Water molecules are omnipresent in nature and are a key part of many life processes. Due its ability of hydrogen binding water molecule plays an important role in the packing of small molecule crystal structures. Over the past years, the structure of a water molecule has been intensively studied. [1] The experimental values for a free water molecule in the gas phase are the bond angle (H–O–H) of 104.52 ± 0.05° and the bond (O–H) length of 0.9572 ± 0.0003 Å. [2] Neutron diffraction experiments of liquid water showed that the bond angle increases to 106.1 ± 1.8° and the bond length increases to 0.970 ± 0.005 Å. [3] Most of the bond angles in structures of ice have values close to a tetrahedral angle. However, in some of the ice structures, the bond angles and bond lengths remarkably deviates. Calculations based on the spectroscopic potential energy surface showed the equilibrium structure of a water molecule with the bond angle of 104.501 ± 0.005° and the bond length of 0.95785 ± 0.00005 Å. [4] In this study, [5] we performed an analysis of non-coordinated water containing structures archived in Cambridge Structural Database (CSD) as well as ab-initio calculations on a range of bond angles and bond lengths of water molecule. The results of the analysis of crystal structures solved by neutron as well as by X-ray diffraction analysis showed a large discrepancy of both the bond angle and bond length values. Namely, the ranges of the bond angle and the average bond lengths of neutron solved structures having R factor ≤ 0.05 are from 100.74° to 113.92° and from 0.91 Å to 0.99 Å respectively. The corresponding range of the bond angle of X-ray solved structures is from 13.27° to 180.00°. High level ab initio calculations predicted a possibility for energetically low-cost (±1 kcal mol–1) changes of both the bond angle and bond lengths in a wide range, from 96.4° to 112.8° (Fig. 1) and from 0.930 A to 0.989 A (Fig. 1), respectively. Consequently, it would lead to at least 15% of X-ray solved structures that contain questionable water molecule geometries.

In recent years, quantum crystallography, which combines X-ray crystal structure analysis and quantum chemistry, has attracted attention. Attempts have been made to reproduce wave function from electron density obtained from experiments and to refine the crystal structure using computational chemistry [1,2]. However, these attempts require a high degree of expertise and such programs are not widely used. The purpose of this study is to discuss electronic states of azo-Schiff base metal complexes based on quantum chemical calculations and to verify whether quantum crystallography can be performed easily by using conventional programs.

The samples investigated were two Schiff base metal complexes having azobenzene moiety (new Cu of trans-[CuN₂O₂] and known Mn of cis-[CuN₂O₂X₂]) (Fig. 1) studied on photochemical behavior [3]. Experimental electron density was drawn using a PLATON program. DFT calculation was carried out with a Gaussian09, and electron density analysis and bond order analysis were also performed. Additionally, a CRYSTAL EXPLORER program was used for Hirschfeld surface analysis. Experimental and calculated electron density maps exhibited good agreement (Fig. 2) and gave additional information such as bond strength only with the aid of DFT.

[1] Grabowsky, S.; Genoni, A.; Burgide, H.-B. Chem. Sci., 2017, 8, 4159–4176.

[2] Genoni, A.; Bučinský, L. Claiser, N.; Contreras-García, J.; Dittrich, B.; Dominiak, P. M.; Espinosa, E.; Gatti, C.; Giannozzi, P.; Gillet, J.-M.; Jayatilaka, D.; Macchi, P.; Madsen, A. Ø.; Massa, L.; Matta, C. F.; Merz Jr., K. M.; Nakashima, P. N. H.; Ott, H.; Ryde, U.; Schwarz, K.; Sierka, M.; Grabowsky, S. Chem. Eur. J., 2018, 24, 10881-10905.

[3] Takiguchi, Y.; Onami, Y.; Haraguchi, T.; Akitsu, T. Molecules, 2021, 26, 551 (12 pages).

Hydrogen bonds are of greate importance for understanding of different processes in chemistry, crystalography and biology. [1] Properties of hydrogen bonds were subject of numerous experimental and theoretical studies. [1] Specially interesting case represent hydrogen bonds involving transition metal complexes since coordination of ligands can have significant influence on electrostatic potentials of coordinated molecules. [2] Here we present detailed analysis of crystalographic data combined with quantum chemical calculations of very strong hydrogen bonds between water and acetylacetonate ligand of different transition metal complexes.

Cambridge Structural Database (CSD) was searched for all structures containing O-H/O interactions between water molecule and acetylacetonato ligands of transition metal complexes. Geometrical parameters extracted from crystall structures were analyzed and compared with quantum chemical calculations performed on model systems. The O-H/O interactions were studied on model systems containing water and neutral or charged square-planar complexes of Ir, Rh, Pd, and Pt. The strongest interactions were found in charged model systems and these results are in agreement with the predominant electrostatic nature of hydrogen bond (-16.54 kcal/mol). However, suprisingly strong O-H/O interactions were identified also in neutral model systems. The calculated energies of these interactions are -7.98 and -8.22 kcal/mol in [M(acac)(en)]/H₂O (M = Ir(I), Rh(I)) model system, respectively.

Using geometrical cristeria for hydrogen bonds 82 structures with 220 O-H/O contacts involving water molecule and coordinated acetylacetonato fragment were found. All extracted structures were statistically analyzed and results of analysis were in agreement with the results of quantum chemical calculations on model systems.

Although the metal is not directly involved in hydrogen bonding, the results of theoretical studies show that the nature of metal atom has significant influence on the strength of hydrogen bonds of ligands.

[1] T. Steiner, Angew. Chem. Int. Ed., 2002, 48-76.

[2] G. V. Janjić, M. D. Milosavljević, D. Ž. Veljković, S. D. Zarić, Phys. Chem. Chem. Phys, 2017, 8657-8660.

Molecular crystals with structural phase transitions are expected as novel materials for actuators and sensors. The design of crystals with phase transitions has been challenged by experimental and theoretical approaches. However, the former approach requires huge time and cost for experiments, and the latter requires high computational accuracy due to the high degree of freedom of molecular conformation. In recent years, the inductive approach to obtain knowledge from known data has been attracting attention as materials informatics. In this study, the intermolecular interactions in the (S)-N-3,5-di-tert-butylsalicylidene-1-(1-naphthyl)ethylamine (enol-(S)-1) crystals were analyzed, and we found new insights on structural phase transitions.

The enol-(S)-1 crystal has three crystal phases, namely α (<-80°C), β (-80~40°C), and γ (<40°C) phases, which are reversible through single-crystal-to-single-crystal phase transition depending on temperature change[1]. First, the lattice energies were calculated at each temperature point of the enol-(S)-1 crystals to elucidate the whole strength of intermolecular interactions. The lattice energies showed a discontinuous temperature dependence according to the phase transition, with a slight decrease at the phase transition from α to β phase and then an increase with rising temperature (Fig.1a). Then, the analysis of intermolecular interactions by Hirshfeld surfaces and 2D fingerprint plots was performed to reveal which interaction has contribution on the phase transitions. The Hirshfeld analysis uncovered that the proportion of π···π interactions decreased while the proportion of CH···π interactions increased at the phase transition from α to β phase. In addition, while the proportion of π···π interaction did not change, the proportion of CH···π interaction decreased in the β phase as the temperature rose but increased slightly after the transition from the β to γ phase. As to interaction energies, it was shown that the intermolecular interaction of CH···π stabilized at the transitions from α to β phase and from β to γ phase (Fig.1b). CH···π interactions had the unique temperature dependence compared to other main interactions of π···π and CH···π interactions. These results clarify the contribution of CH···π interaction to the stability of the high-temperature crystal phases and may provide new insights for designing crystals with phase transitions.

The development of new procedures to refine experimental diffraction data lead to an increased number of individual software packages to perform these analyses. They may require setup of specific input files, learning of configuration files, and sometimes result in file types unique to each package, which can make comparisons, changes between methods, and the overall workflow time consuming and only available to a trained specialist. NoSpherA2 – Non-Spherical-Atoms-in-Olex2 [1] – provides a possibility to interface any type and source of atomic form factors to the refinement engine olex2.refine [2], itself based on the cctbx [3].

This interface makes it possible to combine any level of sophistication in the calculation of the form factors, ranging from tabulated spherical atoms to tailor-made form factors from quantum mechanical calculations with the established refinement engine. Restraints, constraints, disorder modeling, solvent masking, and intuitive handling using the well-known Graphical-User-Interface of Olex2 [4] are the main advantages, which in combination with easy selection of options, automatic completion of CIFs, and no required manual handling of input files make the treatment of diffraction data using non-spherical models easier than ever.

While NoSpherA2 provides a variety of possibilities and generally better results of refinements (compare Figure 1), some questions about handling of various structures arise: Is it possible to mix-and-match different approaches? How to handle network compounds like MOFs and inorganic Salts? If we describe the non-spherical density distribution of atoms, what information possibly left in the data might need improved treatment? What resolution is required to use NoSpherA2?

The benefits, common practices, and strategies to tackle problems when using NoSpherA2 will be presented with examples and the philosophy of the development: Making the best-suited model available for the refinement task to obtain the best possible results without the need of individual file-conversion, in-depth training or specialized extra software.

[1] Kleemiss, F., Dolomanov, O. V., Bodensteiner, M., Peyerimhoff, N., Midgley, L., Bourhis, L. J., Genoni, A., Malaspina, L. A., Jayatilaka, D., Spencer, J. L., White, F., Grundkötter-Stock, B., Steinhauer, S., Lentz, D., Puschmann, H., Grabowsky, S. (2021), Chem. Sci. 12, 1675-1692. [2] Bourhis, L. J., Dolomanov, O. V., Gildea, R. J., Howard, J. A. K., Puschmann, H. (2015), Acta Cryst. A71, 59-75. [3] Grosse-Kunstleve, R.W., Sauter, N.K., Moriarty, N. W., Adams, P. D. (2002), J. Appl. Cryst. 35, 126-136. [4] Dolomanov, O. V., Bourhis, L. J., Gildea, R. J., Howard, J. A. K., Puschmann, H. (2009), J. Appl. Cryst. 42, 339-341.

Analysis of charge density distribution is a powerful method to recover information about non-covalent interactions in crystals and such important physical quantities as lattice energies. These quantities allow evaluation of the mutual stability of polymorphs and comparison of the strength of supramolecular associates in solids that is valuable for crystal engineering. Moreover, in the case of compounds that serve as active pharmaceutical ingredients (API) the energies of individual intermolecular interactions and lattice energies can be associated with the energies of ligand-receptor binding.

Herein we present the results of experimental charge density studies, quantum chemical calculations and Voronoi partitioning for several APIs (abiraterone acetate [1], bicalutamide [2] and lamivudine) used in common practice to treat tumors and HIV infection. As result the energies of individual interatomic interactions were evaluated for single crystals of API and simplified models describing ligand-receptor interaction constructed using PDB data as starting points. The characterization of intermolecular interactions was carried out with a variety of theoretical approaches including deformation electron density, QTAIM theory, NCI method, molecular electrostatic potential and solid bond angles (Fig. 1). The data on intermolecular interaction obtained for single crystals and models of ligand-receptor binding demonstrated the similarity of lattice energy values with those for the energies of interactions between API and receptor despite of conformational changes.

The analysis of experimental electron density and Voronoi analysis of intermolecular bonding of compounds studied was financially supported by the Russian Science Foundation (project 20-13-00241).

There are already several examples of successful potential antineoplasic copper complexes which have completed pre-clinical trials, such as Casiopeína-III-ia [1] with copper(II) and HydroCuP® [2] with copper(I), that show high selectivity towards cancer cells. This work presents some of the structure-property relationships that we have determined in different potential antitumor agents during the last years in the laboratory. We focused particularly in intermolecular interactions in the crystal structure and their relationship with the lipophilicity of the compounds and the extent and mode of interaction with DNA, considered one of the main target biomolecule for complexes with planar aromatic ligands.

In the study of copper(I) ternary complexes with diimine and triphenylphosphine ligands there is a relationship between the percentage of polar interactions seen in the 2D fingerprint plot derived from Hirshfeld surface analysis and the experimentally determined lipophilicity [3]. Heteroleptic copper(II)-neocuproine complexes were also studied using L-dipeptides as co-ligands to regulate the lipophilicity of the obtained complexes and their ability to interact with DNA (Fig. 1). Studies confirmed that the complex with the most C-C interactions is the only one that interacts with DNA through partial intercalation, whereas the rest interact through groove-binding [4]. We are currently working in complex-model membrane interactions through an interdisciplinary approach that includes the use of full interaction maps to aid the understanding of the experimental and molecular modelling docking results.

This work was supported by: FAPESP, CAPES (Brazil), CSIC and PEDECIBA (Uruguay).

Drug-drug cocrystals involve the combination of two or more active pharmaceutical ingredients (API) with their original chemical characteristics maintained without breaking or forming new covalent bonds, thus ensuring its effectiveness.[1] Its pharmaceutical properties are determined by the polarity of functional groups, the electrostatic potential and the available intermolecular interactions, which in turn are characterized by the three-dimensional crystalline arrangement and governed by its molecular electronic structure.[2] These molecular electron properties and their relationships with the charge density topology can be analysed by experimental and theoretical studies.

In this manner, the experimental charge density analysis of the pharmaceutical drug-drug cocrystal involving the antimetabolite prodrug 5-Fluorocytosine (5-FC) and the tuberculostatic drug Isoniazid (INH), named 5FC-INH, [3] has been carried out based on the Hansen & Coppens aspherical multipolar model refinement,[4] using high resolution X-ray diffraction data at low temperature ((sin(θ_max)/λ=1.15 Å^-1, 150K). The experimental model was compared with those derived from corresponding theoretical calculations for solid-state and gas-phase conditions using density functional theory (DFT) methods at the B3LYP6-311++G** level of theory.[5] The detailed study of the molecular electron density, its corresponding topology and charge distribution were based on the quantum theory of atom in molecules (QTAIM).[6] The charge density distribution and analysis of topological properties revealed that the C—F bond may have a transit closed-shell configuration (Fig. 1). This analysis also allowed to verify the charge delocalization due to resonance-assisted hydrogen bond (RAHB) in the formation of the heterosynthon that stabilizes the crystal packing.[7]

Solvatochromism, thermochromism and photochromism are just a few examples of phenomena involving stimulated colour change. The external factors influencing absorption can be physical and/or chemical and might lead to variations of materials hue as well as cause limitations of light transmission [1]. Those effects find multiple applications in photochromic lenses, as smart self-dimming windows, paints and indicators, thermal papers, visual displays or biochemical probes. It is of particular interest for those processes to be controllable (selective absorption) and reversible. To achieve this goal it is necessary to gain more knowledge on the origin of those effects.

Recently we have studied (pseudo)polymorphs of tyraminium violurate showing both crystallochromic and solvatochromic effects [2]. We have deduced the origin of colour for each of the three phases using a set of quantum crystallography tools. The fluctuations in the electron density within the oxime group of violurate ions (target) was proven to be one of the factors influencing the absorption. This research forced us to think outside the box and formulate more general guidance criteria on how to design new chromic materials based on a common target molecule (chromogen) and changing selectively co-formers.

In this work, we have engineered a series of chromic materials based on violuric acid and its derivatives. We have chosen co-formers e.g. aromatic, aliphatic amines or pyridine derivatives to obtain a group of distinctly coloured products. The obtained crystal phases were further analysed using UV-vis and NMR spectroscopy as well as XRD and hot stage microscopy.

QTAIM analysis [3,4] in combination with H¹ NMR has enabled us to formulate a more general mechanism of colour generation in the violurate family. The possible phase transitions and influence of the temperature on the colour change in the solid-state were examined using hot stage microscopy and PXRD. The obtained results will contribute to a better understanding of chromic effects in the solid-state as well as in solution. Uncovering the origin of colour in one family of chromogens enables us to influence the absorption of a material by means of cocrystallization.

[1] Bamfield, P., Hutchings, M. (2018) Chromic Phenomena: Technological Applications of Colour Chemistry, Royal Society of Chemistry. [2] Gryl, M., Rydz, A., Wojnarska, J., Krawczuk, A., Koziel, M., Seidler, T., Ostrowska, K., Marzec, M. & Stadnicka, K. M. (2019). IUCrJ 6,
226-237. [3] Bader, R. F. W. (2003). Atoms in Molecules: A Quantum Theory, International Series of Monographs on Chemistry, Vol. 22. Oxford: Clarendon Press. [4] AIMAll (Version 19.10.12), Todd A. Keith, TK Gristmill Software, Overland Park KS, USA, 2019 (aim.tkgristmill.com)

This research was supported by National Science Centre Poland, grant number UMO-2018/30/E/ST5/00638

Intermolecular interactions are very crucial point to understand exhaustively in the part of rational drug design. Interestingly, the electronic level information of these intermolecular interactions is not possible to determine without high resolution x-ray diffraction measurements. However, this measurement is more familiar for small molecules and still demand to challenge the protein-ligand complexes. The design of drug with improved physical and chemical properties are major driving forces in the medicinal chemistry. To achieve this, quantum crystallographic approach helps to estimate the stability of interactions obtained from the ligand molecule with their target amino acid residues. Indeed, recent methodology development reports [ref] helped us to study as well as compute intermolecular interaction energies of protein-ligand complexes. Therefore, the present study is mainly focused to determine the different type of interactions between protein and ligand at electronic level. To accomplish this, the desirable protein-ligand complexes like enzyme-drug, enzyme-inhibitor and metal proteins-inhibitor complexes were subjected to QM/MM calculation followed by quantum theory of atoms in molecules (QTAIM) analysis which helps to understand the strength of intermolecular interaction and charge density distribution of protein-ligand complexes and these results compared with already reported experimental results. Electron density, Laplacian of electron density and hydrogen bond dissociation energy of metal interactions are very higher than the other interactions which confirms that the metal coordination is partially covalent bond. Hirshfeld surface analysis along with subsequent fingerprint maps were plotted to quantify the intermolecular contacts between ligand and amino acid residues. Non-Covalent Interaction (NCI) analysis has proved method for the qualitative analysis of hydrogen bonds which plays a crucial role and the accurate NCI energies of these bonds are essential to understand the binding mechanism in the formation drug-receptor complexes. NCI isosurface map of intermolecular interactions of protein-ligand complex clearly visualized the strong and weak interactions. Therefore, the quantum crystallographic based interaction energy calculation is a better alternative to docking score-based modelling. The results will be discussed at the time presentation.

In recent years, Single Molecule Magnets (SMMs) have gained significant attention, primarily due to their potential technological applications in the field of information storage and processing. SMMs are molecules that behaves as small nanomagnets and represents the smallest possible bit that can be used to store binary information. A common method to increase the total spin of such molecule is by engaging several metal centers in a strong ferromagnetic coupling. This strategy is applied in a recent study [1] of the tetranuclear transition metal compounds with formulas M₄(NP^tBu₃)₄ (M = Ni, Cu), and the oxidised forms, [M₄(NP^tBu₃)₄]⁺. These results suggest a strongly coupled, large-spin ground state in the two nickel compounds.

The two non-oxidised compounds, Ni₄(NP^tBu₃)₄ and Cu₄(NP^tBu₃)₄, are studied here with respect to the bonding interactions between the metal centers. X-ray diffraction experiments have been performed on crystals of both compounds at the synchrotron facility SPring-8 in Japan. Based on the data from the experiments, a multipole model of the charge density is achieved for both complexes. Theoretical structure factors are also calculated based on the experimental atomic positions of the complex containing nickel, and a theoretical multipole model is also developed.

A topological analysis is performed on all three datasets. The plots of the critical points in the charge density show no bond critical points between the metal centers as shown on Fig. 1. This indicates that no bonding is present between the metal centers, which is supported by the results from deformation density plots such as the one in Fig. 2. Plots of the critical points in the Laplacian for the nickel containing complex show that the charge density around the nickel atoms are affected by one another but not strongly, which is supported by calculations of the delocalization index.

Tautomeric systems such as keto-enol and enamine-imine incorporating both donor and acceptor in the same molecule, can be utilised to design single-component molecular ferroelectrics. This type of prototropic system reinforce intra or intermolecular hydrogen bonds synergistically with the augmentation of the π-system delocalization, defined as RAHB.¹ The ferroelectric systems developed utilizing the aforementioned phenomenon are defined as proton tautomerizarion type ferroelectrics and the corresponding mechanism is known as proton tautomerism mechanism (PTM).² However, this mechanism was never probed from electron density distributions. Here, we report the charge density analysis of 2-(4-(trifluoromethyl)phenyl)-1H-phenanthro [9,10-d] imidazole (1), the single-component molecular ferroelectric material (Figure 1).³ For elucidating the PTM in this molecular crystal, we have performed multipole model⁴ based experimental charge density analysis using high-resolution X-ray diffraction data. Thereby, we studied the deformation of electron densities (ED), the bond paths along with the bond critical points and the electrostatic potential (ESP) map along the N-H…N hydrogen bonds and performed topological analysis of the electron densities for understanding the underlying mechanism behind the proton transfer. The analysis also highlights the amphoteric characteristic of the enamine-imine unit (Figure 1). The different degrees of polarization of the electron densities of the N-H group and the N-atom of the adjacent molecule clearly supports the asymmetric N-H…N hydrogen bond characteristic in this displacive type ferroelectric. Such accurate analysis of multipole based electron densities further strengthen the understanding of the proton tautomerism effect in hydrogen bonded molecular ferroelectrics. To the best of our knowledge, this is the first report of charge density analysis on a molecular ferroelectric material. Our study points to the necessity of charge density analysis for improved understanding in this niche area of molecular ferroelectric research.

Figure 1: (left) P-E loop and remanent polarization and (right) PTM pathway highlighting in terms of ED and ESP in 1

[1] P. Gilli, V. Bertolasi, V. Ferretti, G. Gilli, J. Am. Chem. Soc. 1994, 116 (3), 909–915.

[2] (a) S. Horiuchi, K. Kobayashi, R. Kumai, S. Ishibashi, Nat. Commun. 2017, 8, 1–9. (b) S. Dutta, V. Vikas, A. Yadav, R. Boomishankar, A. Bala, V. Kumar, T. Chakraborty, S. Elizabeth and P. Munshi, Chem. Commun., 2019, 55, 9610–9613.

[3] S. Dutta, Vikas, T. Vijayakanth and P. Munshi, Ferroelectric and Negative Thermal Expansion Properties in a purely organic multifunctional material. 2021 (to be published).

[4] N. K. Hansen and P. Coppens, Acta Crystallogr. Sect. A 1978, 34 (6), 909–921.

In-situ crystallization is a powerful tool to obtain the structure of compounds that are liquid at room temperature [1]. The technique involves cooling the sample in a glass capillary below its liquid-solid phase transition temperature, initializing crystallization and using the crystalline powder so obtained as starting material for crystal growth. In favorable cases, a crystal suitable for X-ray analysis can be obtained at the liquid-solid phase boundary by translation of the capillary through the cold nitrogen gas stream (zone melting). If the quality of the in-situ grown crystal is sufficiently good, experimental charge density studies (resolution up to 0.5 Å) become possible. Nowadays there are many models used to describe the experimental diffraction intensities (IAM, multipole model, maximum entropy method etc.) [2]. With the appearance of quantum crystallographic methods (Hirshfeld Atom Refinement HAR and X-Ray Constrained Wavefunction analysis XCW) that combine the experimental results with quantum mechanical calculations, it is possible to get precise insights into the "true" nature of the wavefunction and the charge density of a molecule [3]. The charge density of a molecule can also be described as a superposition of theoretical calculated Invariom (Invariant atom) or directly by the theoretically calculated charge density [4]. But which of these methods is best suited for which scientific question or experimental resolution? Different charge density models, such as the multipole-model, HAR and the Invariom-approach, were investigated using experimental diffraction intensities of different in-situ crystallized liquids. The residual density in difference Fourier maps as used to evaluate the quality of the theoretical model (Fig. 1) [5]. It was found that multipole parameters derived from fitting to theoretically calculated charge density describe the experiment almost as well as unrestricted multipole refinement. In addition, using theoretical parameters the reflection to parameter ratio is improved.