Niobium oxyiodide cluster compounds Li₃Nb₇O₅I₁₅ and Nb₈O₅I₁₇(NbI₅) with expanding cluster architectures and multicentre Nb–Nb bonding

Abstract

A series of niobium oxyiodide compounds has recently been identified using a non-conventional reduction method. The continuation of these studies of heterogeneous solid-state reactions in a closed system has led to the crystallization and structural analysis of two novel compounds Li₃Nb₇O₅I₁₅ and Nb₈O₅I₁₇(NbI₅). Both crystal structures are derived from the pentanuclear [Nb₅O₄] cluster core and are expanded through the incorporation of additional niobium atoms, forming new [Nb₇O₅] and [Nb₈O₅] cluster cores. Furthermore, it is shown that oxyiodide clusters containing between four and eight niobium atoms can form from the same mixture of reactants by simply adjusting the composition and temperature. A comprehensive assembly model for these clusters is presented, and electronic structure calculations provide insight into the nature of niobium–niobium bonding within the [Nb₇O₅] core.

---

Introduction

Metal-rich halide cluster compounds of group 5 and 6 transition metals are well known from the research initiated by the groups of Schäfer, von Schnering and Simon.¹ Preparations of these compounds are typically performed by metallothermic reduction, in which a metal halide (MX_n) is reacted with metal (M), which can be the same metal (M) or an electropositive alkali metal or Al.² Attempts to employ non-conventional reduction agents have been demonstrated to provide insights into the reduction pathway (namely through metal intercalation followed by metal halide elimination) by using non-conventional reduction agents, such as Cr, Fe, Co, or Ni, etc., assisted by thermal scanning.³ Recently, we have successfully employed lithium carbodiimide as a reduction agent, as will be explained subsequently.⁴

The vast majority of transition metal cluster compounds is based on an octahedral cluster core of metal (M) atoms which most commonly appear with [M₆X₈] with eight face-capping X or [M₆X₁₂] with twelve edge-capping X halide (X) environments.⁵ In addition, interstitially (Z) stabilized octahedral clusters of the type [M₆ZX₁₂] are reported for electron-poor group 4 metals.⁶

More than twenty binary tungsten iodide compounds have been discovered and structurally characterized; most of them are based on the [M₆X₈] type cluster.⁷ In contrast, binary niobium iodide compounds involve only Nb₆I₁₁, Nb₃I₈, NbI₃, and NbI₄.⁸ An interesting extension of the niobium halide chemistry has been achieved by the employment of an extra non-metal element, leading to heteroanionic cluster compounds, which may provide similar or completely distinct cluster architectures. Thus, heteroanionic clusters can be envisioned to constitute anion replacements in given architectures and to induce a modified structure or connectivity pattern of clusters, as exemplified for Nb₆I₉S (derived from Nb₆I₁₁), Nb₃X₇S (X = Cl, Br, I), and ANb₃Br₇S with A = Rb, Cs (both derived form Nb₃X₈).⁹ Different cluster architectures are obtained in compounds with chalcogenides, such as Nb₃TeI₇, Nb₄Se₄I₄, Nb₇S₂I₁₉, and with pnictide anions in Nb₄PnX₁₁ with Pn = N, P, X = Cl, Br, I.^9b,10 Beside the difference in the number of Nb atoms forming the cluster core, the cluster design and connectivity are showing a great variety and so do the properties.

Niobium oxyiodide compounds are reported for NbOI₃, NbO₂I, and the metal-rich example NbOI₂.¹¹ Condensed metal–oxide-halides volatilize easily and can be crystallized only within a specific equilibrium window. This observation highlights the sensitivity of these systems to changes in conditions, such as temperature and composition, which can significantly impact the formation of different niobium oxyiodide phases.

The Nb–O–I system was recently a subject of investigations with a focus on metal-rich compounds by employing Li₂(CN₂) as a reducing agent. The reaction, i.e. the reduction, involves the decomposition of the (NCN)²⁻ ion into C₃N₄, as evidenced by IR spectroscopy.^4b Previous reactions have shown, that the interplay between NbI₄, Li₂(CN₂) and Li₂O can form a heterogeneous system in which local temperatures have an important impact on the phase formation. As a result, we have already reported the compounds Nb₄OI₁₂, Nb₄OI₁₁, and Nb₄OI₁₀ with their crystal structures, all based on the rectangular [Nb₄O] core.^4a With decreasing iodide content, the series of Nb₄OI_12−x compounds is showing structures with increasing connectivity between clusters, increasing electrical conductivities, and an increasing number of electrons from six (x = 2) to eight electrons (x = 0) per cluster. Furthermore, we have conducted a detailed analysis of the properties of oxyiodides of the type M₅O₄I₁₁, where M = Nb or Ta.¹²

In continuation of this research, we now present two additional niobium oxyiodides, Li₃Nb₇O₅I₁₅ and Nb₈O₅I₁₇(NbI₅), featuring [Nb₇O₅] and [Nb₈O₅] cluster cores, each accommodating 13 electrons available for niobium–niobium bonding.

Results and discussion

Synthesis and crystal structure

In our synthesis, NbI₄ was reacted with Li₂(CN₂) and Li₂O. The specific combination of these reactants (see Experimental section) constitutes a heterogeneous reaction system governed by temperature-dependent equilibria, leading to the formation of various solid and gaseous phases. Within this system, Li₂(CN₂) is proposed to function as an unconventional reducing agent for NbI₄, thereby undergoing decomposition of the (CN₂)²⁻ anion to form C₃N₄.

To gain insight into the reaction mechanism, we performed a series of reactions at different temperatures and durations. Heating the mixture to 300 °C yielded an unknown crystalline phase, along with LiI, NbOI₂, and an amorphous powder, as determined by powder X-ray diffraction (PXRD) (see Fig. S1, SI). Upon heating at 400 °C for 1 hour, the unidentified phase disappeared, and the compound Li₃Nb₇O₅I₁₅ was formed, accompanied by NbOI₂ and NbI₅. Prolonged heating at 400 °C for 24 hours resulted in the crystallization of the known Nb₅O₄I₁₁, which contains the [Nb₅O₄] cluster core.¹²

Further heating at 500 °C for 1 hour produced crystals characterized as Nb₈O₅I₁₇(NbI_5.37), which were found on the walls of the ampoule. Extending the heating time at 500 °C to 24 hours led to the formation of Nb₄OI₁₀, a previously reported compound featuring a [Nb₄O] cluster core.^4a

These results, while focused on solid-state products, do not account for the role of gaseous species that significantly influence the formation pathways of the observed materials. A quenching experiment of the gas phase from 400 °C to room temperature revealed the presence of a black fluid that appeared as small droplets on the ampoule walls (Fig. S2a). After 48 hours at ambient temperature, these droplets transformed into dark red, filamentous crystallites (Fig. S2b), which subsequently decomposed over time (Fig. S2c).

The two new compounds presented in this study, Li₃Nb₇O₅I₁₅ and Nb₈O₅I₁₇(NbI_5.37), were structurally characterized by single-crystal X-ray diffraction (see Table 1). Li₃Nb₇O₅I₁₅ crystallizes as rod-like crystals, whereas Nb₈O₅I₁₇(NbI_5.37) forms plate-like crystals. Several single crystals of the latter were refined, and in each case, the iodide content of the enclosed NbI₅ molecules were found to have slightly more than five iodide atoms, associated with the presence of some polyiodide. For clarity and simplicity, this compound will hereafter be referred to as Nb₈O₅I₁₇(NbI₅).

Table 1 Crystallographic and refinement data of Li₃Nb₇O₅I₁₅ and Nb₈O₅I₁₇(NbI₅)

	Li3Nb7O5I15	Nb8O5I17(NbI5.37)
CCDC	2311571	2428720
Space group	Cmcm	Pnma
Temperature/K	230	200
Unit cell dimensions/Å	a = 14.5783(3)	a = 11.4282(4)
	b = 17.6445(3)	b = 13.4336(6)
	c = 13.5823(2)	c = 31.479(1)
Volume/Å^3	3493.7(1)	4848.1(3)
Z	4	4
Wavelength/Å	1.54184	1.54184
μ/mm^−1	122.292	128.729
2θ range for data collection	7.866 to 136.454	5.598 to 133.306
Total number of reflections	17 730	47 074
Independent reflections	1742	4437
Refined parameters	85	240
Rint	0.0316	0.0589
R1	0.0197	0.0274
wR2	0.0534	0.0516
Goodness-of-fit on F^2	1.060	1.036

---

The crystal structure of Li₃Nb₇O₅I₁₅ features an oxygen-stabilized Nb₇ cluster arrangement, in which a central niobium atom (Nb2) is surrounded by six additional niobium atoms (Nb1 and Nb3), forming a trigonal prismatic coordination environment around Nb2 (Fig. 1).

Fig. 1 Structure of the [Nb₇O₅] cluster core inside Li₃Nb₇O₅I₁₅.

Cluster cores composed of seven metal atoms are relatively rare and often adopt significantly different geometries compared to the [Nb₇O₅] cluster core observed in Li₃Nb₇O₅I₁₅. For instance, in [N(PPh₃)₂][Os₆Au(CO)₂₀H₂], the [Os₆Au] cluster is described as two butterfly-shaped units sharing a wing tip.¹³ Another example is the asymmetric [MoFe₆] cluster core in [Ph₄As]₂[MoFe₆S₆(CO)₁₆].¹⁴ Octahedral M₆ clusters containing a central transition-metal atom are also known, particularly among rare-earth and zirconium-based compounds.¹⁵

In contrast, the [Nb₇O₅] cluster core can be regarded as an extension to the [M₅O₄] cluster core found in the M₅O₄I₁₁ compounds (M = Nb, Ta) through addition of three more atoms, namely two Nb3 atoms and the O3 ion depicted in Fig. 1. The [Nb₇O₅] cluster core represents six short Nb–Nb distances between the central Nb2 with the surrounding six niobium atoms measuring 2.7883(7) Å (Nb2–Nb3) and 2.8050(4) Å (Nb1–Nb2). These distances are significantly shorter than the Nb1 and Nb3 separation, which measure 3.0667(6) Å. The cluster is surrounded by iodide ligands, including four edge-capping ligands, four terminal ligands, and twelve iodide ligands that bridge between adjacent cluster cores in a 12/2 fashion, as illustrated in Fig. 2. Using the notation originally developed for octahedral cluster compounds, distinguishing between inner (i), outer (a), and outer-bridging (a–a) ligand functionalities, the connectivity pattern of the cluster can be described as (Nb₇O₅I₄ⁱ)I₄^aI_12/2^a–a.

Fig. 2 Iodide coordination of the [Nb₇O₅] cluster core, with twelve iodide atoms having bridging functionality (indicated by dashed lines). Niobium atoms are coloured in blue, oxygen in red and iodine in pink.

This cluster fragment is further stabilized by three lithium ions and one additional iodide ion, completing the overall formula Li₃Nb₇O₅I₁₅. Four outer iodide ligands from the cluster, along with the additional iodide ion (I5), surround each lithium ion (Li1 and Li2) in a square-pyramidal fashion (see Fig. 3). The less flexible cluster-bound iodide ligands (I2 and I3) form the base of the square pyramid, with Li–I distances ranging from 2.904(6) to 3.17(1) Å, while the more flexible additional iodide ion (I5) occupies the apical position, showing Li–I distances of 2.64(2) Å and 2.76(1) Å.

Fig. 3 Square-pyramidal lithium (grey) coordination in the structure with the additional iodide atom (I5).

The complete structure of Li₃Nb₇O₅I₁₅ is constructed by linking each [Nb₇O₅] cluster to six neighbouring clusters, forming a three-dimensional framework. As illustrated in Fig. 4, a view along the c-axis of the structure reveals an alternating orientation of the adjacent clusters, highlighting the spatial arrangement and connectivity within the extended structure.

Fig. 4 Projection of the crystal structure of Li₃Nb₇O₅I₁₅, looking along the c-axis. Niobium atoms are shown in blue, oxygen in red, iodine in pink and lithium in grey.

The second cluster compound obtained from the same basic reaction is Nb₈O₅I₁₇(NbI₅). The [Nb₈O₅] cluster core contained within this compound can be viewed as an extension of the [Nb₇O₅] core found in Li₃Nb₇O₅I₁₅, incorporating just one additional niobium atom. As shown in Fig. 5, this extension occurs through the addition of a niobium atom (Nb5) positioned at one edge of the [M₅O₄] base, effectively expanding the cluster architecture.

Fig. 5 The [Nb₈O₅] cluster core of Nb₈O₅I₁₇(NbI₅).

As observed for the previously discussed [Nb₇O₅] cluster core, the Nb–Nb distances between the central Nb2 atom and the six surrounding niobium atoms (Nb1, Nb3, Nb4) in Nb₈O₅I₁₇(NbI₅) are relatively short, ranging from 2.7911(3) to 2.8600(7) Å. In contrast, the Nb–Nb distances between Nb3–Nb1 and Nb3–Nb4 are notably longer, falling in the range of 2.9981(6) to 3.1812(7) Å. The Nb–Nb distance involving the additional Nb5 atom is with 3.1418(7) Å in the same range. A comprehensive comparison of Nb–O bond lengths for both compounds is provided in Table 2. This comparison shows a good overall agreement between the structural metrics observed in the two cluster compounds.

Table 2 Nb–O distances in the cluster cores of the two compounds and the average Nb–O distances with the corresponding coordination number (CN) of the oxygen

Compound distance/Å	Li3Nb7O5I15	Nb8O5I17(NbI5)
Nb1–O1	2.0005(9)	1.958(5)
Nb1–O2	1.9568(5)	1.9564(6)
Nb2–O1	2.197(6)	2.205(6)
Nb2–O2	2.097(6)	2.093(8)
Nb2–O3	2.080(8)	2.079(7)
Nb2–O4		2.084(7)
Nb3–O1	2.247(6)	2.233(5)
Nb3–O3	1.9464(6)	1.9415(7)
Nb4–O1		2.010(5)
Nb4–O4		2.122(1)
Nb5–O4		2.073(7)
∅ Nb–O1 (CN = 4)	2.11(1)	2.10(1)
∅ Nb–O2 (CN = 3)	2.00(1)	2.00(1)
∅ Nb–O3 (CN = 3)	1.99(1)	1.99(1)
∅ Nb–O4 (CN = 4)		2.10(1)

---

The presence of the additional niobium atom in the structure Nb₈O₅I₁₇(NbI₅) alters, both the local environment and the connectivity pattern of the cluster. The connectivity changes from (Nb₇O₅I₄ⁱ)I₄^aI_12/2^a–a in Li₃Nb₇O₅I₁₅ to (Nb₈O₅I₈ⁱ)I₅^aI_8/2^a–a) in Nb₈O₅I₁₇(NbI₅) (Fig. 6). As a result, the number of connections to neighboring clusters is reduced to just four, leading to a dimensional reduction in the overall connectivity, from a three-dimensional framework in Li₃Nb₇O₅I₁₅ to a layered arrangement in Nb₈O₅I₁₇(NbI₅) with zig-zag layers extending along the ab-plane (Fig. 7).

Fig. 6 Iodide coordination of the [Nb₈O₅] cluster core with eight iodide atoms with bridging functionality (indicated by dashed lines) connecting to four adjacent clusters. Niobium atoms are coloured in blue, oxygen in red and iodine in pink.

Fig. 7 Projected structure with the Nb₈O₅I₁₇ section of Nb₈O₅I₁₇(NbI₅). The NbI₅ molecules between the layers are omitted for clarity. Niobium atoms are shown in blue, oxygen in red and iodine in pink.

Between the layers of the Nb₈O₅I₁₇(NbI₅) structure, channels are formed that accommodate NbI₅ units (Fig. S3). The observed disorder in these regions arises from the partial incorporation of excess iodine, likely in the form of coordinated polyiodide species such as I₂⁻ or I₃⁻. These species preferentially align along the crystallographic a-axis, contributing to the structural complexity and variability of the interlayer region.

The intercalation of molecular species within a solid-state framework is relatively uncommon but well-documented in the literature. Such molecules can reside in interlayer spaces or structural channels when the host framework provides cavities large enough to accommodate them. A notable example is the crystal structure of Nb₇S₂I₁₉, which can be described as (Nb₃SI₇)₂(NbI₅).^10d In this compound, six [Nb₃S] units form iodide-bridged ring arrangements that create central channels occupied by NbI₅ molecules. Another example is Ta₅O₄I₁₁(TaI₅), where TaI₅ molecules are intercalated between corrugated layers.¹² In this case, TaI₅ acts as a structural template for layer formation, as a similar cluster connectivity is observed in m-Ta₅O₄I₁₁. A comparable template effect is reported in [K₅(Ti₂Cl₉)][(Nb₆Cl₁₂O₄)₃(Ti₃Cl₄)₂],¹⁶ where the complex anion [(Nb₆Cl₁₂O₄)₃(Ti₃Cl₄)₂]²⁻ forms a layered framework with large channels, which are filled by [K₅(Ti₂Cl₉)]²⁺ units. A related scenario is found in Cs₃(ZrCl₅)Zr₆Cl₁₅Mn, where ZrCl₅⁻ units are incorporated into a three-dimensional framework constructed from [Zr₆(Mn)Cl₁₂ⁱ]Cl_6/2^a–a.^15a Such a monoanionic molecule can also be intercalated into a simple metal halide, as demonstrated by the compound (Ta₆Br₁₅)(TaBr₆)_0.86.¹⁷ These examples show how molecular species can be stabilized within solid-state matrices by occupying voids or acting as templates that influence framework formation.

Scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDX)

Scanning electron microscopy was performed with the aid of a custom-designed vacuum transfer device to transfer the samples under inert conditions. Fig. 8 shows some crystals of Li₃Nb₇O₅I₁₅, having a rod-like morphology with right-angled edges and a rough surface.

Fig. 8 Scanning electron micrographs of Li₃Nb₇O₅I₁₅.

To confirm the composition determined by X-ray diffraction of Li₃Nb₇O₅I₁₅, EDX measurements on multiple crystals were performed. These measurements result in an average Nb : I ratio of 7 : 15.2(6).

Electronic band structure and ELF calculations

The electronic structure of Li₃Nb₇O₅I₁₅ was investigated using density functional theory (DFT). The calculated band structure, shown in Fig. 9, reveals seven metal-centred energy bands located below the Fermi level. Among these, the highest-energy band intersects the Fermi level at multiple points in the Brillouin zone. These seven bands correspond to one formula unit of Li₃Nb₇O₅I₁₅ and primarily arise from niobium–niobium bonding interactions within the [Nb₇O₅] cluster, rather than in between clusters. In total, they accommodate 13 electrons per cluster. However, this number of electrons cannot be straightforwardly assigned to specific niobium–niobium contacts within the cluster core, especially when considering a notably large number of relatively short Nb–Nb distances within the cluster, discussed previously. The presence of a high density of bands at and above the Fermi level indicates delocalization of electrons across these bands. This suggests a more delocalized distribution of bonding interactions across the cluster core.

Fig. 9 DFT-calculated electronic band structure of Li₃Nb₇O₅I₁₅ with bands coloured by their niobium character. Special points in and paths through the Brillouin zone were chosen following the literature.¹⁸

Beside the reasonably high niobium content of the bands, there is also some content arising from oxygen orbitals in bands below the Fermi energy (Fig. S4). Hence, there seems to be oxygen states hybridizing with niobium states which leads to some covalent character of some Nb–O bonds. Electron localization function (ELF) calculations of the cluster core support this interpretation. As shown in (Fig. 10, left), there is notable electron density extending from the O1 atom into the Nb1–Nb3 bonding region, indicating participation of oxygen in metal–metal bonding. Furthermore, a pronounced region of electron density is observed between Nb1, Nb2, and Nb3 (Fig. 10, right), consistent with the presence of a three-centred bond. The ELF analysis suggests the presence of four such three-centred bonds, in addition to four Nb–Nb bonds with an admixture of electron density coming from oxygen within the cluster. These bonding interactions highlight the significant role of orbital mixing between niobium d-orbitals and oxygen p-orbitals in stabilizing the cluster core and the cluster framework.

Fig. 10 Electron localisation functions of the [Nb₇O₅] cluster core in Li₃Nb₇O₅I₁₅.

Conclusion

With the discovery of the niobium oxyiodide cluster compounds Li₃Nb₇O₅I₁₅ and Nb₈O₅I₁₇(NbI₅), the family of cluster architectures featuring [Nb_xO_y] cores is expanded by two new members. The smallest cluster in this series is the [Nb₄O] core in Nb₄OI_12−x, followed by the [Nb₅O₄] core in Nb₅O₄I₁₁, and now the [Nb₇O₅] and [Nb₈O₅] cores reported in the present work. These clusters are illustrated in Fig. 11.

Fig. 11 The expansion of cluster architectures of niobium oxide cluster cores in various niobium oxyiodides.

All compounds were synthesized via heterogeneous solid-state reactions between NbI₄ and Li₂(CN₂), which served as an unconventional reducing agent, in the presence of an oxide. The formation of the reported compounds occurs via temperature-dependent equilibria, influenced by slight variations in composition and the generation of various solid and gaseous phases (Fig. S5). Importantly, the gas phase and the specific gaseous species involved is thought to play a key role in guiding the formation pathways of the resulting cluster compounds.

Experimental section

Materials and methods

All manipulations of starting materials and products were performed in a glovebox under dry argon with moisture and oxygen levels below 1 ppm. Li₂O (ABCR, 95%) was used as purchased. NbI₄ and Li₂(CN₂) were synthesized as described in the literature.¹⁹

Synthesis

Li₃Nb₇O₅I₁₅ was synthesized from NbI₄, Li₂O and Li₂(CN₂). For this purpose, NbI₄ (160.8 mg, 0.268 mmol), Li₂O (2 mg, 0.067 mmol), and Li₂(CN₂) (7.2 mg, 0.135 mmol) were encapsulated into a fused silica ampoule with 3 cm length and a volume of about 1.5 cm³. The ampoule was heated in a box furnace from room temperature to 500 °C with a rate of 0.1 °C min⁻¹. The holding time was 24 h before the reaction was cooled down to room temperature with a rate of 2 °C min⁻¹. Rod-like crystals of Li₃Nb₇O₅I₁₅ were found at the bottom of the ampoule in the powder, beside LiI and an amorphous phase, produced by the decomposition of Li₂(CN₂). The product was washed with dried ethanol to get rid of the LiI. A powder pattern of pure product war recorded (Fig. S6). The estimated yield of Li₃Nb₇O₅I₁₅ was about 80%; crystals appear black and decompose in air due to moisture after some time.

Nb₈O₅I₁₇(NbI₅) was synthesized from NbI₄, Li₂O and Li₂(CN₂). For this purpose, NbI₄ (160.8 mg, 0.268 mmol), Li₂O (2 mg, 0.067 mmol) and Li₂(CN₂) (7.2 mg, 0.135 mmol) were encapsulated into a fused silica ampoule with 3 cm length and a volume of about 1.5 cm³. The ampoule was heated in a Simon-Müller furnace from room temperature to 500 °C with a rate of 0.1 °C. The holding time was 1 h before the reaction was cooled down to room temperature with a rate of 2 °C min⁻¹. Plate-like crystals of Nb₈O₅I₁₇(NbI₅) were found at the wall in the middle of the ampoule with NbOI₂ and NbI₅ at the top of the ampoule and LiI, Li₃Nb₇O₅I₁₅ and amorphous C₃N₄ in the powder. An X-ray powder pattern of pure product was recorded (Fig. S7). Crystals appear black and decompose in air due to moisture quite fast.

Crystallography

A rod-like Li₃Nb₇O₅I₁₅ and a plate-like Nb₈O₅I₁₇(NbI₅) single-crystal were mounted on a Rigaku XtaLab Synergy-S X-ray diffractometer using Cu-K_α (λ = 1.54184 Å) radiation. The single crystal was kept under N₂ cooling at 230 K during the data collection. Corrections for absorption effects were applied with CrysAlisPro 1.171.42.70a (Rigaku Oxford Diffraction, 2022). The crystal structure was solved by the integrated space group and crystal-structure determination routine of SHELXT²⁰ and full-matrix least-squares refinement with SHELXL-2018/3²⁰ implemented in Olex2 1.5.²¹

Reaction products were investigated by powder X-ray diffraction (PXRD) using a StadiP diffractometer (Stoe, Darmstadt) with Ge[111]-monochromated Cu-K_α1 radiation and a Mythen1 detector.

Density functional theory

Density functional theory (DFT) calculations were performed using the Abinit software package (v. 9).²² The Perdew–Burke–Ernzerhof exchange–correlation functional²³ was used with the vdw-DFT–D3(BJ) dispersion correction of Grimme.²⁴ A Monkhorst–Pack grid of k-points with real-space basis vectors (3 3 0) (−2 2 0) and (0 0 3) was used to construct the electronic structure of crystalline Li₃Nb₇O₅I₁₅.²⁵ Plane-wave calculations were performed using an energy cutoff of 40 Ha outside of the PAW spheres and a 120 Ha cutoff inside them. Pseudopotentials sets were used as received from the Abinit library.²² These computational parameters were chosen following convergence studies.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

The data that support the findings of this study are available on request from the corresponding author, H.-J. Meyer. Computational data are avaliable at https://doi.org/10.5281/zenodo.15366543.

Supplementary information is available. See DOI: https://doi.org/10.1039/d5dt01819f.

CCDC 2311571 (Li₃Nb₇O₅I₁₅) and 2428720 (Nb₈O₅I₁₇(NbI_5.37)) contains the supplementary crystallographic data for this paper.^26a,b

Acknowledgements

This research was supported by the Deutsche Forschungsgemeinschaft (ME 914-32/1) and the state of Baden-Württemberg through bwHPC and the German Research Foundation (DFG) through grant no INST 40/467-1 FUGG (JUSTUS cluster). C. P. R. acknowledges support from the project FerrMion of the Ministry of Education, Youth and Sports, Czech Republic, co-funded by the European Union (CZ.02.01.01/00/22\_008/0004591), the European Union and Horizon 2020 through grant no. 810451.

References

1. (a) H. Schäfer and H.-G. von Schnering, Angew. Chem., 1964, 76, 833–849; (b) A. Broll, A. Simon, H.-G. von Schnering and H. Schäfer, Z. Anorg. Allg. Chem., 1969, 367, 1–18.
2. G. Brauer, Handbuch der Präparativen Anorganischen Chemie, Ferdinand Enke Verlag, Stuttgart, 1981, vol. 3.
3. (a) A. Mos, C. Castro, S. Indris, M. Ströbele, R. F. Fink and H.-J. Meyer, Inorg. Chem., 2015, 54, 9826–9832; (b) A. Mos, M. Ströbele and H.-J. Meyer, Z. Anorg. Allg. Chem., 2015, 641, 1722–1727; (c) A. Mos, M. Ströbele and H.-J. Meyer, J. Cluster Sci., 2015, 26, 187–198; (d) A. Mos-Hummel, M. Ströbele and H.-J. Meyer, Eur. J. Inorg. Chem., 2016, 2016, 4234–4240.
4. (a) J. Beitlberger, M. Ströbele, F. Strauss, M. Scheele, C. P. Romao and H.-J. Meyer, Eur. J. Inorg. Chem., 2024, 27, e202400329; (b) J. Beitlberger, M. Martin, M. Scheele, P. Schmidt, M. Ströbele and H.-J. Meyer, Dalton Trans., 2025, 54, 5486–5493.
5. L. Pauling, The Nature of the Chemical Bond and the Structure of Molecules and Crystals: An Introduction to Modern Structural Chemistry, Cornell University Press, 1960.
6. R. P. Ziebarth and J. D. Corbett, J. Solid State Chem., 1989, 80, 56–67.
7. M. Ströbele and H.-J. Meyer, Dalton Trans., 2019, 48, 1547–1561.
8. (a) A. Simon and H.-G. von Schnering, J. Less-Common Met., 1966, 11, 31–46; (b) A. Simon, H.-G. von Schnering and H. Schäfer, Z. Anorg. Allg. Chem., 1967, 355, 295–310; (c) P. W. Seabaugh and J. D. Corbett, Inorg. Chem., 1965, 4, 176–181; (d) L. F. Dahl and D. L. Wampler, Acta Crystallogr., 1962, 15, 903–911.
9. (a) H.-J. Meyer and J. D. Corbett, Inorg. Chem., 1991, 30, 963–967; (b) P. J. Schmidt and G. Thiele, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1997, 53, 1743–1745; (c) G. V. Khvorykh, A. V. Shevelkov, V. A. Dolgikh and B. A. Popovkin, J. Solid State Chem., 1995, 120, 311–315; (d) C. M. Pasco, Dissertation, Johns Hopkins University, 2021; (e) F. Grahlow, F. Strauß, M. Scheele, M. Ströbele, A. Carta, S. F. Weber, S. Kroeker, C. P. Romao and H.-J. Meyer, Phys. Chem. Chem. Phys., 2024, 26, 11789–11797; (f) H.-J. Meyer, Z. Anorg. Allg. Chem., 1994, 620, 863–866.
10. (a) M. Ströbele, O. Oeckler, M. Thelen, R. F. Fink, A. Krishnamurthy, S. Kroeker and H.-J. Meyer, Inorg. Chem., 2022, 61, 17599–17608; (b) M. D. Smith and G. J. Miller, J. Alloys Compd., 1998, 281, 202–205; (c) H. B. Yaich, J. C. Jegaden, M. Potel, M. Sergent, A. K. Rastogi and R. Tournier, J. Less-Common Met., 1984, 102, 9–22; (d) G. J. Miller and J. Lin, Angew. Chem., Int. Ed. Engl., 1994, 33, 334–336.
11. (a) J. Rijnsdorp and F. Jellinek, J. Less-Common Met., 1978, 61, 79–82; (b) S. Hartwig, Dissertation, University of Bayreuth, 2003.
12. F. Grahlow, J. Beitlberger, M. Martin, E. Juriatti, H. Peisert, M. Scheele, M. Ströbele, C. P. Romao and H.-J. Meyer, unpublished work.
13. B. F. G. Johnson, D. A. Kaner, J. Lewis and P. R. Raithby, J. Chem. Soc., Chem. Commun., 1981, 753–755.
14. P. A. Eldredge, K. S. Bose, D. E. Barber, R. F. Bryan, E. Sinn, A. Rheingold and B. A. Averill, Inorg. Chem., 1991, 30, 2365–2375.
15. (a) J. Zhang and J. D. Corbett, Inorg. Chem., 1995, 34, 1652–1656; (b) T. Hughbanks and J. D. Corbett, Inorg. Chem., 1988, 27, 2022–2026; (c) J. Zhang and J. D. Corbett, Inorg. Chem., 1993, 32, 1566–1572.
16. T. Duraisamy, Y. Zhihua, E. V. Anokhina, C. S. Choi, C. S. Day and A. Lachgar, J. Solid State Chem., 2003, 175, 46–51.
17. K. Habermehl, A.-V. Mudring and G. Meyer, Eur. J. Inorg. Chem., 2010, 2010, 4075–4078.
18. Y. Hinuma, G. Pizzi, Y. Kumagai, F. Oba and I. Tanaka, Comput. Mater. Sci., 2017, 128, 140–184.
19. (a) G. Brauer, Handbuch der präparativen anorganischen Chemie, Enke, 1975; (b) R. Srinivasan, M. Ströbele and H.-J. Meyer, Inorg. Chem., 2003, 42, 3406–3411.
20. G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., 2015, 71, 3–8.
21. O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
22. X. Gonze, B. Amadon, G. Antonius, F. Arnardi, L. Baguet, J.-M. Beuken, J. Bieder, F. Bottin, J. Bouchet, E. Bousquet, N. Brouwer, F. Bruneval, G. Brunin, T. Cavignac, J.-B. Charraud, W. Chen, M. Côté, S. Cottenier, J. Denier, G. Geneste, P. Ghosez, M. Giantomassi, Y. Gillet, O. Gingras, D. R. Hamann, G. Hautier, X. He, N. Helbig, N. Holzwarth, Y. Jia, F. Jollet, W. Lafargue-Dit-Hauret, K. Lejaeghere, M. A. L. Marques, A. Martin, C. Martins, H. P. C. Miranda, F. Naccarato, K. Persson, G. Petretto, V. Planes, Y. Pouillon, S. Prokhorenko, F. Ricci, G.-M. Rignanese, A. H. Romero, M. M. Schmitt, M. Torrent, M. J. van Setten, B. Van Troeye, M. J. Verstraete, G. Zérah and J. W. Zwanziger, Comput. Phys. Commun., 2020, 248, 107042.
23. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
24. S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys., 2010, 132, 154104.
25. H. J. Monkhorst and J. D. Pack, Phys. Rev. B: Condens. Matter Mater. Phys., 1976, 13, 5188–5192.
26. (a) J. Beitlberger, M. Ströbele, P. Schmidt, C. P. Romao and H.-J. Meyer, CCDC 2311571: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2hlct9; (b) J. Beitlberger, M. Ströbele, P. Schmidt, C. P. Romao and H.-J. Meyer, CCDC 2428720: Experimental Crystal Structure Determination, 2025,  DOI:10.5517/ccdc.csd.cc2mj8t8.

---