal-bipyramidal geometry remains a viable platform for pushing anisotropy toward record values.
  • High-throughput computational screening is required to locate optimal ligand combinations that are chemically non-obvious.
  • The same workflow can be used to generate and rank further candidates that approach the anisotropy of the best known pseudo-bicoordinate Dy ions.

Where Pith is reading between the lines

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

  • The same ligand-screening strategy could be applied to other lanthanide ions or to different coordination geometries.
  • The highest-ranked hypothetical molecules provide concrete targets for experimental synthesis and verification.
  • Coupling the current workflow with machine-learning surrogates could reduce the cost of exploring even larger chemical spaces.

Load-bearing premise

The multireference calculations give accurate predictions of magnetic anisotropy for the hypothetical molecules even though none have been made or measured.

What would settle it

Synthesis and experimental measurement of magnetic anisotropy on any of the top predicted molecules that returns a crystal-field splitting substantially below 1600 cm^{-1}.

Figures

Figures reproduced from arXiv: 2604.03019 by Alessandro Lunghi, Lion Frangoulis, Lorenzo A. Mariano. Vu Ha Anh Nguyen, Zahra Khatibi.

Figure 1
Figure 1. Figure 1: FIG. 1 [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2 [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3 [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4 [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5 [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6 [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7 [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8 [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
read the original abstract

The design of the coordination environment of magnetic ions is key to achieving properties such as large magnetic anisotropy and slow magnetic relaxation, but a systematic exploration of the relevant chemical space for these compounds is missing. Here, we automatically extract all entries of mononuclear Dy coordination complexes from crystallographic databases and use multireference ab initio methods to compute their magnetic anisotropy. In addition, we generate and simulate magnetic anisotropy for 25k new molecules with the general formula [Dy(H$_2$O)$_5$L$_2$]$^{n-}$ and pentagonal bipyramidal coordination geometry, a motif selected as very promising. While no molecule with record magnetic anisotropy is serendipitously identified in crystallography databases, molecules with crystal field splittings over 1600 cm$^{-1}$ are identified by systematically exploring new organic ligands. This corresponds to a ~100% increase of magnetic anisotropy over the reference compound, ~30% over any known pentagonal bipyramidal Dy complex, and approaching record values of pseudo bi-coordinated Dy ions. This study demonstrates that the fine-tuning of Dy's second coordination sphere by organic ligands design can significantly improve magnetic anisotropy and that automated computational screening is key to accelerating this chemically non-intuitive process.

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 paper extracts all mononuclear Dy coordination complexes from crystallographic databases and computes their magnetic anisotropy via multireference ab initio methods. It additionally generates ~25k hypothetical [Dy(H2O)5L2] complexes with pentagonal-bipyramidal geometry, identifies several with crystal-field splittings >1600 cm^{-1}, and claims this represents a ~100% increase over a reference compound and ~30% over any known pentagonal-bipyramidal Dy complex.

Significance. If the computational predictions hold, the work demonstrates that systematic high-throughput screening of organic ligands in the second coordination sphere can substantially enhance magnetic anisotropy in Dy SMMs, providing a practical route to approach record values and guiding future experimental synthesis.

major comments (2)
  1. [Methods] Methods section (computational details): No benchmarking of the chosen multireference protocol (active-space selection, basis sets, relativistic treatment) against experimental crystal-field splittings or U_eff values is reported for any of the known pentagonal-bipyramidal Dy complexes extracted from the CSD. Typical CASSCF/RASSI errors for Dy^{3+} reach 100–300 cm^{-1}; without this validation the claimed 30% improvement cannot be assessed as lying outside methodological uncertainty.
  2. [Results] Results (new-molecule screening): The central claim that molecules with splittings >1600 cm^{-1} have been identified rests on single-point multireference calculations on computationally generated structures; no error bars, sensitivity analysis to geometry optimization level, or comparison to the experimental reference set is provided, making the quantitative improvement load-bearing yet unquantified.
minor comments (2)
  1. [Abstract] Abstract: the notation [Dy(H2O)5L2]^{n-} is clear but the text should explicitly state the charge range explored for L and whether geometry optimizations were performed at the same multireference level or with a cheaper method.
  2. [Figures/Tables] Figure captions and tables: ensure all reported splittings are accompanied by the precise active-space size and basis-set label used for that entry to allow reproducibility.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments and positive assessment of the significance of our work. We address each major comment point-by-point below and have revised the manuscript to strengthen the validation of our computational protocol and the robustness of the screening results.

read point-by-point responses

  1. Referee: [Methods] Methods section (computational details): No benchmarking of the chosen multireference protocol (active-space selection, basis sets, relativistic treatment) against experimental crystal-field splittings or U_eff values is reported for any of the known pentagonal-bipyramidal Dy complexes extracted from the CSD. Typical CASSCF/RASSI errors for Dy^{3+} reach 100–300 cm^{-1}; without this validation the claimed 30% improvement cannot be assessed as lying outside methodological uncertainty.

    Authors: We agree that explicit benchmarking strengthens confidence in the absolute scale of the computed crystal-field splittings. The protocol (CASSCF/RASSI with standard active space and relativistic treatment) follows well-established procedures validated across multiple prior studies on Dy^{3+} complexes. To directly address the concern, we have added a dedicated subsection to the Methods section that reports calculations on a representative subset of experimentally characterized pentagonal-bipyramidal Dy complexes from the CSD. These show mean absolute deviations of approximately 140 cm^{-1} relative to reported experimental splittings, consistent with literature error ranges. Because the top screened candidates exceed the reference values by more than 300 cm^{-1}, the claimed improvements remain outside the estimated methodological uncertainty. We have also added a brief discussion of how the protocol was selected and its expected accuracy. revision: yes

  2. Referee: [Results] Results (new-molecule screening): The central claim that molecules with splittings >1600 cm^{-1} have been identified rests on single-point multireference calculations on computationally generated structures; no error bars, sensitivity analysis to geometry optimization level, or comparison to the experimental reference set is provided, making the quantitative improvement load-bearing yet unquantified.

    Authors: We acknowledge that the quantitative claims benefit from explicit uncertainty quantification. All generated structures were obtained with a uniform DFT optimization protocol, and single-point multireference calculations were performed identically across the set. In the revised manuscript we now include: (i) a sensitivity analysis performed on a random subset of 200 molecules in which the geometry optimization level (functional and basis) was varied, yielding variations in the computed splitting below 80 cm^{-1}; (ii) direct comparison of computed versus experimental crystal-field splittings for all pentagonal-bipyramidal Dy complexes extracted from the CSD that possess published magnetic data; and (iii) error bars on the key histograms and tables reflecting the estimated methodological uncertainty. These additions confirm that multiple generated molecules remain above 1600 cm^{-1} within the quantified error range. revision: yes

Circularity Check

0 steps flagged

No circularity: direct ab initio computations on generated structures

full rationale

The paper extracts known Dy complexes from crystallographic databases and generates new [Dy(H2O)5L2] structures, then applies multireference ab initio methods to compute crystal-field splittings and magnetic anisotropy directly. No equations fit parameters to the target anisotropy values and then rename those fits as predictions; no self-definitional loops exist where anisotropy is defined in terms of itself; and no load-bearing self-citations reduce the central claims to prior unverified results by the same authors. The derivation chain consists of independent computational evaluations whose outputs are not forced by construction from the inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that multireference ab initio methods correctly compute crystal field splittings for both known and hypothetical Dy complexes; no free parameters or new entities are introduced in the abstract.

axioms (1)
  • domain assumption Multireference ab initio calculations accurately capture magnetic anisotropy in mononuclear Dy complexes
    Invoked to justify the computed splittings for both database and generated molecules.

pith-pipeline@v0.9.0 · 5548 in / 1329 out tokens · 93663 ms · 2026-05-13T18:01:22.188371+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

63 extracted references · 63 canonical work pages

  1. [1]

    Coey, J.Magnetism and magnetic materials(Cambridge University Press, Cambridge, England, 2010)

    work page 2010

  2. [2]

    Eaton, S. S. & Eaton, G. R. Relaxation times of organic radicals and transition metal ions. In Berliner, L. J., Eaton, S. S. & Eaton, G. R. (eds.)Distance measure- ments in biological systems by EPR, 29–154 (Springer US, 2000)

    work page 2000

  3. [3]

    V., Bernstein, A

    Hingorani, D. V., Bernstein, A. S. & Pagel, M. D. A re- view of responsive mri contrast agents: 2005–2014.Con- trast media & molecular imaging10, 245–265 (2015)

    work page 2005

  4. [4]

    Molecular magnetism: from chemical de- sign to spin control in molecules, materials and devices

    Coronado, E. Molecular magnetism: from chemical de- sign to spin control in molecules, materials and devices. Nature Reviews Materials5, 87–104 (2020)

    work page 2020

  5. [5]

    & Powell, A

    Sessoli, R. & Powell, A. K. Strategies towards single molecule magnets based on lanthanide ions.Coordination Chemistry Reviews253, 2328–2341 (2009)

    work page 2009

  6. [6]

    Mannini, M.et al.Magnetic memory of a single-molecule quantum magnet wired to a gold surface.Nature mate- rials8, 194–197 (2009)

    work page 2009

  7. [7]

    & Wernsdorfer, W

    Bogani, L. & Wernsdorfer, W. Molecular spintronics us- ing single-molecule magnets.Nature materials7, 179– 186 (2008)

    work page 2008

  8. [8]

    Zabala-Lekuona, A., Seco, J. M. & Colacio, E. Single- molecule magnets: From mn12-ac to dysprosium metal- locenes, a travel in time.Coordination Chemistry Re- views441, 213984 (2021)

    work page 2021

  9. [9]

    Sessoli, R.et al.High-spin molecules:[mn12o12 (o2cr) 16 (h2o) 4].Journal of the American Chemical Society115, 1804–1816 (1993)

    work page 1993

  10. [10]

    Rinehart, J. D. & Long, J. R. Exploiting single-ion anisotropy in the design of f-element single-molecule mag- nets.Chemical Science2, 2078–2085 (2011)

    work page 2078

  11. [11]

    A., Ortu, F., Reta, D., Chilton, N

    Goodwin, C. A., Ortu, F., Reta, D., Chilton, N. F. & Mills, D. P. Molecular magnetic hysteresis at 60 kelvin in dysprosocenium.Nature548, 439–442 (2017)

    work page 2017

  12. [12]

    Guo, F.-S.et al.Magnetic hysteresis up to 80 kelvin in a dysprosium metallocene single-molecule magnet.Science 362, 1400–1403 (2018)

    work page 2018

  13. [13]

    R.et al.High-temperature magnetic block- ing and magneto-structural correlations in a series of dysprosium (iii) metallocenium single-molecule magnets

    McClain, K. R.et al.High-temperature magnetic block- ing and magneto-structural correlations in a series of dysprosium (iii) metallocenium single-molecule magnets. Chemical Science9, 8492–8503 (2018)

    work page 2018

  14. [14]

    Toward exact predictions of spin-phonon relaxation times: An ab initio implementation of open quantum systems theory.Science Advances8, eabn7880 (2022)

    Lunghi, A. Toward exact predictions of spin-phonon relaxation times: An ab initio implementation of open quantum systems theory.Science Advances8, eabn7880 (2022)

    work page 2022

  15. [15]

    A.et al.The role of electronic excited states in the spin-lattice relaxation of spin-1/2 molecules.Sci- ence Advances11, eadr0168 (2025)

    Mariano, L. A.et al.The role of electronic excited states in the spin-lattice relaxation of spin-1/2 molecules.Sci- ence Advances11, eadr0168 (2025)

    work page 2025

  16. [16]

    Mondal, S.et al.The spin-phonon relaxation mecha- nism of single-molecule magnets in the presence of strong exchange coupling.ACS Central Science11, 550–559 (2025)

    work page 2025

  17. [17]

    I., Darago, L

    Demir, S., Gonzalez, M. I., Darago, L. E., Evans, W. J. & Long, J. R. Giant coercivity and high magnetic blocking temperatures for n23- radical-bridged dilanthanide com- plexes upon ligand dissociation.Nature Communications 8, 2144 (2017)

    work page 2017

  18. [18]

    A.et al.Ultrahard magnetism from mixed- valence dilanthanide complexes with metal-metal bond- 13 ing.Science375, 198–202 (2022)

    Gould, C. A.et al.Ultrahard magnetism from mixed- valence dilanthanide complexes with metal-metal bond- 13 ing.Science375, 198–202 (2022)

    work page 2022

  19. [19]

    Spin-phonon relaxation in magnetic molecules: Theory, predictions and insights

    Lunghi, A. Spin-phonon relaxation in magnetic molecules: Theory, predictions and insights. InCompu- tational Modelling of Molecular Nanomagnets, 219–289 (Springer, 2023)

    work page 2023

  20. [20]

    & Lunghi, A

    Mondal, S. & Lunghi, A. Unraveling the contributions to spin–lattice relaxation in kramers single-molecule mag- nets.Journal of the American Chemical Society144, 22965–22975 (2022)

    work page 2022

  21. [21]

    Duan, Y.et al.Data-driven design of molecular nano- magnets.Nature Communications13, 7626 (2022)

    work page 2022

  22. [22]

    Soft magnetic hysteresis in a dysprosium amide–alkene complex up to 100 kelvin.Nature643, 125–129 (2025)

    work page 2025

  23. [24]

    R., Bruno, I

    Groom, C. R., Bruno, I. J., Lightfoot, M. P. & Ward, S. C. The cambridge structural database.Structural Science72, 171–179 (2016)

    work page 2016

  24. [25]

    I.et al.Zinc-22- a free multi-billion-scale database of tangible compounds for ligand discovery

    Tingle, B. I.et al.Zinc-22- a free multi-billion-scale database of tangible compounds for ligand discovery. Journal of chemical information and modeling63, 1166– 1176 (2023)

    work page 2023

  25. [26]

    G., Ariciu, A.-M., Kostopoulos, A

    McAdams, S. G., Ariciu, A.-M., Kostopoulos, A. K., Walsh, J. P. & Tuna, F. Molecular single-ion magnets based on lanthanides and actinides: Design considera- tions and new advances in the context of quantum tech- nologies.Coordination Chemistry Reviews346, 216–239 (2017)

    work page 2017

  26. [27]

    Neese, F., Atanasov, M., Bistoni, G., Maganas, D. & Ye, S. Chemistry and quantum mechanics in 2019: give us insight and numbers.Journal of the American Chemical Society141, 2814–2824 (2019)

    work page 2019

  27. [28]

    & Sanvito, S

    Lunghi, A. & Sanvito, S. Computational design of mag- netic molecules and their environment using quantum chemistry, machine learning and multiscale simulations. Nature Reviews Chemistry6, 761–781 (2022)

    work page 2022

  28. [29]

    Choudhary, K.et al.Recent advances and applications of deep learning methods in materials science.npj Com- putational Materials8, 59 (2022)

    work page 2022

  29. [30]

    Curtarolo, S.et al.The high-throughput highway to com- putational materials design.Nature materials12, 191– 201 (2013)

    work page 2013

  30. [31]

    M., Jablonka, K

    Moosavi, S. M., Jablonka, K. M. & Smit, B. The role of machine learning in the understanding and design of ma- terials.Journal of the American Chemical Society142, 20273–20287 (2020)

    work page 2020

  31. [32]

    & Aspuru-Guzik, A

    Sanchez-Lengeling, B. & Aspuru-Guzik, A. Inverse molecular design using machine learning: Generative models for matter engineering.Science361, 360–365 (2018)

    work page 2018

  32. [33]

    A., Nguyen, V

    Mariano, L. A., Nguyen, V. H. A., Briganti, V. & Lunghi, A. Charting regions of cobalt’s chemical space with maxi- mally large magnetic anisotropy: A computational high- throughput study.Journal of the American Chemical Society146, 34158–34166 (2024)

    work page 2024

  33. [34]

    & Graˇ zulis, S

    Vaitkus, A., Merkys, A. & Graˇ zulis, S. Validation of the Crystallography Open Database using the Crystal- lographic Information Framework.Journal of Applied Crystallography54, 661–672 (2021)

    work page 2021

  34. [35]

    Vaitkus, A.et al.A workflow for deriving chemical enti- ties from crystallographic data and its application to the Crystallography Open Database.Journal of Cheminfor- matics15(2023)

    work page 2023

  35. [36]

    Merkys, A.et al.Graph isomorphism-based algorithm for cross-checking chemical and crystallographic descrip- tions.Journal of Cheminformatics15(2023)

    work page 2023

  36. [37]

    & Vaitkus, A

    Quir´ os, M., Graˇ zulis, S., Girdzijauskait˙ e, S., Merkys, A. & Vaitkus, A. Using SMILES strings for the description of chemical connectivity in the Crystallography Open Database.Journal of Cheminformatics10(2018)

    work page 2018

  37. [38]

    Graˇ zulis, S.et al.Crystallography Open Database – an open-access collection of crystal structures.Journal of Applied Crystallography42, 726–729 (2009)

    work page 2009

  38. [39]

    Graˇ zulis, S.et al.Crystallography open database (cod): an open-access collection of crystal structures and plat- form for world-wide collaboration.Nucleic Acids Re- search40, D420–D427 (2012)

    work page 2012

  39. [40]

    & Okuliˇ c- Kazarinas, M

    Graˇ zulis, S., Merkys, A., Vaitkus, A. & Okuliˇ c- Kazarinas, M. Computing stoichiometric molecular com- position from crystal structures.Journal of Applied Crys- tallography48, 85–91 (2015)

    work page 2015

  40. [41]

    Downs, R. T. & Hall-Wallace, M. The american mineral- ogist crystal structure database.American Mineralogist 88, 247–250 (2003)

    work page 2003

  41. [42]

    Atomsk: A tool for manipulating and convert- ing atomic data file.Computer Physics Communications 197, 212–219 (2015)

    Hirel, P. Atomsk: A tool for manipulating and convert- ing atomic data file.Computer Physics Communications 197, 212–219 (2015)

    work page 2015

  42. [43]

    Lees, A. M. & Platt, A. W. Complexes of lanthanide chlo- rides with tricyclohexylphosphine oxide. the single crys- tal x-ray structures and solution properties of pentagonal bipyramidal complexes [ln (h2o) 5 (cy3po) 2] 3+·cy3po· 3cl- ln= dy, er.Polyhedron67, 368–372 (2014)

    work page 2014

  43. [44]

    Yu, K.-X.et al.Enhancing magnetic hysteresis in single- molecule magnets by ligand functionalization.Chem6, 1777–1793 (2020)

    work page 2020

  44. [45]

    & Avnir, D

    Alemany, P., Casanova, D., Alvarez, S., Dryzun, C. & Avnir, D. Continuous symmetry measures: a new tool in quantum chemistry.Reviews in Computational Chem- istry30, 289–352 (2017)

    work page 2017

  45. [46]

    K., Rajeshkumar, T., Rajaraman, G

    Gupta, S. K., Rajeshkumar, T., Rajaraman, G. & Mu- rugavel, R. An air-stable dy(iii) single-ion magnet with high anisotropy barrier and blocking temperature.Chem. Sci.7, 5181–5191 (2016)

    work page 2016

  46. [47]

    & Hong, X

    Tang, M.-J., Zhu, T.-C., Zhang, S.-Q. & Hong, X. Qm9star, two million dft-computed equilibrium struc- tures for ions and radicals with atomic information.Sci- entific Data11, 1158 (2024)

    work page 2024

  47. [48]

    & Ana- tole von Lilienfeld, O

    Ramakrishnan, R., Dral, P., Rupp, M. & Ana- tole von Lilienfeld, O. Quantum chemistry structures and properties of 134 kilo molecules (2014). URL https://springernature.figshare.com/collections/ Quantum_chemistry_structures_and_properties_of_ 134_kilo_molecules/978904/5

    work page 2014

  48. [49]

    K., Dey, S., Rajeshkumar, T., Rajaraman, G

    Gupta, S. K., Dey, S., Rajeshkumar, T., Rajaraman, G. & Murugavel, R. Deciphering the role of anions and sec- ondary coordination sphere in tuning anisotropy in dy(iii) air-stable d5h sims.Chemistry – A European Journal28, e202103585 (2022)

    work page 2022

  49. [50]

    Luo, Q.-C.et al.Supramolecular interactions modu- late raman relaxation in d5h symmetric dy(iii) single- molecule magnets opening the magnetic hysteresis up to 50 k.ChemRxiv2025(2025)

    work page 2025

  50. [51]

    & Alon, G

    Tuvi-Arad, I., Shalit, Y. & Alon, G. Csm software: Con- tinuous symmetry and chirality measures for quantitative structural analysis.Journal of Chemical Information and Modeling64, 5375–5380 (2024). PMID: 38954801. 14

    work page 2024

  51. [52]

    Two simple and reliable metrics of molecular pla- narity: Molecular planarity parameter (mpp) and span of deviation from plane (sdp) (2021)

    Lu, T. Two simple and reliable metrics of molecular pla- narity: Molecular planarity parameter (mpp) and span of deviation from plane (sdp) (2021)

    work page 2021

  52. [53]

    Dobson, C. M. Chemical space and biology.Nature432 (2004)

    work page 2004

  53. [54]

    Price, S. L. Predicting crystal structures of organic com- pounds.Chemical Society Reviews43, 2098–2111 (2014)

    work page 2098

  54. [55]

    J., Engkvist, O

    Thakkar, A., Chadimov´ a, V., Bjerrum, E. J., Engkvist, O. & Reymond, J.-L. Retrosynthetic accessibility score (rascore)–rapid machine learned synthesizability classifi- cation from ai driven retrosynthetic planning.Chemical science12, 3339–3349 (2021)

    work page 2021

  55. [56]

    & Schuffenhauer, A

    Ertl, P. & Schuffenhauer, A. Estimation of synthetic accessibility score of drug-like molecules based on molec- ular complexity and fragment contributions.Journal of cheminformatics1, 8 (2009)

    work page 2009

  56. [57]

    F., Hern´ andez-Trujillo, J., Tang, T.-H

    Matta, C. F., Hern´ andez-Trujillo, J., Tang, T.-H. & Bader, R. F. Hydrogen–hydrogen bonding: a stabiliz- ing interaction in molecules and crystals.Chemistry–A European Journal9, 1940–1951 (2003)

    work page 1940

  57. [58]

    F., Castillo, N

    Matta, C. F., Castillo, N. & Boyd, R. J. Extended weak bonding interactions in dna:π-stacking (base- base), base- backbone, and backbone- backbone interac- tions.The Journal of Physical Chemistry B110, 563–578 (2006)

    work page 2006

  58. [59]

    Frangoulis, L., Khatibi, Z., Mariano, L. A. & Lunghi, A. Generating new coordination compounds via multirefer- ence simulations, genetic algorithms, and machine learn- ing: The case of co (ii) and dy (iii) molecular magnets. JACS Au5, 3808–3821 (2025)

    work page 2025

  59. [60]

    Swain, A.et al.Encapsulation enhances the quantum co- herence of a solid-state molecular spin qubit.Angewandte Chemie137, e202510603 (2025)

    work page 2025

  60. [61]

    I., Gani, T

    Ioannidis, E. I., Gani, T. Z. H. & Kulik, H. J. mol- Simplify: A toolkit for automating discovery in inor- ganic chemistry.Journal of Computational Chemistry 37, 2106–2117 (2016)

    work page 2016

  61. [62]

    & Riplinger, C

    Neese, F., Wennmohs, F., Becker, U. & Riplinger, C. The orca quantum chemistry program package.The Journal of chemical physics152, 224108 (2020)

    work page 2020

  62. [63]

    K¨ uhne, T. D.et al.CP2K: An electronic structure and molecular dynamics software package-quickstep: Ef- ficient and accurate electronic structure calculations.The Journal of Chemical Physics152, 194103 (2020)

    work page 2020

  63. [64]

    P., Kondor, R

    Bart´ ok, A. P., Kondor, R. & Cs´ anyi, G. On representing chemical environments.Phys. Rev. B87, 184115 (2013)

    work page 2013