Software for building transition metal complexes in 3D

Snowdrop Science Ltd develops software for building organometallic complexes atom by atom and preparing their 3D structures for quantum chemical calculations. The work so far has focused on half-sandwich ruthenium(II) arene complexes, a class of compounds studied as anticancer agents [1, 2].

0:00 / 0:40

Figure 1. Building RM175, [(η⁶-biphenyl)Ru(en)Cl]⁺, in the Snowdrop builder. The build is shown at 15 times normal speed. The biphenyl atoms are placed and bonded, then chloride and ethylenediamine (en) are bonded to ruthenium, and ruthenium is bonded to the six carbon atoms of one ring, giving the half-sandwich structure. The final part, at normal speed, shows the atom inspector listing the bonds and coordinates of the ruthenium atom and then of one ethylenediamine nitrogen atom. RM175 was first reported by Morris et al. [1] and its crystal structure is deposited as CCDC 170362 [3].

Software

Building

Atoms and bonds are placed by hand in a 3D view. Each structure records the hapticity of every ligand bound to a metal through several atoms, such as the η⁶ arene in RM175, together with chelating ligands, dative metal–ligand bonds and formal charges.

Clean up

The Clean up command positions atoms using structural chemistry rules. These set valence shell electron pair repulsion (VSEPR) geometries, close aromatic rings, place chelating ligands and seat a ligand bound through several atoms, such as an η⁶ arene, on one face of an octahedral metal centre. The result is a starting geometry for the Minimise command.

Minimise

The Minimise command optimises the structure with the semi-empirical method GFN2-xTB in implicit water, using the ALPB solvation model [4, 5]. Weak harmonic restraints are applied to the metal coordination sphere during the optimisation. A result is accepted only if the optimisation converges to a gradient norm below 0.001 hartree per bohr and no bond is broken or formed. For the ruthenium arene complexes, all six ruthenium–carbon distances to the arene must also stay between 1.80 and 2.60 Å and within 0.25 Å of each other.

Export

Each exported structure records, for every atom, whether its coordinates came from Clean up or from Minimise. Before a file is saved under an existing compound name, the software checks that the molecule’s bond graph matches the graph recorded for that name.

Validation

Thirteen ruthenium(II) arene complexes with published crystal structures [1, 6] were optimised with GFN2-xTB in implicit water, without restraints, starting from their crystal coordinates. Table 1 compares each optimised structure with its crystal structure.

Table 1. Absolute differences between the GFN2-xTB and crystal structures, in Å. RMSD is the heavy-atom root-mean-square deviation after superposition [7]. Complex numbers follow Habtemariam et al. [6].

Complex	CCDC	Arene	Chelating ligand	RMSD	Ru to arene centroid	Ru–N	Ru–Cl
RM175	170362	biphenyl	ethylenediamine	0.161	0.131	0.007	0.022
16	607563	fluorene	ethylenediamine	0.117	0.122	0.006	0.028
22	607565	tetralin	ethylenediamine	0.123	0.116	0.003	0.024
27	607573	indan	1,3-diamino-2-propanol	0.100	0.140	0.016	0.014
28	607562	indan	homopiperazine	0.080	0.147	0.024	0.032
36	607564	tetrahydroanthracene	1,2-diaminobenzene	0.506	0.129	0.021	0.004
38	607570	indan	2,2′-bipyridine	0.083	0.189	0.106	0.029
39	607572	biphenyl	2,2′-bipyridine	0.122	0.157	0.095	0.014
43	607571	indan	1,10-phenanthroline	0.072	0.177	0.089	0.035
53	607567	p-cymene	8-hydroxyquinolinate	0.176	0.158	0.085	0.022
60	607574	p-cymene	2,2,6,6-tetramethylheptane-3,5-dionate	0.354	0.179	n/a	0.021
61	607566	p-cymene	hexafluoroacetylacetonate	0.464	0.203	n/a	0.043
62	607568	p-cymene	chromenone diketonate	0.580	0.187	n/a	0.019

All 13 optimisations converged with no bonds broken or formed. The median differences were 0.022 Å for Ru–N bonds in the 10 complexes with nitrogen donors, 0.022 Å for Ru–Cl bonds and 0.68° for the chelate bite angle. The median heavy-atom RMSD was 0.12 Å.

Two differences are systematic. GFN2-xTB placed ruthenium on average 0.16 Å further from the arene centroid than in the crystal structures, and it shortened bonds to aromatic nitrogen donors by 0.085 to 0.106 Å. In the four complexes with the largest RMSD, 0.35 to 0.58 Å, the Ru–Cl differences are still at most 0.043 Å. Their RMSD comes from the orientation of fused rings and ligand substituents. The crystal structures are solids that include counter-ions, while the calculations model a single complex in implicit water.

RM175 compared with density functional theory

RM175 was also optimised with density functional theory, using the ωB97M-V functional [8], the def2-SVP basis set [9] with an effective core potential for ruthenium [10] and the SMD water model [11]. The calculation ran on a cloud GPU with PySCF and GPU4PySCF [12, 13]. A vibrational frequency calculation supports a local minimum. Table 2 follows RM175 through each stage.

Table 2. RM175 at each stage compared with its crystal structure. Distances in Å and angles in degrees. The Clean up structure comes from a run on 11 September 2026. RMSD is over the 18 heavy atoms.

Structure	Ru–Cl	Ru–N	Mean Ru–C	Ru height above arene	N–Ru–N	RMSD to crystal
Crystal structure, CCDC 170362	2.408	2.111, 2.120	2.178	1.661	79.2	0.00
Clean up	2.480	2.170, 2.170	2.220	1.723	90.0	1.53
Clean up, then GFN2-xTB	2.389	2.120, 2.131	2.278	1.786	79.2	1.46
DFT, ωB97M-V/def2-SVP	2.406	2.116, 2.122	2.192	1.669	79.7	1.05

The DFT structure matches the crystal structure to within 0.014 Å for the Ru–Cl, Ru–N and mean Ru–C distances. Its arene is rotated by 69° about the axis from ruthenium to the arene centroid relative to the crystal structure, which accounts for most of its 1.05 Å RMSD. Clean up alone sets an ideal 90° N–Ru–N angle, and GFN2-xTB relaxes it to 79.2°, the crystal value. The GFN2-xTB structure keeps a different ligand orientation from the crystal structure, so its RMSD stays at 1.46 Å.

Company

Snowdrop Science Ltd is a private limited company registered in England and Wales, company number 16848046, incorporated in November 2025.

Jed Walsh founded the company and is its director. In his MChem research at the University of Leeds, he developed an inline method for determining the encapsulation efficiency of RNA lipid nanoparticles. In 2024 he took second place in the Reckitt Sustainability Challenge. The name Snowdrop Science is a dedication to his grandmother. The chemotherapy drug she received had poor selectivity for cancer cells, and that experience led him to work on metal-based anticancer compounds beyond platinum that could act more selectively.

Jed Walsh on LinkedIn

Contact

For questions about the software or the validation results, email jed.walsh@snowdropscience.co.uk.

References

1R. E. Morris et al., J. Med. Chem. 2001, 44, 3616. doi.org/10.1021/jm010051m
2R. E. Aird et al., Br. J. Cancer 2002, 86, 1652. doi.org/10.1038/sj.bjc.6600290
3CCDC 170362, Experimental Crystal Structure Determination, Cambridge Crystallographic Data Centre, 2001. doi.org/10.5517/cc5q8kp
4C. Bannwarth, S. Ehlert and S. Grimme, J. Chem. Theory Comput. 2019, 15, 1652. doi.org/10.1021/acs.jctc.8b01176
5S. Ehlert, M. Stahn, S. Spicher and S. Grimme, J. Chem. Theory Comput. 2021, 17, 4250. doi.org/10.1021/acs.jctc.1c00471
6A. Habtemariam et al., J. Med. Chem. 2006, 49, 6858. doi.org/10.1021/jm060596m
7W. Kabsch, Acta Crystallogr. A 1976, 32, 922. doi.org/10.1107/S0567739476001873
8N. Mardirossian and M. Head-Gordon, J. Chem. Phys. 2016, 144, 214110. doi.org/10.1063/1.4952647
9F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297. doi.org/10.1039/B508541A
10D. Andrae, U. Häußermann, M. Dolg, H. Stoll and H. Preuß, Theor. Chim. Acta 1990, 77, 123. doi.org/10.1007/BF01114537
11A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B 2009, 113, 6378. doi.org/10.1021/jp810292n
12Q. Sun et al., J. Chem. Phys. 2020, 153, 024109. doi.org/10.1063/5.0006074
13X. Wu et al., WIREs Comput. Mol. Sci. 2025, 15, e70008. doi.org/10.1002/wcms.70008