**Carborane: Dedicated to the Work of Professor Alan Welch**

Editors

**Marina Yu. Stogniy Georgina Rosair**

MDPI • Basel • Beijing • Wuhan • Barcelona • Belgrade • Manchester • Tokyo • Cluj • Tianjin

*Editors* Marina Yu. Stogniy A.N.Nesmeyanov Institute of Organoelement Compounds of Russian Academy of Sciences Moscow Russia

Georgina Rosair Heriot Watt University Edinburgh UK

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *Crystals* (ISSN 2073-4352) (available at: https://www.mdpi.com/journal/crystals/special issues/ Carborane).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365–3619-4 (Hbk) ISBN 978-3-0365-3620-0 (PDF)**

© 2022 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

## **Contents**


## **About the Editors**

**Marina Yu. Stogniy** was born in Kazakhstan. In 2003-2009 she studied at the M.V. Lomonosov Moscow Institute of Fine Chemical Technology. She received her PhD degree in chemistry in 2013 under the supervision of Dr. Igor Sivaev. Dr Stogniy currently works as a Senior Researcher at the A.N. Nesmeyanov Institute of Organoelement Compounds of Russian Academy of Sciences (Laboratory of Organoaluminum and Organoboron Compounds) and teaches a course in general and inorganic chemistry at MIREA - Russian Technological University. Her areas of interest are the chemistry of polyhedral boron hydrides and metallocarboranes.

**Georgina Rosair** was born in Essex, England. She studied chemistry at Bristol University, obtaining a BSc in 1989, MSc in Analytical Chemistry in 1990 and a PhD in X-ray crystallography and computational chemistry with Prof Guy Orpen in 1994. In 1994 she became a postdoctoral worker in Prof. Alan Welch's group at Heriot Watt University, Edinburgh, to determine single crystal structures of carboranes and metallacarboranes, as well as crystal structure determinations for other research groups at Heriot Watt. In 1997 she became the staff crystallographer at Heriot Watt, the role expanded into powder diffraction and teaching inorganic chemistry, particularly practical laboratory courses. Since Prof Welch's retirement, her structure determination work now involves organometallic catalysts, supramolecular chemistry and organic photocatalysts and photosensitisers.

### *Editorial* **Carborane: Dedicated to the Work of Professor Alan Welch**

**Georgina M. Rosair 1,\* and Marina Yu. Stogniy 2,\***


In the 1950s, borohydrides arose as promising components of new rocket and aviation fuels. This led to the discovery of a new class of compounds—polyhedral boron hydrides and the creation of a new chapter in the chemistry of organoelement compounds, which are intrinsically attractive structures. It was one of the most important discoveries of the 20th century in the field of chemistry.

Polyhedral boron hydrides lie at the intersection of organic and inorganic chemistry. The main theoretical interest in the chemistry of these compounds is due to their unusual type of chemical bond and their three-dimensional aromaticity. The aromatic nature of polyhedral boron hydrides determines many properties that distinguish them from most boron hydrides and organoboron compounds: high thermal stability, kinetic stability of the borane cluster, a pronounced tendency towards substitution reactions and isomerisations.

The replacement of one or more boron atoms in a polyhedral cluster by atoms of other elements enables further diversification. The formation of carboranes, i.e., the inclusion of one or two carbon atoms in a boron cluster, leads to some radical changes. On one hand, it is the acidic character of the CH protons that makes it possible to replace the hydrogen atom(s) with various functional groups using standard organic synthesis methods. On the other hand, it becomes possible to remove one or more boron vertices, which significantly expands the range of structural types of carboranes. Thus, in addition to closed (*closo*-carboranes) structures, open (*arachno*- and *nido*-carboranes) ones become accessible. Moreover, open structures, such as the 7,8-dicarba-*nido*-undecaborate anion (*nido*-carborane) and its derivatives, are very promising ligands for the synthesis of metal complexes. The deprotonated form of *nido*-carborane (dicarbollide dianion [7,8-C2B9H11] <sup>2</sup>−) is a three-dimensional cluster with an open pentagonal face capable of forming strong *π*-bonds with transition metal cations, which makes it a unique ligand with unusual steric, electronic and chemical properties that are sometimes inaccessible for organic ligands.

For more than half a century, scientists from all over the world have been studying the properties of carboranes, as well as the possibility of obtaining new substances and materials with desired properties. The study of these compounds significantly expanded the modern understanding of molecular structures and the nature of chemical bonds, such as Wade–Mingos rules and the three-dimensional aromaticity concept, which is currently used to describe the structure of not only polyhedral boron hydrides but also transition metal clusters, fullerenes and their derivatives, etc.

One of the scientists who made a significant contribution to the development of the chemistry of carboranes is the British chemist Alan Welch.

Alan Welch undertook his PhD with Mike Hursthouse (who established the National Crystallography Service in the UK), then a postdoctoral degree in heteroboranes with FGA Stone in Bristol, and another postdoctoral degree with H-B Bürgi at ETH Zürich. From a lectureship at Edinburgh University, he joined Heriot Watt University in 1994 and

**Citation:** Rosair, G.M.; Stogniy, M.Y. Carborane: Dedicated to the Work of Professor Alan Welch. *Crystals* **2021**, *11*, 1365. https://doi.org/ 10.3390/cryst11111365

Received: 4 November 2021 Accepted: 4 November 2021 Published: 9 November 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

established the crystallography facility at Heriot Watt, as well as leading his internationally recognized research group in heteroborane chemistry.

Later, Alan Welch worked on the synthesis of new heteroborane compounds, in particular, metallacarboranes. His group investigated their spectroscopic and structural characterization and studied their isomerisations and reactivity. The chemistry of supraicosahedral heteroboranes, bis(carboranes), nitrosocarboranes, and non-Wadian metallacarboranes was significantly expanded by Prof. Alan Welch's group.

These eight papers form a Special Issue of *Crystals* to commemorate the excellent contribution made to carborane chemistry by Prof. Alan Welch, who retired from Heriot Watt University, Edinburgh, this year.

The papers illustrate the very comprehensive world of heteroborane chemistry, from liquid crystals to BNCT agents, di-halogen bonding to quantum chemical calculations of tetrel complexes of the carbonium ylide CB11H11, nickellacarboranes as potential acid-base sensors to revealing how the selective formations of metallacarborane diastereomers can arise and metallacarboranes as radical cation salts with dielectric or semiconductor properties.

A computational study by Drahomír Hnyk and co-workers used DFT to successfully describe the reactions of experimentally known *closo*-C2B8H10 with bases such as hydroxides and amines. The formation of [*arachno*-4,5-C2B6H11] − was established computationally when this was not demonstrable experimentally [1].

Laskova and co-workers developed BNCT (boron neutron capture therapy) agents that incorporate amino acids that are potentially taken up by malignant brain tumour cells. The recent synthesis and biological evaluation of *m*-carboranyl-cysteine as an agent for boron neutron capture therapy inspired the group to synthesize the analogue based on readily available 1-mercapto-*o*-carborane. The synthesis was optimised by using the "free of base" method [2].

Mandal was one of Alan Welch's PhD students, and his work is on bis(nickelation) of bis(o-carborane), which forms diastereoisomeric mixtures on metalation of the second carborane cage and additionally undergoes isomerisation. It was found that stereospecificity was influenced by intramolecular dihydrogen bonding, whereas a specific isomerisation outcome was related to the stereo-electronic nature of bis(phosphine) ligands [3].

Liquid crystals incorporating carboranes were explored by Núñez and co-workers. They varied substituents on *o*-carborane to tune liquid crystal properties employing the mesogen cholesteryl benzoate. They found that the methyl substituent produced a blue phase, whilst the phenyl substituent species was not mesogenic [4].

Stogniy, Sivaev and co-workers synthesized half-sandwich nickel(II) complexes with amidine ligands where breakage of the Ni–N bond on acidification results in a colour change, which gives these complexes potential as acid-base indicators [5].

High-level quantum-chemical computations (G4MP2) were employed by Oliva-Enrich and co-workers to examine tetrel bonding (interaction between any electron donating system and a group of 14 elements acting as a Lewis acid) to predict the formation of tetrel complexes between the icosahedral carbonium ylide CB11H11 and a set of simple molecules and anions. The electronic structure of the complexes was analysed with AIM and ELF methods, showing the C\_\_\_ X sharing and closed-shell interactions in the complexes [6].

Intermolecular halogen bonding, in this case, the diiodo bond, was investigated by Sivaev and co-workers. They obtained 1,12-diiodo-*ortho*-carborane, and its crystal structure was determined by X-ray diffraction, which revealed the existence of the I\_\_\_ I halogen bond in its crystal structure. Such dihalogen bonding is not found in 1,12-dibromo*ortho*-carborane. Quantum chemical calculations determined the noncovalent interaction preferences in 1,12-diiodo- and 1,12-dibromo-*ortho*-carboranes, which were in agreement with experimental findings [7].

Radical-cation salts based on tetramethyltetrathiafulvalene (TMTTF) and tetramethyltetraselenefulvalene (TMsTSF) with metallacarborane anions were also explored by Sivaev and co-workers. The iron bis(1,2-dicarbollide) and chromium bis(1,2-dicarbollide) salts were synthesized by electrocrystallisation, and their characterisation revealed that the resulting TMTTF radical-cation salts are dielectrics, whilst the TMTSF species is a narrow-gap semiconductor [8].

Thus, this Special Issue combines the latest achievements in the field of theoretical and experimental chemistry of carboranes. We thank all the authors who took part in this issue and look forward to further fruitful and impressive developments in the chemistry of carboranes.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


#### *Communication*

## **New Radical-Cation Salts Based on the TMTTF and TMTSF Donors with Iron and Chromium Bis(Dicarbollide) Complexes: Synthesis, Structure, Properties †**

**Denis M. Chudak 1, Olga N. Kazheva 2,3,\*, Irina D. Kosenko 4, Gennady V. Shilov 2, Igor B. Sivaev 4,5, Georgy G. Abashev 6, Elena V. Shklyaeva 6, Lev I. Buravov 2, Dmitry N. Pevtsov 2, Tatiana N. Starodub 7, Vladimir I. Bregadze <sup>4</sup> and Oleg A. Dyachenko <sup>2</sup>**


**Abstract:** New radical-cation salts based on tetramethyltetrathiafulvalene (TMTTF) and tetramethyltetraselenefulvalene (TMsTSF) with metallacarborane anions (TMTTF)[3,3- -Cr(1,2-C2B9H11)2], (TMTTF)[3,3- -Fe(1,2-C2B9H11)2], and (TMTSF)2[3,3- -Cr(1,2-C2B9H11)2] were synthesized by electrocrystallization. Their crystal structures were determined by single crystal X-ray diffraction, and their electrophysical properties in a wide temperature range were studied. The first two salts are dielectrics, while the third one is a narrow-gap semiconductor: <sup>σ</sup>RT = 5 <sup>×</sup> <sup>10</sup>−<sup>3</sup> Ohm−1cm−1; Ea <sup>≈</sup> 0.04 eV (aprox. 320 cm<sup>−</sup>1).

**Keywords:** iron bis(1,2-dicarbollide); chromium bis(1,2-dicarbollide); tetramethyltetrathiafulvalene; tetramethyltetraselenafulvalene; radical-cation salts; crystal and molecular structure; electric conductivity

#### **1. Introduction**

Radical-cation salts and charge transfer complexes based on derivatives of tetrathiafulvalene (TTF) constitute a wide class of organic materials with transport properties ranging from insulating to superconducting [1–4]. This work is part of the systematic study of radical-cation salts of tetrathiafulvalene and its derivatives with metallacarborane anions, of which earlier results were summarized in works [5–7].

Transition metal bis(dicarbollide) complexes [3,3- -M(1,2-C2B9H11)2] − (M = Fe, Co, or Ni) are of great interest as counterions for the synthesis of TTF-based molecular conductors due to the unique high stability, possibility of tuning the charge and nature of the metal, and wide range of options for modification with dicarbollide ligands via hydrogen substitution by other atoms and functional groups [5,6]. Although most of the compounds studied were BEDT-TTF-based radical-cation salts, recently, we have synthesized radical-cation

5

**Citation:** Chudak, D.M.; Kazheva, O.N.; Kosenko, I.D.; Shilov, G.V.; Sivaev, I.B.; Abashev, G.G.; Shklyaeva, E.V.; Buravov, L.I.; Pevtsov, D.N.; Starodub, T.N.; et al. New Radical-Cation Salts Based on the TMTTF and TMTSF Donors with Iron and Chromium Bis(Dicarbollide) Complexes: Synthesis, Structure, Properties. *Crystals* **2021**, *11*, 1118. https://doi.org/10.3390/ cryst11091118

Academic Editors: Georgina Rosair and Marina Yu. Stogniy

Received: 11 August 2021 Accepted: 9 September 2021 Published: 14 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

salts based on such unconventional and rather exotic donors as bis(1,3-propylenedithio) tetrathiafulvalene [8,9], dibenzotetrathiafulvalene [10], and 4,5-ethylenedithio-4- ,5- -(2-oxa-1,3-propylenedithio)-tetrathiafulvalene [9]. On the other hand, although compounds of the composition (TMTXF)2Y (X = T, S) are usually classical organic metals among which the first organic superconductors were discovered [4,7], and TMTTF and TMTSF radical-cation salts continue to attract the attention of researchers [11–15], very little attention has been paid to TMTTF and TMTSF radical-cation salts with metallacarborane anions [16–19]. This prompted us to prepare and investigate new TMTTF and TMTSF radical-cation salts with metallacarborane anions.

This contribution describes the synthesis, structure, and electrical conductivity of new salts with TMTTF and TMTSF radical-cations and metallacarborane anions: (TMTTF)[3,3- -Cr(1,2- C2B9H11)2] (**1**), (TMTTF)[3,3- -Fe(1,2-C2B9H11)2] (**2**), and (TMTSF)2[3,3- - Cr(1,2-C2B9H11)2] (**3**).

#### **2. Results and Discussion**

Single crystals of compounds **1**–**3** suitable for X-ray diffraction studies in the form of thin plates were obtained by electrochemical crystallization (See Supplementary Materials and Table 1). The crystal structure of **1** is formed by the TMTTF radical-cations and the [3,3- -Cr(1,2-C2B9H11)2] − anions occupying general positions in the unit cell (Figure 1). (TMTTF)[3,3- -Cr(1,2-C2B9H11)2] has a pseudo-layered structure, in which anionic layers alternate along the *ac* diagonal with layers formed by radical-cation dimers (Figure 2). The dimer formation corresponds to the stoichiometry of the salt: in this case due to the Peierls instability a phase transition should occur with doubling of the stacks period [7]. The distances between the averaged planes of the TMTTF donors in the dimers are 3.38 Å (the planes are drawn through all S atoms), and the dihedral angle between the planes is 0◦ by symmetry conditions. There are short intermolecular S ... S interactions (3.426(1)–3.432(1) Å) of the "face-to-face" type between the TMTTF donors in the dimers.

**Table 1.** Crystal data and structure refinement for (TMTTF)[3,3- -Cr(1,2-C2B9H11)2] (**1**), (TMTTF)[3,3- - Fe(1,2-C2B9H11)2] (**2**), and (TMTSF)2[3,3- -Cr(1,2-C2B9H11)2] (**3**).


**Figure 1.** TMTTF radical-cation and anion in (**1**). Thermal ellipsoids are given at 30% probability level. Cage H atoms omitted for clarity.

**Figure 2.** Crystal packing fragment of (**1**). A view along the *b* axis. The unit cell is outlined. H atoms are omitted for clarity.

The TMTTF<sup>+</sup> radical-cations are non-planar and have a "boat" conformation: the maximum deviations of terminal C(9), C(10), C(11), and C(12) atoms from the plane of the averaged molecule drawn through all sulfur atoms are 0.30–0.36 Å.

The Cr-C and Cr-B bond lengths are 2.173(2)–2.180(2) and 2.232(2)–2.279(2) Å, correspondingly. The distances from the chromium atom to the C2B3 faces of the dicarbollide ligands are equal to 1.68 Å, which is close to the corresponding distances found in the structures of Cs [3,3- -Cr(1,2-C2B9H11)2] [20], (TTF)[3,3- -Cr(1,2-C2B9H11)2] [21], and (BEDT-TTF)2[3,3- -Cr(1,2-C2B9H11)2] [22,23]. The dicarbollide ligands in the [3,3- -Cr(1,2- C2B9H11)2] − anion are turned relative to each other by 180◦, forming the *transoid* conformation. The C2B3 faces deviate slightly from parallel, being inclined by 178.7◦ to each other.

The electrical conductivity measurements have shown that 1 is an insulator with σ293~10−<sup>11</sup> Ohm<sup>−</sup>1cm−1. The low value of electrical conductivity is apparently connected with the absence of conducting layers and dimerization of the radical-cations stacks.

It should be noted that compound 1 is the first TMTTF radical-cation salt with an unsubstituted transition metal bis(dicarbollide), while the radical-cation salts (TMTTF)[8- HO-3,3- -Co(1,2-C2B9H10)(1- ,2- -C2B9H11)] and (TMTTF)(8,8- -Cl2-3,3- -Co(1,2-C2B9H10)2]2 obtained earlier contained substituted bis(dicarbollide) anions [16,17].

The crystal structure of (TMTTF)[3,3- -Fe(1,2-C2B9H11)2] (**2**) is formed by a quarter of the TMTTF radical-cation in a special position placed on the *m* plane and a quarter of the [3,3- -Fe(1,2-C2B9H11)2] − anion in the *2/m* special position of the unit cell (Figure 3). The compound **2** is characterized by a structure where the TMTTF cations and the metallacarborane anions form staggered stacks (Figures 4 and 5). The distances between the averaged

planes of the TMTTF donors in the dimers are 3.38 Å, and the dihedral angle between the planes is 0◦ by symmetry conditions.

**Figure 4.** Crystal packing fragment of (**2**). A view along the *c* axis. The unit cell is outlined. H atoms are omitted for clarity.

**Figure 5.** A stack and radical-cations overlapping in (**2**).

The Fe-C and Fe-B bond lengths are 2.0790(9)–2.1001(8) and 2.1001(8)–2.1494(8) Å, correspondingly, and the overlapping values are due to the statistical disordering of carbon and boron atoms in the dicarbollide ligands. The distances from the iron atom to the C2B3 faces of the dicarbollide ligands are equal to 1.53 Å, which is close to the distances in analogous salts of the iron bis(dicarbollide) anion [19,24,25]. The dicarbollide ligands are turned relative to each by 180◦, forming the *transoid* conformation. The C2B3 faces are parallel by symmetry conditions.

According to the electric conductivity measurements, compound 2 is an insulator with conductivity ~10−<sup>10</sup> Ohm−1cm−1. The low value of electroconductivity is in an agreement with the 1:1 stoichiometry and non-layered structure of the salt, as well as with the inclination angle of the radical-cations in the stack, at which there is only slight overlap between neighboring radical-cations.

The (TMTSF)2[3,3- -Cr(1,2-C2B9H10)2] (3) crystals are isostructural to (TMTSF)2[3,3- - Co(1,2-C2B9H11)2] and (TMTSF)2[3,3- -Fe(1,2-C2B9H11)2] salts studied earlier, containing cobalt and iron bis(dicarbollide) anions [18,19]. The crystal structure of 3 is formed by the TMTSF cation in a general position and the [3,3- -Cr(1,2-C2B9H11)2] − anion in a special centrosymmetrical position (Figure 6). Compound 3 possesses a structure (Figures 7 and 8) where the TMTSF+• radical-cations and anions form staggered stacks. The distances between the averaged planes of the TMTSF donors in the dimers are 3.70 and 3.73 Å, and the dihedral angle between the planes is 0◦ by symmetry conditions.

The Cr-C and Cr-B bond lengths are 2.175(7)–2.176(7) and 2.226(8)–2.277(8) Å, correspondingly. The distances from the chromium atom to the C2B3 faces of the dicarbollide ligands are equal to 1.68 Å, and the dicarbollide ligands in the [3,3- -Cr(1,2-C2B9H10)2] − anion are turned relative to each other by 180◦, forming the *transoid* conformation. The C2B3 faces are parallel to each other by the symmetry conditions.

The electroconductivity measurements have revealed that compound 3 in the range of 41–195 K behaves like a dielectric. However, above 195 K, the delocalization of the positive charge disappears due to the numerous intermolecular S ... S contacts and an inconspicuous dielectric–semiconductor structural phase transition occurs, caused by charge ordering: stacks contain both TMTSF molecules and TMTSF radical-cations. The room temperature electric conductivity <sup>σ</sup><sup>293</sup> = 5·10−<sup>3</sup> Ohm−1cm−<sup>1</sup> and activation energy Ea ∼= 0.04 eV (Figure 9). It should be noted that analogous salts (TMTSF)2[3,3- -Co(1,2- C2B9H11)2] and (TMTSF)2[3,3- -Fe(1,2-C2B9H11)2] were characterized by electroconductivity values σ<sup>293</sup> of 15 and 0.1 Ohm<sup>−</sup>1cm−1, correspondingly [18,19].

**Figure 6.** Anion in (**3**). Thermal ellipsoids are given at 30% probability level.

**Figure 7.** Crystal packing fragment of (**3**). A view along the *a* axis. The unit cell is outlined. H atoms are omitted for clarity.

**Figure 8.** A stack of radical-cations in (**3**).

**Figure 9.** Temperature dependence of electrical resistivity of (**3**).

In conclusion, new salts with the TMTTF and TMTSF radical-cations and metallacarborane anions (TMTTF)[3,3- -Cr(1,2-C2B9H11)2] (**1**), (TMTTF)[3,3- -Fe(1,2-C2B9H11)2] (**2**), and (TMTSF)2[3,3- -Cr(1,2-C2B9H11)2] (**3**) were electrochemically synthesized and investigated. Their crystal structures were determined by X-ray study and electroconductivities were measured. Salts (**1**) and (**2**) are insulators, which is explained by the 1:1 stoichiometry and the absence of an extended network of interdonor interactions, whereas (**3**) is a semiconductor at room temperature with electroconductivity <sup>σ</sup><sup>293</sup> = 5·10−<sup>3</sup> Ohm−1cm−1, which is lower than in (TMTSF)2[3,3- -Fe(1,2-C2B9H11)2] and (TMTSF)2[3,3- -Co(1,2-C2B9H11)2] salts (electroconductivity values σ<sup>293</sup> of 0.1 and 15 Ohm−1cm−1, correspondingly). The tendency of a rise in conductivity (5·10−<sup>3</sup> < 0.1 < 15) is apparently connected with decreasing the cation size in the order Cr3+ > Fe3+ > Co3+ [26], which leads to decreasing the corresponding metallacarborane anion size and, in turn, to unit cell compression and a tighter radical-cation packing of the salts.

**Supplementary Materials:** Details of experimental data including synthesis of the title compounds, their X-ray diffraction studies, and electric resistivity measurements are available online at https: //www.mdpi.com/article/10.3390/cryst11091118/s1.

**Author Contributions:** Synthesis, D.M.C., I.B.S., I.D.K., G.G.A., E.V.S.; measurements, L.I.B., D.N.P., T.N.S.; X-ray diffraction study, G.V.S.; data analysis and writing, O.N.K.; research conception, V.I.B., O.A.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The crystallographic data for this paper (the CCDC numbers 2091714, 2091713, 2091715 for (**1**), (**2**), (**3**), respectively) can be obtained free of charge via www.ccdc.cam.ac. uk/data\_request/cif (accessed on 11 August 2021).

**Acknowledgments:** This work was partly performed in accordance with the state task of the Institute of Problems of Chemical Physics, Russian Academy of Sciences, State Registration No. AAAA-A19- 119092390076-7. Synthesis and characterization of the starting metallacarboranes were performed at A.N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences with support of the Ministry of Science and Higher Education of the Russian Federation.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


## *Communication* **1,12-Diiodo-***Ortho***-Carborane: A Classic Textbook Example of the Dihalogen Bond †**

**Kyrill Yu. Suponitsky 1,\*, Alexei A. Anisimov 1,2, Sergey A. Anufriev 1, Igor B. Sivaev 1,\* and Vladimir I. Bregadze <sup>1</sup>**

	- <sup>2</sup> Higher Chemical College at the Russian Academy of Sciences, D.I. Mendeleev Russian Chemical Technological University, 9 Miusskaya Sq., 125047 Moscow, Russia
	- **\*** Correspondence: kirshik@yahoo.com (K.Y.S.); sivaev@ineos.ac.ru (I.B.S.)
	- † Dedicated to Professor Alan J. Welch in occasion of his retirement at Herriot-Watt University and in recognition of his outstanding contribution in the carborane chemistry.

**Abstract:** The crystal structure of 1,12-diiodo-*ortho*-carborane 1,12-I2-1,2-C2B10H10 was determined by single crystal X-ray diffraction. In contrary to earlier studied 1,12-dibromo analogue 1,12-Br2-1,2- C2B10H10, its crystal packing is governed by the presence of the intermolecular I··· I dihalogen bonds between the iodine atom attached to the carbon atom (acceptor) and the iodine atom attached to the antipodal boron atom (donor) of the carborane cage. The observed dihalogen bonds belong to the II type and are characterized by classical parameters: shortened I··· I distance of 3.5687(9) Å, C–I··· I angle of 172.61(11)◦ and B–I··· I angle of 92.98(12)◦.

**Keywords:** carborane; iodo derivatives; dihalogen bond; X-ray structure; quantum chemical calculationsy

#### **1. Introduction**

Carboranes [CB11H12] − and C2B10H12, in which one or two vertices in boron icosahedron are replaced by a carbon unit, are a fascinating family of compounds with exceptional chemical and thermal robustness, unique geometry, rigidity, and synthetic versatility [1]. Selective chemical substitution of hydrogen atoms at carbon or boron atoms in these clusters allows for their use as rigid, three dimensional scaffolds upon which to construct new materials, such as liquid crystals [2–5], nonlinear optical materials [6–9], carboranebased anticrowns [10], and even in drug design [11,12]. As rigid molecules of fixed length, carboranes can be used as building blocks ("molecular tinkertoys") [13–18] for supramolecular assemblies, such as porous coordination polymers or metal–organic frameworks (MOFs) [19–22]. Another type of supramolecular structures with the participation of carboranes is based on the acidity of their CH groups, which demonstrate a high potential for hydrogen bonding. Indeed, intermolecular C–H··· O and C–H··· N hydrogen bonding, including bifurcated interactions, features in much of the supramolecular chemistry of carboranes [23,24]. Functionalization of carboranes with different substituents including halogen atoms opens an opportunity to the formation diverse noncovalent interactions [25,26]. Thus, the intermolecular C–H··· X–B hydrogen bonds were found to stabilize crystal structures of fluoro- [27], bromo- [28] and iodo- [29–34] derivatives of *ortho*-carborane. Alternatively, hydrogen atoms bonded to the carbon atoms can also be replaced by halogen atoms. It should be noted that when substituted at a carbon atom, carborane acts as an electron-withdrawing group with respect to a substituent, while when substituted at boron atoms it plays the role of an electron-releasing group. The further the location of a substituent is from carbon atoms, the higher the electron-releasing ability is the carborane cage [35].

Halogen bonds are one of the strongest noncovalent intermolecular interactions, and are formed between the σ-hole of a halogen atom and nucleophile [36–39]. In the case

**Citation:** Suponitsky, K.Y..; Anisimov, A.A.; Anufriev, S.A.; Sivaev, I.B.; Bregadze, V.I. 1,12-Diiodo-*Ortho*-Carborane: A Classic Textbook Example of the Dihalogen Bond. *Crystals* **2021**, *11*, 396. https://doi.org/10.3390/ cryst11040396

Academic Editors: Georgina Rosair and Marina Yu. Stogniy

Received: 15 March 2021 Accepted: 6 April 2021 Published: 8 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

13

of halogen bonds in which both atoms are halogens, the σ-hole is activated by electron acceptor substitution of a halogen while donor substituents are necessary to increase ability of lone pair donation of the second halogen atom. Therefore, in the case when halogen atoms are simultaneously introduced to the carborane carbon atom and the boron atom is antipodal to it, this makes the formation of intermolecular halogen bonds possible, where the halogen atom attached to the carbon atom plays the role of an acceptor, and the halogen atom bonded to boron acts as a donor. Thus, the *B,C*-dihalogen-substituted carboranes represent a unique class of small molecules, in the crystals of which the formation of intermolecular dihalogen bonds could be possible without the participation of the second component. In particular, one might expect a formation of the intermolecular dihalogen bonds for 1,12-dihalo-*ortho*-carboranes 1,12-X2-1,2-C2B10H10. However, the recent study of the crystal structure of 1,12-Br2-1,2-C2B10H10 showed that in this case, instead of the formation of the intermolecular C-Br··· Br-B dihalogen bonds, the formation of the C–H··· Br–B hydrogen and C-Br··· H-B halogen bonds occurs [28]. On the other hand, the σ-hole size, which is the determining factor in the formation of a halogen bond, depends on both the electronic effect of the substituent and the electronegativity of the halogen atom [40–42]. This prompted us to study intermolecular interactions in an analogous diiodine derivative 1,12-I2-1,2-C2B10H10 using single crystal X-ray diffraction and quantum chemical calculations.

#### **2. Results and Discussion**

Despite the fact that the syntheses of the *C*-iodo derivatives of *ortho*-carborane were first reported more than 50 years ago [43,44], they were on the periphery of mainstream carborane chemistry developments, and were not even well characterized [34,45]. In this way, they radically differ from the *B*-iodo derivatives of *ortho*-carborane, which have found active use in the synthesis of *B*-alkyl and aryl derivatives by means of Pd-catalyzed cross-coupling reactions [46–52]. Therefore, the synthesis of the *C-*iodo derivatives of *ortho*-carborane is not an easy task.

In this respect, the synthesis and characterization of the *C*-halogen derivatives of the carba-*closo*-dodecaborate anion [1-X-1-CB11H11] − (X = F, Cl, Br, I) are described much better [53]. Moreover, the preparation of its 1,12-diiodo derivative [1,12-I2-1-CB11H10] −, containing iodine atoms in opposite positions of the boron backbone, has recently been described [54]. However, in the case of anionic carboranes, it is rather difficult to find a cation that, on the one hand, will not form additional non-covalent bonds with the anion and, on the other hand, will be small enough not to hinder the formation of intermolecular dihalogen bonds between the anions. In addition, unlike the *C*-substituted *ortho*-carborane, the *C*-substituted carba-*closo*-dodecaborate anion has no or negligible electron-withdrawing effects [55].

An attempt to prepare 1,12-I2-1,2-C2B10H10 by the reaction of the lithium derivative of 9-iodo-*ortho*-carborane with iodine in 1,2-dimethoxyethane resulted in the expected formation of a mixture of 1,9-I2-1,2-C2B10H10 and 1,12-I2-1,2-C2B10H10 derivatives (1:1). However, in contrast to the similar dibromo derivatives 1,9-Br2-1,2-C2B10H10 and 1,12- Br2-1,2-C2B10H10 [56], we failed to separate this mixture. Nevertheless, we managed to obtain the desired 1,12-diiodo derivative as a by-product of the cross-coupling reaction of 9-iodo-*ortho*-carborane with phenylmagnesium bromide. Another by-product of this reaction was 1-iodo-*ortho*-carborane, which we also obtained by direct reaction of the lithium derivative of *ortho*-carborane with iodine. Notably, formation of similar products of iodine migration in the process of cross-coupling of *B*-iodo carboranes with Grignard reagents was noted earlier [57].

As mentioned in the Introduction, no Br··· Br halogen bond was observed in the crystal structure of 1,12-dibromo-*ortho*-carborane 1,12-Br2-1,2-C2B10H10 [28]. It should be noted that similarities and differences between bonding preferences of the bromine atom in comparison to iodine atom, on the one hand, and chlorine atom, on the other hand, was the subject of extensive studies [58–61]. Based on comparison of the crystal packing of 1-Ph-2-X-*ortho*-carboranes (X = F, Cl, Br, I), it was shown that both Br and I form Hal··· π interactions, while neither Cl or F participate in such interactions [58]. Study of N-(2-halo-2,2-dinitroethyl)pyrrolidine- 2,5-diones (Hal = F, Cl, Br) [59] has revealed that both Cl and Br participate in halogen bonding, but bromine interacts with the carbonyl oxygen atom (the strongest donor site), while chlorine prefers to connect to much weaker donors, namely, oxygen atoms of the nitro group. Based on the above, it becomes unclear a priori which packing motif should be expected in the crystal of 1,12-diiodo-*ortho*-carborane.

Single crystals of 1,12-I2-1,2-C2B10H10 suitable for X-ray study were obtained in the form of thin plates by slow evaporation of chloroform solution. An asymmetric unit cell of 1,12-I2-1,2-C2B10H10 contains one molecule (Figure 1). The I1–C1 bond length (2.121(2) Å) is slightly longer than average X-ray value for I–C (aromatic) bonds (2.095 Å [62]) and is significantly shorter than the B12–I12 bond (2.179(2) Å).

**Figure 1.** General view of 1,12-I2-1,2-C2B10H10. Thermal ellipsoids are given at 50% probability level.

The main packing motif of the crystal structure of 1,12-I2-1,2-C2B10H10 is represented by infinite chains along the *b* direction formed by the C–I··· I–B dihalogen bonds of II type [37,63] (the I(12)··· I(1') distance is 3.5687(9) Å, the B(12)-I(12)··· I(1') and I(12)··· I(1')-C(1') angles are 92.98(12) and 172.61(11)◦, respectively) (Figure 2), which is very different from 1,12-Br2-1,2-C2B10H10 studied earlier.

In order to find out the reason of observed differences as well as peculiarities of the crystal packing of 1,12-I2-1,2-C2B10H10, we used energetic analysis of intermolecular contacts, that is frequently invoked for crystal packing study [64–66]. We calculated halogen bonded dimer for both compound 1,12-I2-1,2-C2B10H10 and similar dimer in which iodine atoms are replaced with bromines. The results are presented in Figure 3 and Table 1. The calculated dimer of 1,12-I2-1,2-C2B10H10 is characterized by the structure similar to that found experimentally. The I··· I distance is somewhat shorter, while C–I··· I and B–I··· I angles and B–H··· I distances are close to experimentally observed values. Topological analysis of calculated electron density for 1,12-I2-1,2-C2B10H10 dimer has revealed additional stabilization of the dimeric structure with the B–H··· I hydride–halogen bonds that was not evident from the consideration of bare X-ray data. From Table 1, it can be seen that energy of the I··· I contact is sizably higher than that of the B–H··· I contacts; therefore, the I··· I dihalogen bond can be considered as the structure-forming interaction in the crystal of 1,12-I2-1,2-C2B10H10. In contrary, optimized geometry of dimeric 1,12-dibromo-*ortho*carborane appeared to be quite different. The C–Br··· Br angle significantly deviates from 180◦. As a consequence, energy of the Br··· Br interactions is relatively small and becomes comparable to the B–H··· Br interactions which are also formed between two molecules in the dimer. It means that the Br··· Br interactions are no more structure-forming ones. These results are in qualitative agreement with a previous experiment [26]; according to which, no Br··· Br halogen bond is observed in the crystal of 1,12-Br2-1,2-C2B10H10.

**Figure 2.** Crystal packing fragment of 1,12-I2-1,2-C2B10H10. Halogen bonded chains are formed along axis *b*.

**Figure 3.** Noncovalent bonding in dimers of 1,12-I2-1,2-C2B10H10 (**left**) and 1,12-Br2-1,2-C2B10H10 (**right**).

**Table 1.** Characteristics of intermolecular noncovalent interactions for dimers of 1,12-I2-1,2-C2B10H10 and 1,12-Br2-1,2-C2B10H10.


The above results have demonstrated computational ability for, at least, qualitative explanation and prediction of the main crystal packing motif for dihalogen derivatives of *ortho*-carborane. Based on that, we made an attempt to predict the possibility of halogen bond formation in 1,3-I2-1,2-C2B10H10 and 1,9-I2-1,2-C2B10H10. Those isomers were chosen because they can be experimentally obtained from the available 3- and 9-iodo*ortho*-carboranes, while synthesis of other possible isomers is troublesome. In Figure 4 and Table 2, the results of calculation of dimers of the 1,3- and 1,9-isomers are presented.

**Figure 4.** Noncovalent bonding in dimers of 1,9-I2-1,2-C2B10H10 (**left**) and 1,3-I2-1,2-C2B10H10 (**right**).

**Table 2.** Characteristics of intermolecular noncovalent interactions for halogen-bonded dimers of 1,3-I2-1,2-C2B10H10 and 1,9-I2-1,2-C2B10H10.


One can see that 1,9-I2-1,2-C2B10H10 demonstrates the same system of close contacts and nearly the same energetic properties of halogen-bonded dimer of 1,12-I2-1,2-C2B10H10. The energy of a halogen bond is only 0.2 kcal/mol less; B9-I9··· I1' and I9··· I1'-C1' only slightly deviate from 90 and 180◦, respectively, while energies of B–H··· I interactions are only 0.1 kcal/mol higher. It allows consideration of the I··· I halogen bond as a predominant interaction in the potential crystal structure of 1,9-I2-1,2-C2B10H10.

When looking at halogen-bonded dimers built up of 1,3-I2-1,2-C2B10H10, one can observe a clear trend of weakening of the I··· I halogen bonds and simultaneous strengthening of the B–H··· I interactions and increases in their roles in stabilizing dimeric structures upon approaching the iodine substituent from its position at B12 to B3. During movement of the iodine atom from B12 to B3, quantitative changes due to the weakening of the I··· I halogen bonds can be transformed to qualitative changes, which can result in the disappearance of the I··· I halogen bonds from the crystal structure of 1,3-I2-1,2-C2B10H10.

In conclusion, 1,12-diiodo-*ortho*-carborane was obtained and its crystal structure was determined by X-ray diffraction, which revealed the existence of the I··· I halogen bond in its crystal structure, in contrast to 1,12-dibromo-*ortho*-carborane. Based on quantum chemical calculation, we have determined preferences of the type of noncovalent interactions in 1,12-diiodo- and 1,12-dibromo-*ortho*-carboranes which appeared to be in agreement with experimental findings. Based on our results, we can predict the formation of the I··· I halogen bonds in 1,9-diiodo-*ortho*-carborane, while our results cannot provide solid support for the formation of such bonds in the 1,3-isomer. This question is still open and can be answered experimentally. Synthesis and crystal growth of 1,9- and 1,3-diiodo-*ortho*carboranes is in progress in our group.

#### **3. Materials and Methods**

#### *3.1. General*

Compounds 9-iodo-*ortho*-carborane and bis(triphenylphosphine)palladium(II) dichloride were prepared according to the literature procedures [67,68]. Solvents 1,2-dimethoxyethane and diethyl ether were dried using standard procedures [69]. Phenyl iodide was distilled at boiling point. All other chemical reagents were purchased from Sigma Aldrich, Acros Organics and ABCR and used without purification. All reactions were carried out in an argon atmosphere. The reaction progress was monitored by thin-layer chromatography (Merck F254 silica gel on aluminum plates) and visualized using 0.5% PdCl2 in 1% HCl in aq. MeOH (1:10). Acros Organics silica gel (0.060–0.200 mm) was used for column chromatography. The NMR spectra at 400.1 MHz (1H), 128.4 MHz (11B) and 100.0 MHz ( 13C) were recorded with a Varian Inova 400 spectrometer. The residual signal of the NMR solvent relative to Me4Si was taken as an internal reference for 1H and 13C NMR spectra; 11B NMR spectra were referenced using BF3 . Et2O as an external standard.

#### *3.2. Cross-Coupling of 9-Iodo-Ortho-Carborane with PhMgBr*

Phenyl iodide (0.70 mL, 1275 mg, 6.25 mmol) was added to a mixture of magnesium turnings (228 mg, 9.38 mmol) in fresh distilled diethyl ether (25 mL). The resulting mixture was heated under reflux for 1 h. Then, 9-iodo-*ortho*-carborane (675 mg, 2.50 mmol) in fresh distilled diethyl ether (25 mL) was added, and the reaction was stirred at room temperature for another 1 h. Then, copper(I) iodide (25 mg, 0.13 mmol, catalytic amount) with [(Ph3P)2PdCl2] (83 mg, 0.13 mmol, catalytic amount) were added. The reaction was heated under reflux for 16 h and 6% HCl in water (50 mL) was added. The organic layer was separated; the water layer was washed with diethyl ether (3 × 50 mL). The organic phases were combined, dried over Na2SO4, and concentrated under reduced pressure. The crude product was purified by column chromatography on silica using a mixture of chloroform and petroleum ether (1:3, *v*/*v*) to give, along with the expected 9-phenyl-*ortho*-carborane, pale-yellow solids of 1-iodo-*ortho*-carborane (15 mg, yield 2%) and 1,12-diiodo-*ortho*- carborane (20 mg, yield 2%) as side products.

Compound 1-I-1,2-C2B10H11: 1H NMR (CDCl3, ppm): *δ* 3.78 (1H, br s, C*H*carb), 3.7–0.7 (10H, br m, B*H*). 11B NMR (CDCl3, ppm): *<sup>δ</sup>* −0.7 (1B, d, J = 151 Hz), −4.0 (1B, d, J = 154 Hz), −7.8 (4B, d, J = 125 Hz), −9.0 (2B, d, J = 111 Hz), −11.7 (1B, d, J = 167 Hz).

Compound 1,12-I2-1,2-C2B10H10: 1H NMR (CDCl3, ppm): *δ* 3.86 (1H, br s, C*H*carb), 4.1–0.6 (9H, br m, B*H*). 11B NMR (CDCl3, ppm): *δ* 0.9 (1B, d, J = 156 Hz), –5.8 (2B, d, J = 160 Hz), −7.8 (2B, d, J = 173 Hz), −8.7 (2B, d, J = 183 Hz), −10.7 (2B, d, J = 168 Hz), −16.7 (1B, s).

#### *3.3. General Synthetic Procedure of C-Iodination of Ortho-Carborane and Its B-I Derivatives*

The 2.25 M BuLi in hexanes was added to a mixture of carborane in fresh distilled 1,2-dimethoxyethane (10 mL). The mixture was stirred for 1 h at room temperature and I2 was added by one portion. The reaction was stirred at room temperature overnight and Na2S2O3·5H2O (1000 g, 4.03 mmol) in water (10 mL) and diethyl ether (15 mL) were added. The organic layer was separated; the water layer was washed with diethyl ether (25 mL). The organic phases were combined, dried over Na2SO4, and concentrated under reduced pressure. The crude product was purified by column chromatography on silica using diethyl ether as the eluent to give the corresponding C–I derivative.

Compound 1,2-I2-1,2-C2B10H10: 2.25 M BuLi in hexanes (1.00 mL, 2.25 mmol), *ortho*carborane (144 mg, 1.00 mmol) and I2 (635 mg, 2.50 mmol) were used; a yellow crystalline solid was obtained (381 mg, yield 96%). 1H NMR (CDCl3, ppm): *<sup>δ</sup>* 3.9–0.8 (10H, br m, B*H*). 11B NMR (CDCl3, ppm): *<sup>δ</sup>* <sup>−</sup>2.5 (2B, d, J = 154 Hz), <sup>−</sup>4.5 (2B, d, J = 183 Hz), <sup>−</sup>6.8 (6B, d, J = 183 Hz).

Compounds 1,9- and 1,12-I2-1,2-C2B10H10: 2.25 M BuLi in hexanes (0.49 mL, 1.10 mmol), 9-iodo-*ortho*-carborane (135 mg, 0.50 mmol) and I2 (305 mg, 1.20 mmol) were used; a pale-grey crystalline solid was obtained (80 mg, yield 20%). 1H NMR (CDCl3, ppm): *δ* 4.07 (1H, br s, C*H*carb, 1,9-isomer), 3.87 (1H, br s, C*H*carb, 1,12-isomer), 3.9–0.6 (20H, br m, B*H,* 1,9 + 1,12-isomers).

#### *3.4. Synthesis of 1-Iodo-Ortho-Carborane*

The 2.25 M BuLi in hexanes (0.40 mL, 0.90 mmol) was added to a mixture of *ortho*carborane (144 mg, 1.00 mmol) in fresh distilled 1,2-dimethoxyethane (10 mL). The mixture

was stirred for 1 h at room temperature, and I2 (381 mg, 1.50 mmol) was added by one portion. The reaction was stirred at room temperature overnight and Na2S2O3·5H2O (1000 g, 4.03 mmol) in water (10 mL) and diethyl ether (15 mL) were added. The organic layer was separated; the water layer was washed with diethyl ether (25 mL). The organic phases were combined, dried over Na2SO4, and concentrated under reduced pressure. The crude product was purified by column chromatography on silica using petroleum ether as an eluent to give a pale-grey crystalline solid of 1-I-1,2-C2B10H11 (160 mg, yield 59%).

#### *3.5. X-ray Diffraction Study*

Single crystal X-ray diffraction experiments were carried out using a SMART APEX2 CCD diffractometer (λ(Mo-Kα) = 0.71073 Å, graphite monochromator, ω-scans) at 120 K. Collected data were processed by the SAINT and SADABS programs incorporated into the APEX2 program package [70]. The structures were solved by the direct methods and refined by the full-matrix least-squares procedure against *F*<sup>2</sup> in anisotropic approximation. The refinement was carried out with the SHELXTL program [71]. The CCDC numbers (2070233 for 1,12-I2-C2B10H10 and 2074102 for 1-I-C2B10H11) contain the supplementary crystallographic data for this paper. These data can be found in the Supplementary Materials or obtained free of charge via www.ccdc.cam.ac.uk/data\_request/cif, accessed on 15 March 2021.

Crystallographic data for 1,12-I2-1,2-C2B10H10: C2H10B10I2 are monoclinic, space group *P*21/*c*: *a* = 7.1919(8)Å, *b* = 15.8202(17)Å, *c* = 11.1509(12)Å, *β* = 108.809(2)◦, *V* = 1201.0(2) Å3, *<sup>Z</sup>* = 4, *<sup>M</sup>* = 396.00, *<sup>d</sup>*cryst = 2.190 g·cm−3. *wR*2 = 0.0365 calculated on *<sup>F</sup>*<sup>2</sup> *hkl* for all 3557 independent reflections with 2*θ* < 60.4◦, (*GOF* = 1.067, *R* = 0.0161 calculated on *Fhkl* for 3314 reflections with *I* > 2*σ*(*I*)).

Crystallographic data for 1-I-1,2-C2B10H11: C2H11B10I are orthorhombic, space group *Pnma*: *a* = 13.8323(9)Å, *b* = 8.9644(6)Å, *c* = 8.4539(5)Å, *V* = 1048.27(12) Å3, *Z* = 4, *M* = 270.11, *<sup>d</sup>*cryst = 1.711 g·cm<sup>−</sup>3. *wR*2 = 0.0747 calculated on *<sup>F</sup>*<sup>2</sup> *hkl* for all 1344 independent reflections with 2*θ* < 56.1◦, (*GOF* = 1.143, *R* = 0.0327 calculated on *Fhkl* for 1203 reflections with *I* > 2*σ*(*I*)).

#### *3.6. Quantum Chemical Calculations*

All quantum chemical calculations were carried out with the Gaussian09 program [72]. The PBE0 functional with the triple zeta basis set was found to be reliable for the calculation of noncovalent intra- and intermolecular interactions [73–75] and was adopted throughout this study. Initial geometries for the optimization of all dimers considered in this study were based on the X-ray structure of a dihalogen-bonded dimer of 1,12-I2-C2B10H10 (symmetry code is 1 − x, –0.5 + y, 0.5 − z). All dimeric associates were fully optimized and converged to the energy minima. Theoretical electron density was treated within the AIM approach [76] using the AIMAll program package [77]. For energy (*E*) estimation, we used the *E* = 1/2*V*(*r*) formula [78,79], in which *V*(*r*) is the potential energy density at the bond critical point between interacting atoms. It has frequently been shown that this approach to describe noncovalent interactions demonstrates realistic energetic characteristics [80–82].

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/cryst11040396/s1, Figure S1: Asymmetric part of 1-I-1,2-C2B10H11 molecule showing numbering scheme and the disorder of the C2/B4 atoms; Figure S2: General view of 1-I-1,2-C2B10H11; Figure S3: Halogen bonded dimer of 1-I-1,2-C2B10H11, and complete crystallographic data (cif-files) for compounds 1-I-1,2-C2B10H11 and 1,12-I2-1,2-C2B10H10.

**Author Contributions:** Synthesis and NMR spectroscopy study, S.A.A.; quantum-chemical calculations, A.A.A.; X-ray diffraction experiment and final manuscript writing, K.Y.S.; general manuscript concept and final manuscript writing, I.B.S.; technical editing, V.I.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the Russian Science Foundation—synthesis by Grant No. 19-73-00353 and crystal packing analysis and quantum-chemical calculations by Grant No. 19-13-00238. **Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The supporting data are available on request from the authors.

**Acknowledgments:** The NMR spectroscopy and X-ray diffraction data were obtained by using equipment from the Center for Molecular Structure Studies at A.N. Nesmeyanov Institute of Organoelement Compounds, operating with support from the Ministry of Science and Higher Education of the Russian Federation.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


### *Article* **Carboranes as Lewis Acids: Tetrel Bonding in CB11H11 Carbonium Ylide**

**Maxime Ferrer 1, Ibon Alkorta 1, José Elguero <sup>1</sup> and Josep M. Oliva-Enrich 2,\***

<sup>1</sup> Instituto de Quimica Médica (CSIC), Juan de la Cierva, 3, E-28006 Madrid, Spain; maxime.ferrer@iqm.csic.es (M.F.); ibon@iqm.csic.es (I.A.); iqmbe17@iqm.csic.es (J.E.)

<sup>2</sup> Instituto de Química-Física "Rocasolano" (CSIC), Serrano, 119, E-28006 Madrid, Spain

**\*** Correspondence: j.m.oliva@iqfr.csic.es; Tel.: +34-915619400

**Abstract:** High-level quantum-chemical computations (G4MP2) are carried out in the study of complexes featuring tetrel bonding between the carbon atom in the carbenoid CB11H11—obtained by hydride removal in the C-H bond of the known *closo*-monocarbadodecaborate anion CB11H12(−) and acting as Lewis acid (LA)—and Lewis bases (LB) of different type; the electron donor groups in the tetrel bond feature carbon, nitrogen, oxygen, fluorine, silicon, phosphorus, sulfur, and chlorine atomic centres in neutral molecules as well as anions H(−) , OH(−) , and F(−) . The empty radial 2*pr* vacant orbital on the carbon centre in CB11H11, which corresponds to the LUMO, acts as a Lewis acid or electron attractor, as shown by the molecular electrostatic potential (MEP) and electron localization function (ELF). The thermochemistry and topological analysis of the complexes {CB11H11:LB} are comprehensively analysed and classified according to shared or closed-shell interactions. ELF analysis shows that the tetrel C··· X bond ranges from very polarised bonds, as in H11B11C:F(−) to very weak interactions as in H11B11C··· FH and H11B11C··· O=C=O.

**Keywords:** Lewis acid; carborane; carbonium ylide; tetrel bond; quantum chemistry; electron density; ELF

#### **1. Introduction**

The very stable B12H12(2−) dianion and its neutral dicarbon counterparts *ortho*-(1,2- C2B10H12), *meta*-(1,7-C2B10H12), and *para*-carborane (1,12-C2B10H12) are icosahedral systems that are closely related to elemental boron. Their isoelectronic analogue, *closo*monocarbadodecaborate anion CB11H12(−) , first prepared in 1967 [1] and further with other synthesis methods [2,3], is similarly resistant to cage degradation, and many derivatives have been synthesized as described in the literature [4]. The stability and threedimensional aromaticity of CB11H12(−) has also been explained using quantum-chemical computations [5]. Extraction of hydride H(−) in the C-H bond from CB11H12(−) leads to a carbocation ylide or carbenoid (**1**) with a vacant radial 2*pr* orbital on the cage carbon atom as shown in Figure 1. On the other hand, reaction mechanisms of polyhedral (car)boranes and their derivatives are scarce in the literature and further research is needed in this respect [6–8]. Thus, the permethylated carbenoid analogue CB11Me11 has been postulated as a reaction intermediate during the extraction of the L substituent from L-CB11Me11 carboranes (L = BrCH2CH2 or (CF3)2CHO) by electrophiles [6,7], further reacting with arenes in the presence of (CF3)2CHOH to generate 1-aryl-CB11Me11 products [6–8]. The methyl groups in the permethylated anion CB11Me12(−) have substantial CH3 (−) (methide) character according to DFT computations [6–8] and can easily bind to transition metal and main group elements.

On the other hand, in recent years, tetrel bonding—defined as an interaction between any electron donating system (ED) and a group 14 element acting as Lewis acid—has called the attention of both experimentalists [9] and theoreticians [10–12]. Here the carbon centre

**Citation:** Ferrer, M.; Alkorta, I.; Elguero, J.; Oliva-Enrich, J.M. Carboranes as Lewis Acids: Tetrel Bonding in CB11H11 Carbonium Ylide. *Crystals* **2021**, *11*, 391. https:// doi.org/10.3390/cryst11040391

Academic Editor: Marina Yu. Stogniy

Received: 12 March 2021 Accepted: 2 April 2021 Published: 7 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

25

in (**1**) is clearly an acceptor of electrons or Lewis acid; hence, we can define a tetrel bonding interaction with an electron donor (ED) or Lewis base, as shown in Figure 1c.

**Figure 1.** Removal of hydride from the C-H bond in (**a**) *closo*-monocarbadodecaborate anion CB11H12(−) leads to (**b**) carbocation ylide or carbenoid CB11H11 (**1**). (**c**) Complex formation between (**1**) and a Lewis base (LB). All vertices correspond to B-H moieties except for the carbon vertex.

The goal of this work is to study the electronic interaction between the naked carbon vertex in the carbenoid (**1**) with a series of electron donor molecules and anions leading to tetrel C-X bonds. The chosen 18 LB systems, including the anions H(−) , F(−) , and OH(−) , are displayed below in Scheme 1 with the corresponding label.


**Scheme 1.** The chosen set of molecules acting as Lewis base (LB) and forming a tetrel bond with the C atom from carbenoid (**1**) according to Figure 1c.

#### **2. Computational Methods**

Electronic structure quantum-chemical computations were carried out using the G4MP2 model [13], which is a fourth-generation method available in the Gaussian16 scientific software [14]. This method combines density-functional theory [15,16] and second-order perturbation theory [17] and provides an accurate and economical method for thermochemical predictions. The G4(MP2) model works as follows: The geometries of the molecules are optimized at the B3LYP/6-31G(2df,p) level of theory, and then a series of single point energy calculations at higher levels of theory are computed. The zero-point energy, E(ZPE), is based on B3LYP/6-31G(2df,p) frequencies scaled by 0.9854, the same as in G4 theory. The first energy calculation is at the triples-augmented coupled cluster level of theory, CCSD(T), with the 6-31G(d) basis set, i.e., CCSD(T)/6-31G(d). This energy is then modified by a series of energy corrections to obtain a total energy E0. For more details on the G4(MP2) method, the reader is referred to Reference [13]. In the particular case of the **1**:LB complexes, we computed the enthalpy and free energy differences between the complex and separated systems **1** and LB at room temperature as indication of stability of the complex. All complexes included in this work correspond to energy minimum structures, checked with frequency computations. The quantum theory of atoms-in-molecules (QTAIM) [18,19] was used in the topological analysis of the electron density of the **1**:LB complexes with the scientific software AIMAll [20]. This method is based on the analysis of the electron density *<sup>ρ</sup>*, its gradient ∇*ρ*, and the corresponding Laplacian ∇2*ρ*. For further aspects of this methodology, the reader is referred to the above References [17,18] and Section 3.3.1 below. The electron localisation function (ELF) [21,22] was also used in the topological analysis of the complexes. The ELF is a distribution function which measures the probability of finding two electrons with the same spin, as further described in Section 3.3.2 below. The TopMod09 package [23] was used for the ELF calculations.

#### **3. Results**

#### *3.1. Geometries of Complexes (1:n), n = 2–19*

In Figure 2a,b we display the G4MP2 optimized geometries of CB11H12(−) with C5v symmetry—the (**1**:**2**) complex and the carbonium ylide CB11H11 (**1**), with the different tetrel complexes (**1**:*n*), *n* = **2**–**19** and structural parameters when necessary, in order to highlight the atomic rearrangements undergone due to the complexation process. The loss of H(−) in the C-H leads also to a structure with C5v symmetry—with a geometrical change which involves a considerable flattening of the CB5 pentagonal pyramid with expansion of the corresponding B5 pentagon, since there is an increase of the B-B bond distance of Δ = +0.022 Å. As we proceed down the cage from the top (C atom), the B-B bond differences are +0.022 Å, +0.022 Å, and −0.015 Å. Quite noticeable is the shortening of the apical intracage C··· B distance (Δ∼−0.3 Å). The B-H bond distances hardly change at all, with a very slight shortening upon loss of H(−) with <sup>Δ</sup> <sup>=</sup> −0.007 Å, −0.008 Å, and −0.001 Å, from top to bottom, respectively. If we take the C and B cage nuclei as point charges and define a distorted C5v icosahedron, the corresponding volumes are V(CB11H12(−) ) = 12.03 Å3 and V(CB11H11) = 11.90 Å3, and therefore there is a shrinkage of the cage by <sup>Δ</sup>V = −0.13 Å<sup>3</sup> (1%). In summary, the extraction of hydride in the C-H bond from CB11H12(−) implies a minor change in the cage volume with a flattening of the top CB5 pentagonal pyramid and minor changes as we proceed down the cage from the top C atom. In Figure 2c–s, we display the optimised geometries for the remaining complexes with the coordinates gathered in the Supplementary Material (SI, Tables S1–S10). The shortest and longest C··· X tetrel interactions correspond to the original anion CB11H12(−) or complex (**1**:**2**) and the CB11H11··· O=C=O tetrel complex (**1**:**13**), respectively. In the latter case, the interaction is clearly non-covalent in origin with d(C··· O) = 2.693 Å, as will be discussed later on. We should also emphasize the C··· X interaction of the CH2 and SiH2 complexes (**1**:**3**) and (**1**:**16**), respectively, with the LB groups tilted from the C5 axis of rotation. In all other systems, including the CF2 complex (**1**:**4**), the C··· X bond is aligned with the C5 axis of rotation.

**Figure 2.** Structures of the G4MP2 optimized geometries for the 18 tetrel complexes (**1**:*n*)—with *n* = **2**–**19**—considered in this work, following Scheme 1: (**a**) The known anion CB11H12(−) corresponding to complex (**1**:**2**), (**b**) carbenoid CB11H11 (**1**), (**c**) (**1**:**3**), (**d**) (**1**:**4**), (**e**) (**1**:**5**), (**f**) (**1**:**6**), (**g**) (**1**:**7**), (**h**) (**1**:**8**), (**i**) (**1**:**9**), (**j**) (**1**:**10**), (**k**) (**1**:**11**), (**l**) (**1**:**12**), (**m**) (**1**:1**3**), (**n**) (**1**:**14**), (**o**) (**1**:**15**), (**p**) (**1**:**16**), (**q**) (**1**:**17**), (**r**) (**1**:**18**), and (**s**) (**1**:**19**).

In Figure 3, we display the d(C··· X) distances in the tetrel bonding complexes, ordered from shortest to longest. Clearly, we can classify five groups according to the C··· X distances, in increasing order: (i) The original anion CB11H12(−) or (**1**:**2**) complex; (ii) com-

plexes with d∼1.4–1.5 Å including complexes (**1**:**k2**), with k2 = (3–12, 14); (iii) complexes with d∼1.8 Å, including complexes (**1**:**k3**), with k3 = 16–18; (iv) complexes with d∼2.0 Å, including complexes (**1**:**15**) and (**1**:**19**); and finally (v) the (**1**:**13**) complex with d∼2.7 Å.

**Figure 3.** C··· X distances dCX (Å) in the tetrel complexes (**1**:*n*), *n* = **2**–**19**, from Figure 2, in increasing order; x axis corresponds to Lewis base (LB) *n* and y axis to dCX distances (Å), respectively.

#### *3.2. Thermochemistry of Complexes (1:n), n = 2–19*

In this subsection, we predict the ΔH and ΔG of the tetrel bonding complexes (**1**:*n*), *n* = **2**–**19**. In Table 1, we gather the computed enthalpies and free energies at the G4MP2 level of theory.

**Table 1.** Enthalpy (ΔH) and free energy (ΔG) of formation for complexes (**1**:*n*), *<sup>n</sup>* <sup>=</sup> **<sup>2</sup>**–**19**, in kJ·mol<sup>−</sup>1. LB: Lewis base.


The free energy of formation for complexes (**1**:*n*) is always negative except for (**1**:**13**) and (**1**:**15**) complexes, namely the O=C=O and FH complexes, respectively. Small negative values (|ΔG| < 10 kJ·mol<sup>−</sup>1) are obtained for complexes (**1**:**6**) and (**1**:**19**) with Lewis bases N2 and ClH, respectively. Therefore, all complexes with negative ΔG should be formed at room temperature spontaneously, provided an isolated Lewis acid (**1**) approaches an isolated Lewis base. On the other hand, the enthalpies of formation are negative for all complexes, an indication that the bond energies of a given complex (**1**:*n*) have a lower value than the bond energies of separated systems (**1**) and (*n*). In order to better visualize

the similarities and differences of the thermochemical aspects of complexes (**1**:*n*), we plot ΔG and ΔH vs. *n* in increasing order of each state function.

As shown in Figure 4 from left to right in the abscissa, the ΔG (black) and ΔH (blue) in complex formation follow the same order as function of Lewis base number except for ClH (**1**:**19**) and N2 (**1**:**6**), where the order is inverted in the ΔH tendency as compared to ΔG. Hence, the formation of anionic complexes are the most energetic and favourable ones in the order (**1**:**2**), (**1**:**10**), (**1**:**14**) corresponding to Lewis bases H(−) , OH(−) , and F(−) , respectively; then follow complexes (**1**:**3**), (**1**:**16**), and (**1**:**4**) corresponding to Lewis bases CH2, SiH2 and CF2, respectively, namely the carbene series. A plateau with complexes (**1**:**8**), (**1**:**17**), and (**1**:**7**) follows with Lewis bases NH=CH2, PH3, and NH3, respectively. A smaller (positive) slope of ΔG/ΔH vs. *n* appears with complexes (**1**:**5**), (**1**:**18**), (**1**:**9**), (**1**:**12**), and (**1**:**11**) always in increasing order, which correspond to Lewis bases C≡O, SH2, N≡CH, O=CH2, and OH2, respectively. The weakest bound complexes with ΔG < 0 correspond to (**1**:**19**) and (**1**:**6**) with Lewis bases ClH and N2, respectively. Finally, complexes (**1**:**13**) and (**1**:**15**) with Lewis bases O=C=O and FH, respectively, show a predicted quantum-chemical value of ΔG > 0, and therefore one should not expect a spontaneous formation of these complexes at room temperature. It is noteworthy to mention the tiny value <sup>Δ</sup>H(**1**:**15**) = −0.7 kJ·mol−<sup>1</sup> for FH attachment to (**1**); this number is within the accuracy of the method and therefore a heat of formation for complex (**1**:**15**) or the bond energy on both sides of the equation remains unaltered.

**Figure 4.** <sup>Δ</sup>G and <sup>Δ</sup>H vs. Lewis base *<sup>n</sup>* in respective increasing order, all in kJ·mol<sup>−</sup>1.

#### *3.3. Electronic Structure of Complexes (1:n), n = 2–19*

In Figure 5, we show for (**1**) the molecular electrostatic potential (MEP) and the electron localization function (ELF). These electronic structure features are computed using the optimized geometry of the system with the G4MP2 method—B3LYP/6-31G(2df,p) model chemistry for structure optimization. As noticed in Figure 5a, the shape of the MEP and the corresponding π-hole just on top of the C ylide centre shows the electron-attraction nature of this region of the molecule. In the ELF from Figure 5b, we show disynaptic V(B,H) yellow basins corresponding to the B-H bonds; the ELF distribution around the CB11 icosahedral cage can be partitioned into green disynaptic and trisynaptic basins, as we will describe below in Section 3.3.2.

**Figure 5.** Electronic structure of carbonium ylide (**1**) CB11H11: (**a**) Molecular electrostatic potential V(**r**). Red colour for V(**r**) < −0.015 au, Blue colour for V(**r**) > 0.03 au. The black dot indicates the localization of the π-hole (0.061 au), and (**b**) Electron Localization Function (ELF) with an isosurface of ELF = 0.75. Computations with the G4MP2 level of theory.

The molecular electrostatic potential (MEP) is the potential energy of a proton at a particular location near a molecule. Negative electrostatic potential corresponds to an attraction of the proton by the concentrated electron density in the molecules. The MEP of (**1**)—Figure 5b—shows that the potential energy of a proton is most positive above the C atom with a π-hole of +0.061 au, hence a repulsive region for a proton approaching (**1**), or electron acceptor region. The MEP is smoothly changing from positive to negative values of the potential energy as the proton moves from the C atom down to the B skeleton cage region. A proton would then be attached more favourably to the lower region of the carbonium ylide (**1**). In other words, Lewis bases, electron donors, and nucleophiles should then tend to bind through the C atom of the ylide, hence the study of the tetrel bonding in the complexes (**1**:*n*).

#### 3.3.1. Atoms-in-Molecules (AIM) Topological Analysis of Complexes (**1**:*n*), *n* = **2**–**19**

The Quantum Theory of Atoms in Molecules (QTAIM) [18,19] is a useful tool for analysing the electronic structure of a polyatomic many-electron system, with the electron density *ρ*(**r**) as the central function. The topological properties of *ρ*(**r**) are analysed with the gradient of ∇*ρ*(**r**), the Laplacian of *<sup>ρ</sup>*(**r**), ∇2*ρ*(**r**), and the eigenvalues of the Hessian matrix of the electron density <sup>λ</sup>1, <sup>λ</sup>2, <sup>λ</sup>3. The critical points are those with <sup>→</sup> <sup>∇</sup>*<sup>ρ</sup>* <sup>=</sup> <sup>→</sup> 0 and a bond critical point (BCP) has λ<sup>3</sup> > 0 associated with the bond path direction, and λ<sup>1</sup> < 0, λ<sup>2</sup> < 0 the two latter associated to two directions where ∇2*<sup>ρ</sup>* is a maximum; the BCP (−,−,+) appears at the intersection of the bond path with the interatomic surface S. Other critical points are classified according to the signs of λ*i*: Nuclei positions with (−, −, −); ring critical points with (−,+,+); cage critical points with (+,+,+). We should also introduce the local electron kinetic (G > 0), potential (V < 0), and total (H) energy densities, with H = G + V, also useful parameters at the BCP for the description of the type of bonding interaction between atoms in a many-electron system [24]. In the SI (Table S11), we provide the computed values of *<sup>ρ</sup>*(**r**), ∇*ρ*(**r**), ∇2*ρ*(**r**), G, V, and H for the BCP found between the X atom of the Lewis base in contact with the C ylide centre in (**1**), for all complexes (**1**:*n*), *n* = **2**–**19**.

In Figure 6a, we plot the electron density at the BCP for the C··· X interaction vs. *d*(CX) distance. The largest values of *ρ*BCP correspond to CH2 (*ρ*BCP(CH2) = 0.32 *e*/*a*<sup>0</sup> 3), CF2, and CO, followed by H(−) , OH(−) , NHCH2, N2, and F(−) . Another group follows with lower values, NH3, OCH2, PH3, OH2, SH2, and further down, SiH2 and ClH with similar values. Finally, the lowest *ρ*BCP correspond to FH and CO2, the latter with *ρ*BCP(CO2) = 0.01 *e*/*a*<sup>0</sup> 3. We should emphasize that the ratio *ρ*BCP,max(CH2)/*ρ*BCP,min(CO2) = 32 gives an idea of the topological differences in these BCPs. Given the different type of C··· X interactions in the complexes, the *ρ*BCP vs. *d*(CX) can be fit to an approximate negative exponential curve with *ρ*BCP (*d*CX) = *a* + *b*·exp(−*c*·*d*CX), with *a* = −0.022, *b* = +3.222, and *c* = −1.751, and a

correlation factor of R<sup>2</sup> = 0.99 for closed-shell interacting complexes: CO2, FH, SiH2, N2, and NCH. This curve is displayed in the SI (Figure S1). In general, very good correlations appear if we fix the two interacting atoms both belonging to the same row of the Periodic Table [25–27].

**Figure 6.** *Cont.*

**Figure 6.** (**a**)**.** *ρ*(**r**) at the BCP of the C··· X interaction in complexes (**1**:*n*) vs. *d*(CX), with *n* = **2**–**19**. We label each point with the corresponding LB. (**b**) <sup>∇</sup>2*ρ*(**r**) at the C··· X BCP for complexes (**1**:*n*), *<sup>n</sup>* <sup>=</sup> **<sup>2</sup>**–**19**. (**c**) Plot of G (-), V (), and H (•) vs. Lewis base LB(*n*) in increasing order of G, V, and H, respectively, at the Bond Critical Point (BCP) in the C··· X interaction of complexes (**1**:*n*), *n* = **2**–**19**, as described in Scheme 1.

In order to estimate the type of interaction we need to go beyond the electron density at the BCP and analyse the second derivative, the Laplacian ∇2*ρ*, and the kinetic, potential, and total energy, G, V, and H respectively, of the BCPs in complexes (**1**:*n*). In Figure 6b, the Laplacian is plotted vs. LB(*n*) in increasing order. Clearly, we can distinguish the shared interactions for ∇2*<sup>ρ</sup>* < 0 in the lower left corner and closed-shell interactions for ∇2*<sup>ρ</sup>* > 0 in the upper right corner of Figure 6b. The two-electron sharing in complexes with H(−) , CO, CF2, and CH2 is large, and it is diminished up to OH2. For the complex with HCl, ∇2*<sup>ρ</sup>* <sup>=</sup> −0.0067 *<sup>e</sup>*/Å5, namely in the limit between shared and closed-shell interactions. For positive Laplacians, in increasing order, the LB in the (**1**:*n*) complexes correspond to: NCH, CO2, FH, SiH2, and N2; in these systems, the closed-shell interactions are important.

A further analysis of the BCP in the C··· X interactions of the (**1**:*n*) complexes can be found in the values of G, V, and H—with H=G+V being the total energy—as displayed in Figure 6c. The kinetic energy G is associated with repulsion in the bonding region, and the local potential energy density or local virial field V is a measure of the average effective potential field experienced by a single electron in a many-particle system. Thus, according to Figure 6c, the G and V profiles are inverted for N2, NCH, F(−) , and OH(−) , but due to the nature of different nuclei in the C-X interactions, this is not always the case, as seen when we follow the profiles as function of LB(*n*).

#### 3.3.2. Electron Localisation Function (ELF) Analysis of Complexes (**1**:*n*), *n* = **2**–**19**

We should emphasize that ELF is a function which reports the probability of finding an electron pair with opposite spins in a region of space. Using a certain isovalue, we are able to define regions of space, basins, with a certain probability to find an electron pair. For example, in the plots of the ELF, we used an isovalue of 0.83; in other words, we plot regions of space where we have a high probability to find a pair of electrons. Once the basins are defined, we can integrate the electronic density into those basins, which are the values reported above in Table 2 and correspond to the number of electrons in that basin. In ELF analysis, the partition of space is not based on the electron density, as in AIM, but on the ELF probability function. In order to better understand from the electronic

structure point of view the tetrel bonding in the (**1**:*n*) complexes, we further computed the electron localisation function (ELF) [21,22,28], a measure of the likelihood of finding an electron in the neighborhood space of a reference electron located at a given point and with the same spin; therefore, ELF is a measure of the Pauli repulsion or exchange interaction [29,30]. The ELF for the carbonium ylidene CB11H11 (**1**) is shown in Figure 4. ELF values ranges from zero to one (normalized and without units). In the SI file we provide the ELF for all complexes (**1**:*n*) not shown here (Table S12), and below we have selected four cases with short, medium, long, and very long C··· X distances, according to Figure 3 above: (i) LB = H(−) , complex (**1**:**2**), (ii) LB = N2, complex (**1**:**6**), (iii) LB = PH3, complex (**1**:**17**), and (iv) LB = CO2, complex (**1**:**13**). In Table 2, we gather the ELF function for these four complexes.

**Table 2.** Electron localisation function (ELF) (isovalue 0.83 au) for complexes H11B11C:H(−) <sup>→</sup> (**1**:**2**), H11B11C:N≡<sup>N</sup> <sup>→</sup> (**1**:**6**), H11B11C:PH3 → (**1**:**17**), and H11B11C:O=C=O → (**1**:**13**). Basin labels are depicted for each complex. Population of ELF disynaptic basins V(C, X) in bold.


Below each ELF function of a given complex (**1**:*n*), we also report the function value and average population for the different types of basins. A threshold of 0.2 electrons is considered as to include or not a basin in a group. Below each ELF function of a given complex (**1**:*n*), we report the population of the different basins. In order to avoid dealing with a long list of populations and after observing that basins involving the same elements have similar populations, we decided to report only for each type of basins the average population. A threshold of 0.2 electrons was chosen to decide if two basins belong to the same group or not. For instance, if we consider basins V(B1,B2) and V(B1,B3) with populations of 0.8 and 1.5 electrons, respectively, they belong to different groups. In bold letters, we report the value of the disynaptic basin corresponding to the tetrel C··· X interaction.

In Table 2 we use the following notation:


• 5 V(B,B) = 0.42: There are 5 disynaptic basins involving two boron atoms with an average population of 0.42 electrons.

In the ELF representation, the following colours are used:


Thus, according to Table 2, in the complex (**1**:**2**), the existing anion CB11H12(−) , the 11 core electrons from the boron cage, the 1*s*2(B) ones, are gathered in the C(B) core basins, which have an average population of 2.07 electrons; the same description applies for the single C atom and the C(C) = 2.11 basin. The 11 B-H bonds on each non-naked B vertex of the icosahedron correspond to the 11 V(B,H) = 2.05, basically a two-electron covalent B-H bond. Then the distribution of the remaining valence cage electrons (corresponding to 2*s* and 2*p* electrons from B and C, plus the surplus electron or negative charge of the anion) are distributed in the 10 V(B,B,B) trisynaptic basins with an average population of 0.81 electrons, the 15 V(B,B) disynaptic basins with a population of 0.76 electrons, and the 5 V(C,B) disynaptic basins with ∼1 electron in them. The V(C,H) disynaptic basin, with a population of 2.05, corresponds to the C··· X = C-H bond in (**1**:**2**), a two-centre two-electron bond. Addition of the values of all basins leads to the number of electrons in CB11H12(−) : 74.

As reported in Table 2, the C··· X interaction in the selected complexes can be described by the presence or absence of V(C,X) valence basins and its population. For the complexes (**1**:**6**) and (**1**:**17**), these values are V(**1**:**6**)(C,N) = 2.62 and V(**1**:**17**)(C,P) = 2.21, respectively. Therefore, ELF describes the C··· X for the N2 complex as a bond, with a multiplicity close to 1.5, between the C(ylide) and N nuclei and for the PH3 complex a C(ylide)P single bond, with additional 0.2 electrons. As regard to the (**1**:**13**) complex with CO2, the ELF does not localize a basin between the C(ylide) and the O=C=O molecule; the absence of ELF basins indicate that electron pairs are not shared, and therefore, the interaction is not covalent. However, other interactions such as ionic or non-covalent are possible even in the absence of ELF basins. The lone pairs from N2 and CO2 appear as red monosynaptic basins, as displayed in Table 2.

#### **4. Discussion**

The presence of a filled or empty lone pair on the C atom in the known anion CB11H12(−) , complex (**1**:**2**), depends on whether we remove a proton or a hydride from the C-H bond, leading to a dianion [CB11H11] (2−) (**1b**) or a carbonium ylide CB11H11 (**1**), respectively; the latter process is shown in Figure 1. The tetrel complexes (**1**:*n*) presented in the previous section show a rich variety of thermochemical and electronic structure features with tetrel C··· X interactions from different nuclei: X = {H, C, N, O, F, Si, P, Cl}. The C(ylide) centre in (**1**) confers to this particular molecule with a Lewis acid (LA) character, hence the tetrel denomination. The strength for electronic attachment in (**1**) is given by the computed free energy of formation (**1**)+(*n*) → (**1**:*n*). The strongest complexes correspond to those formed with anions H(−) , OH(−) , and F(−) , and the weakest complexes to those formed with FH, CO2, N2, and ClH. The C··· X distance varies considerably in all complexes, ranging from 1.081 Å for (**1**:**2**), LB = H(−) , to 2.694 Å for (**1**:**13**), LB = CO2. The complex strength is not related to the C··· X distance; namely, complexes with similar C··· X distances may have different free energy of formation, e.g*.,* the free energy of formation for complexes (**1**:**14**) LB = F(−) and (**1**:**6**) LB = N2 is −530 kJ·mol−<sup>1</sup> and −3 kJ·mol−1, respectively, with very similar d(C-X) distances, 1.364 Å and 1.375 Å.

Examples of recent related systems is the 3D analogue of phenyllithium, the lithiacarborane CB11H11:Li(−) , studied in solution as a solid and by quantum-chemical computations [31]. Indeed, Li(−) is a very poor Lewis base but certainly attaches to (**1**), as

recently shown, and defined as the lithiated mono-anion [Li−CB11H11] (−) . On the other hand, this process can also be seen as a carborane dianion [CB11H11] (2−) —a very reactive species—attached to Li(+), as described in Reference [31]. Table 8-1 from the book by Grimes, *Carboranes* [4], reports hundreds of compounds derived from CB11H12(−) , and therefore this is a rich field not only from a synthetic point of view but also for studying the electronic structure of tetrel C··· X bonds in these compounds, and especially if the isolated tetrel complexes (**1**:*n*) could ever be synthesized, taking into account that this work is purely theoretical with predictive quantum-chemical computations.

The electronic structure of the complexes has been analysed thoroughly with AIM and ELF methods, showing the C··· X sharing and closed-shell interactions in the complexes according to the values of the Laplacian of the electron density. In Table 3, we gather the ELF values for disynaptic basins V(C, X) in the C··· X region showing values of ELF: we can find very polarised C-F bonds in (1:14)—only one electron in the C··· X region—single C-X bonds for H(−) and NH=CH2 and intermediate cases, such as in complex (1:5) with a 1.5 multiplicity C-C bond for the CO complex. No V(C,X) disynaptic basins are found for CO2 and FH, an indication of the poor electron-donating ability of these Lewis bases (LB) with indeed long *d*(C-X) distances and positive free energies of formation ΔG > 0, hence confirming the unlikely formation of these two complexes.

**Table 3.** Population of ELF disynaptic basins V(C, X) in tetrel complexes (**1**:*n*), *n* = **2**–**19**, describing the C··· X interaction. LB = Lewis base.


According to the Cambridge structural database(CSD) [32], several tetrel complexes derived from (1) have been characterised [33–39] where the C(ylide) centre interacts with a nitrogen atom from neutral aminoderivatives including pyridine. These structures are shown in the SI file as Table S13. The shortest C(ylide)··· N distance, *d*(C··· N) = 1.477 Å, corresponds to the pyridine complex 1-(4-methoxypyridinium)-1-carba-*closo*-dodecaborane [33]. The longest C(ylide)··· N distance, *d*(C··· N) = 1.554 Å, corresponds to the complex 12 iodo-1-(4-pentylquinuclidine)-1-carba-*closo*-dodecaborane [34]. There is a tetrel complex of (**1**) with NH3, 1-amino-2-fluorocarba-*closo*-dodecaborane [35], where one B-H vertex hydrogen atom on position 2 has been substituted by a fluorine atom with *d*(C··· N) = 1.486 Å. Our (**1**:**7**) tetrel complex H11B11C ←:NH3 has a predicted *d*(C··· N) = 1.498 Å according to the G4MP2 computational model.

#### **5. Conclusions**

The results presented in this work show that by means of quantum-chemical computations we should expect the formation of tetrel complexes between the icosahedral carbonium ylide CB11H11—derived from extraction of H(−) in the known anion CB11H12(−) —and a set of simple molecules and anions. The driving force of formation for these complexes can be accounted for from thermochemical quantum-chemical computations using statistical mechanics implemented in the scientific software Gaussian16 [14], and the results indicate that all the complexes should be formed with the exception of the FH and CO2 molecules, with N2 and ClH complexes with indeed very low, though negative, free energies of formation.

The tetrel C··· X interactions in all complexes have been thoroughly studied by means of AIM and ELF methods, hence defining the type of bond and interaction, ranging from very polarised bonds, with one electron in the C··· X moiety, to intermediate cases as in the carbenes CH2 and CF2 and silane SiH2, with one and a half electrons in the C··· X region.

The existence of known tetrel complexes of the carbonium ylide CB11H11 with amino derivatives, including pyridine, opens the door toward further experimental and theoretical studies in the electronic structure of unusual bonds and interactions between C(ylide) centres in carboranes and other atoms.

We hope that the results from this work can be used for the isolation of reactive species, such as the recently found dianion derived from proton extraction in the wellknown carborane anion CB11H12(−) , a key molecule in the description of 3D aromaticity within boron chemistry.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/10 .3390/cryst11040391/s1, Tables S1–S10. G4MP2 optimised geometries of complexes (**1**:*n*). Table S11. AIM data for complexes (**1**:*n*). Table S12. ELF data for complexes (**1**:*n*) not displayed in the main text.

**Author Contributions:** Writing-original draft preparation, computations, and data curation: J.M.O.- E., M.F., I.A., and J.E. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Spanish MICINN, grant number CTQ2018-094644-B-C22 and Comunidad de Madrid, grant number P2018/EMT-4329 AIRTEC-CM.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Coordination Ability of 10-EtC(NHPr)=HN-7,8-C2B9H11 in the Reactions with Nickel(II) Phosphine Complexes**

**Marina Yu. Stogniy 1,\*, Svetlana A. Erokhina 1, Kyrill Yu. Suponitsky 1,2, Igor B. Sivaev 1and Vladimir I. Bregadze <sup>1</sup>**


**Abstract:** The complexation reactions of *nido*-carboranyl amidine 10-PrNHC(Et)=HN-7,8-C2B9H11 with different nickel(II) phosphine complexes such as [(PR2R')2NiCl2] (R = R' = Ph, Bu; R = Me, R' = Ph) were investigated. As a result, a series of novel half-sandwich nickel(II) π,σ-complexes [3-R'R2P-3-(8-Pr*N*=C(Et)NH)-*closo*-3,1,2-NiC2B9H10] with the coordination of the carborane and amidine components was prepared. The acidification of obtained complexes with HCl led to the breaking of the Ni-N bond with formation of nickel(II) π-complexes [3-Cl-3-R'R2P-8-PrNH=C(Et)NH*closo*-3,1,2-NiC2B9H10]. The crystal molecular structure of [3-Ph3P-3-(8-Pr*N*=C(Et)NH)-*closo*-3,1,2- NiC2B9H10] was determined by single crystal X-ray diffraction.

**Keywords:** boron chemistry; *nido*-carborane; nitrilium derivatives; nickel(II) half-sandwich complexes; synthesis; structure

#### **1. Introduction**

The dicarbollide dianion [7,8-C2B9H11] <sup>2</sup>−, which is the deprotonated form of 7,8 dicarba-*nido*-undecaborate anion [7,8-C2B9H12] − (*nido*-carborane), is known as the inorganic isolobal analogue of the cyclopentadienyl ligand. This makes it the perfect building block in complexation reactions with a wide range of transition metals [1–3]. The possibility to substitute hydrogen atoms at the carbon and boron vertices of the carborane cage with various functional groups [4] makes it possible to vary the properties of ligands based on *nido*-carborane by combining the properties of the *nido*-carborane nest with the properties of an exo-polyhedral substituent. One of the most interesting and promising tasks in this area is the synthesis of heterobifunctional *nido*-carborane-based ligands, that can give a firm bound to capture the metal center along with a weak bond, temporarily protecting a metal coordination site. This allows users to obtain labile complexes of transitional metal representing a promising new type of catalysts [5–10] and molecular switches [11,12]. There are several examples of such stable metal complexes based on *nido*-carborane with a side substituent coordinated through oxygen or nitrogen [5–8,13–15]. The utility of such bifunctional ligand systems with the nitrogen donor atom in the side chain has been demonstrated by the complexation of [7-Me2NCH2-7,8-C2B9H11] - with metals such as nickel [16], iron [17], rhuthenium [17], titanium, zirconium, and hafnium [18,19]. In all cases, the intramolecular coordination of the dimethylamino group of the side substituent with the complexing metal was observed. The possibility of disrupting this coordination in the nickel(II) complexes by displacing the amino group with other soft ligands, such as triethylphosphine or *tert-*butylisocyanide, has been shown [16].

Earlier, we prepared a series of *nido*-carborane-based amidines 10-R(CH2)nNHC(Et)=HN-7,8-C2B9H11 using the reaction of nucleophilic addition of amines to the 10-propionitrilium derivative of *nido*-carborane [20]. Therefore, it was of interest to study the possibility

**Citation:** Stogniy, M.Y..; Erokhina, S.A.; Suponitsky, K.Y..; Sivaev, I.B.; Bregadze, V.I. Coordination Ability of 10-EtC(NHPr)=HN-7,8-C2B9H11 in the Reactions with Nickel(II) Phosphine Complexes. *Crystals* **2021**, *11*, 306. https://doi.org/10.3390/ cryst11030306

Academic Editor: Alexander Kirillov

Received: 28 February 2021 Accepted: 16 March 2021 Published: 19 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

of using obtained amidines in the complexation reactions, where *nido*-carborane itself represents a firm π-acceptor and amidine nitrogen can act as an intramolecular protecting group. In this contribution we report synthesis of a series of new metallacarboranes by the reactions of *nido*-carborane-based amidine 10-PrNHC(Et)=HN-7,8-C2B9H11 with nickel(II) phosphine complexes [(R2R'P)2NiCl2].

#### **2. Results and Discussion**

Recently, we have shown the ease and simplicity of the modification of the *nido-*carborane moiety via nucleophilic addition reactions of various nucleophiles to the highly activated nitrilium group -N+≡C-R attached to the cluster [20–22]. The nucleophilic addition of aliphatic and aromatic amines to the 10-propionitrilium derivative of *nido*-carborane (10- EtC≡N-7,8-C2B9H11) leads to the formation of compounds that are a combination of the *nido*-carborane nest with the amidine fragment [20]. This promises the possibility of synthesizing complexes with simultaneous coordination of the carborane and amidine components. For our study we chose amidine 10-PrNHC(Et)=HN-7,8-C2B9H11 (**1**) prepared by the nucleophilic addition of propyl amine to 10-EtC≡N-7,8-C2B9H11. For the complexation reactions, we used nickel(II) phosphine complexes [(R2R'P)2NiCl2] with ligands having different steric parameters (Tolman cone angles, *θ*)—R2R'P = Me2PhP (*θ* = 122◦), PBu3 (*θ* = 132◦), PPh3 (*θ* = 145◦) [23].

The addition of nickel(II) phosphine complexes [(PR2R')2NiCl2] (R = R' = Ph, Bu; R = Me, R' = Ph) to a solution of the deprotonated amidine **1** in tetrahydrofuran at ambient temperature immediately led to a color change of the reaction mixtures color from pale yellow to dark red. Monitoring the progress of the reactions using thin layer chromatography showed that complexation occurs very quickly and is completed within 5-10 minutes. The column chromatography purification gave the corresponding π,σ-complexes [3-R2R'P-3-(8- Pr*N*=C(Et)NH)-*closo*-3,1,2-NiC2B9H10] (R = R' = Ph (**2**), R = Me, R' = Ph (**3**) and R = R' = Bu (**4**)) in 80–83% yields (Scheme 1).

**Scheme 1.** Synthesis of nickelacarboranes **2**–**4** and **5**–**7**.

The initial analysis of the proposed structure of complexes **2**–**4** was carried out using standard methods of NMR and IR spectroscopy and mass spectrometry. The 1H NMR spectra of complexes **2**-**4** demonstrated the presence of only one N*H* signal of the amidine substituent in the region of 5.04–5.53 ppm, as well as the absence of signals of the *nido*-carborane B-*H*-B bridge, suggesting that nickel was coordinated both by the pentagonal face of the carborane ligand and by one atom nitrogen of the amidine group. The pattern of the 11B NMR spectra of complexes **2**–**4** is characteristic for metallacarboranes and consists of one singlet at ~4.0 ppm from the substituted boron atom and a set of four (complex **2**) or five (complexes **3** and **4**) doublets in the region from −11.2 to −27.4 ppm with total integral ratio of 1:2:3:2:1 (for **2**) and 1:2:2:1:2:1 (for **3** and **4**). The 1H NMR spectra of complexes **2**-**4** also indicated the presence of a single phosphine ligand, while its 13C NMR spectra demonstrated the characteristic splitting of signals from aromatic and/or aliphatic groups of phosphine ligands. In the 31P NMR spectra, the signals of phosphine

ligands appeared at 29.3 ppm for **2**, at −2.6 ppm for **3** and at 11.9 ppm for **4**. Such chemical shifts were in good agreement with the data for similar phosphine complexes of other transitional metals [24].

In the IR spectra of complexes **2**–**4**, the NH stretching bands were observed in the region of 3447–3244 cm<sup>−</sup>1, whereas the BH stretching bands appeared at ~2530 cm−1. The bands corresponding to the N=C bond were at ~1635 cm−<sup>1</sup> for **2** and **3**, and at 1622 cm−<sup>1</sup> for **4**. The mass spectra of complexes demonstrated only peak envelopes corresponding to molecular picks of the supposed structures of complexes **2**–**4**.

The suggested structures of complexes **2**–**4** were confirmed by single crystal X-ray diffraction study on complex **2** (Figure 1).

**Figure 1.** General view of the nickel(II) complex [3-Ph3P-3-(8-Pr*N*=C(Et)NH)-*closo*-3,1,2-NiC2B9H10] (**2**). Thermal ellipsoids are given at 50% probability level. Shortened contacts are shown by dashed lines.

The crystals of complex **2** suitable for X-ray analysis were obtained by slow evaporation from chloroform/hexane (3:1) solution. The nickel atom in the structure of **2** is approximately centered over the pentagonal face of the dicarbollide ligand with the Ni-C2B3 centroid distance of 1.526(2) Å. This is close to the distance found in a similar complex [3-Ph3P-3-(8-MeOCH2CH2*N*=C(Et)NH)-3,1,2- NiC2B9H10] (**2\***) which differs only by a substituent at the N2 atom (1.533 Å) [25] and noticeably longer than in [3- Ph3P-3-(1-Me2NCH2)-*closo*-3,1,2-NiC2B9H10] (1.500 Å) [16], and shorter than in [3-Ph3P-3-Cl-1-(*i*-PrNH)2C-3,1,2-NiC2B9H10] (1.564 Å) [26] and [3,3-(Ph3P)2-*closo*-3,1,2-NiC2B9H11] (1.610 Å) [27]. The Ni-N and Ni-P bond lengths (1.914(2) and 2.2555(7) Å) are close to those found in **2\*** (1.916 and 2.2532 Å) [25] and differ significantly from those found in [3-Ph3P-3- (1-Me2NCH2)-*closo*-3,1,2-NiC2B9H10] (2.061 and 2.166 Å, respectively) [16], reflecting the stronger electron-donating properties of the carboranylamidine ligand as compared to the carborane ligand with pendant NMe2 group. A more detailed comparison of the structures of the complexes **2** and **2\*** is presented in Table 1.


**Table 1.** Selected torsion angles (deg.), Ni-C2B3(centroid) distance and nonbonded intramolecular shortened contacts (Å) that define molecular conformation for experimental (X-ray) and calculated structure of complex **2** and comparison with **2\***.

\* for noncovalent contacts, their attractive energies in kcal/mol are given in parenthesis.

The Ni-C2B3 centroid distances were only slightly different, and differences in orientation of the PPh3 fragment were not so pronounced (within 15◦ of rotation about Ni-P bond). The most significant unequivalence, as expected, was observed for substituents at the N2 and C3 atoms. In spite of that, the system of shortened contacts was quite similar. One can suggest that the observed conformation can be stabilized by intramolecular noncovalent interactions. To confirm that, we carried out quantum chemical calculations of complex **2**. In optimized structure, torsion angles, which define the orientation of PPh3 fragment, differ by ca. 18◦ while differences in orientation of the ethyl and propyl groups are less pronounced. Again, the system of shortened contacts, for which bond critical point were localized, was still nearly the same. Those contacts in total added −5.3 kcal/mol to the stabilization of molecular conformation. These results suggest that the variation of substituents at the N2 and C3 atoms would not significantly affect the orientation of the PPh3 fragment relative to the carborane cage.

An attempt to obtain suitable X-ray diffraction study crystals of **3** and **4** by recrystallization from chloroform unexpectedly led to a change in the color of the solution from dark red to amaranth after ~12 h. Thin layer chromatography confirmed the formation of a new product together with the presence of small amounts of original complexes **3** and **4**. New complexes **6** and **7** were isolated by column chromatography on silica using dichloromethane as an eluent. An analysis of the NMR spectra of complexes **6** and **7** led to the assumption that the metal atom in the obtained complexes was no longer coordinated by the amidine group (Figure 2).

42

In the 1H NMR spectra of complexes **6** and **7**, signals from the second N*H* proton appeared in low field at 10.37 and 9.90 ppm for **6** and **7**, respectively (Figure 2, items a,b). This signal gave cross-pick in the 1H-1H COSY NMR spectrum with the methylene group of the propyl group -C*H*2CH2CH3 (See SI). The 11B NMR spectra of **6** and **7** confirmed the retention of the metallacarborane skeleton, however their spectral patterns differed from those for complexes **3** and **4** (Figure 2, items c,d). Since the newly formed complexes were neutral, we assumed that the violation of the coordination of the amidine fragment was caused by the protonation of the second nitrogen atom, and the electroneutrality of the complexes was achieved due to the coordination of the chloride ion by the nickel atom. The driving force behind this process could be the trace amounts of hydrogen chloride normally present in chloroform. To verify this assumption, we resynthesized complexes **3** and **4**, dissolved them in acetonitrile and acidified them by small amounts of concentrated hydrochloric acid (Scheme 1). This resulted in an immediate change in color of the complexes from dark red to amaranth (Figure 3). The NMR spectra confirmed the formation of complexes **6** and **7** (See SM).

**Figure 3.** The color of complex **4** (left) and complex **7** (right).

For complex **2**, which did not change upon standing in chloroform solution, we carried out a similar acidification procedure with hydrochloric acid. The solution immediately changed its color from dark red to amaranth, but, unlike complexes **3** and **4**, the transformation of complex **2** into a similar complex **5** was not complete. According to the NMR spectroscopy data, the reaction mixture contained approximately 90% of complex **5** and ~10% of the original complex **2** and the addition of more amounts of hydrochloric acid did not change this ratio. An attempt to purify complex **5** by column chromatography on silica gel with dichloromethane as an eluent resulted in a mixture of complexes **2** and **5** with a new ratio of ~5:2, which can be caused by the presence of equilibrium between these complexes and the partial loss of chloride ions on the column (Scheme 2).

**Scheme 2.** Behavior of complex **5** during purification on column chromatography with silica gel.

We supposed that the less donor PPh3 ligand made the Ni-N bond in complex **2** stronger than in complexes **3** and **4** with more donor phosphine ligands PMe2Ph and PBu3.

Like complexes **6** and **7**, the 1H NMR spectrum of complex **5** contained the signals of two different N*H* protons: one at 9.09 ppm from the N*H*Pr group and another one at 5.53 ppm from the B-N*H*-C fragment. However, in contrast to complexes **2**–**4** in the 1H NMR spectra of **5**–**7**, the signals of the C*H*carb groups appeared in low-field at 2.57–2.80 ppm (0.44–1.12 ppm for complexes **2**–**4**). The signals of the ethyl and propyl group of the amidine substituent in **5**–**7** also underwent a number of changes and in general were shifted to the low field. For example, the signal of the methylene group of the propyl fragment -NHC*H*2CH2CH3 in complex **7** was observed at 3.41 ppm, whereas in complex **4** it appeared at 2.75 ppm. The signal of the methylene group of the ethyl substituent was located at 2.54 ppm for **7** in contrast with 2.22 ppm for **4**. At the same time, the signals of the carbon atom of the methylene group of the propyl fragment -NH*C*H2CH2CH3 in 13C NMR spectra of **<sup>5</sup>**–**<sup>7</sup>** underwent the high field shift from ~55 ppm for **<sup>2</sup>**–**<sup>4</sup>** to ~46 ppm, whereas the signal of the methylene group of the ethyl fragment demonstrated a slight high field shift from ~23–24 ppm for **2**–**4** to ~26–30 ppm for **5**–**7**. In the 11B NMR spectra of complexes **5**–**7**, the singlet from the substituted boron atom was observed at 1.9 ppm for **5** and at 3.5 ppm for **6** and **7**. Other signals appeared as groups of four (complex **5**) or five (complexes **6** and **7**) doublets in the region from −10.2 to −26.3 ppm with the total integral ratios 1:2:1:4:1 for **5** and 1:2:1:2:2:1 for **6** and **7**. The chemical shifts of phosphine ligands in the 31P NMR spectra of **5**–**7** were close to those for **2**–**4**. In the IR spectra of **5**–**7**, the NH and BH stretching bands were observed in the region of 3402–3223 cm−<sup>1</sup> and ~2555 cm<sup>−</sup>1, respectively, whereas the bands corresponding to the N=C bond appeared at 1632, 1626 and 1630 cm−<sup>1</sup> for **5**, **6** and **7**, respectively. The mass spectra of complexes **5**–**7** performed using the MS MALDI technique contained two main sets of signals corresponding to the molecular picks of complexes **5**–**7** themselves and complexed with the loss of the chloride ligand. For example, the MALDI mass spectrum of complex **6** contained a typical carborane envelope centered at *m*/*z* 477.253 corresponding to the molecular ion pick and another one centered at *m*/*z* 442.275, that corresponded to the loss of the chloride ligand by complex **6**.

#### **3. Conclusions**

In this work the utility of using the *nido*-carboranyl amidine 10-PrNHC(Et)=HN-7,8-C2B9H11 in the complexation reactions with different nickel(II) phosphine complexes was demonstrated. As a result, a series of novel half-sandwich nickel(II) π,σ-complexes [3-R2R'P-3-(8-Pr*N*=C(Et)NH)-*closo*-3,1,2-NiC2B9H10] (R = R' = Ph, Bu; R = Me, R' = Ph) was prepared. The crystal molecular structure of [3-Ph3P-3- (8-Pr*N*=C(Et)NH)-*closo*-3,1,2- NiC2B9H10] was determined by single crystal X-ray diffraction. The acidification of obtained complexes with HCl led to the breaking of the Ni-N bond with the formation of the corresponding nickel(II) π-complexes [3-Cl-3-R'R2P- 8-PrNH=C(Et)NH-*closo*-3,1,2- NiC2B9H10] (R = R' = Ph, Bu; R = Me, R' = Ph). The process was accompanied by a change in the color of complexes from dark red to amaranth one. In this regard, the obtained complexes can be considered as potential acid-base indicators.

#### **4. Experimental Section**

#### *4.1. Reagents and Methods*

The amidine 10-PrNHC(Et)=HN-7,8-C2B9H11 (**1**) was prepared according to procedure from the literature [20]. Dichlorobis(triphenylphosphine)nickel(II), dichlorobis (dimethylphenylphosphine)nickel(II) and dichlorobis(tributylphosphine)nickel(II) were synthesized according to the previously described methods [28]. Tetrahydrofuran was dried using standard procedure [29]. All manipulations were carried out in air. The reaction progress was monitored by thin-layer chromatography (Merck F254 silica gel on aluminum plates) and visualized using 0.5% PdCl2 in 1% HCl in aq. MeOH (1:10). Acros Organics silica gel (0.060–0.200 mm) was used for column chromatography. The NMR spectra at 400.1 MHz (1H), 128.4 MHz (11B), 100.0 MHz (13C) and 162 MHz (31P) were recorded with a Varian Inova 400 spectrometer. The residual signal of the NMR solvent relative to tetramethylsilane was taken as an internal reference for 1H and 13C NMR spectra. 11B NMR spectra were referenced using BF3 . Et2O as an external standard. 31P NMR spectra were cited relative to 85% H3PO4 as an external standard. Infrared spectra were recorded on an IR Prestige-21 (SHIMADZU, Kyoto, Japan) instrument. UV/Vis spectra in chloroform were recorded with a SF-2000 spectrophotometer (OKB SPECTR LLC, Saint-Petersburg, Russia) using 1 cm cuvettes. MALDI mass spectra (positive ion mode) were acquired using a Bruker AutoFlex II reflector time-of-flight device equipped with an N2 laser (337 nm, 2.5 ns pulse). *Trans*-2-[3-(4-*tert*-butylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB, ≥98%, Sigma-Aldrich, Louis, MO, USA) was chosen as a matrix, matrix-to-analyte molar ratio in spotted probes being more than 1000/1. High resolution mass spectra (HRMS) were measured on a Bruker micrOTOF II instrument using electrospray ionization (ESI). The measurements were done in a positive ion mode with mass range from *m*/*z* 50 to *m*/*z* 3000.

#### *4.2. Synthesis of [3-Ph3P-3-(8-PrN=C(Et)NH)-closo-3,1,2-NiC2B9H10] (2)*

The potassium *tert*-butoxide (0.34 g, 3.00 mmol) was added to a solution of **1** (0.15 g, 0.60 mmol) in dry tetrahydrofuran (15 mL). The mixture was stirred for ~10 min at room temperature and [Ni(PPh3)2Cl2] (0.47 g, 0.72 mmol) was added by one portion. The paleyellow color of the reaction mixture was immediately turned to dark red. The reaction mixture was stirred at room temperature in air for about 30 min and the solvent was evaporated under reduced pressure. The residue was treated with CH2Cl2 (20 mL) and water (20 mL). The insoluble particles were filtered off and the organic layer was separated, washed with water (2 × 20 mL) and evaporated under reduced pressure. The column chromatography on silica gel was used for the purification of the substance with hexane:CH2Cl2 (2:1) as an eluent to give maroon solid of **2** (0.28 g, 83% yield). The crystals suitable for X-ray analysis were obtained by slow evaporation from chloroform/hexane (3:1) solution.

1H NMR (CDCl3, ppm): *δ* 7.82 (6H, Ph), 7.44 (9H, Ph), 5.09 (1H, N*H*), 2.27 (2H, NHC*H*2CH2CH3), 2.14 (2H, C*H*2CH3), 1.25 (2H, NHCH2C*H*2CH3), 1.09 (3H, CH2C*H*3), 0.44 (2H, s, C*H*carb), 0.04 (3H, NHCH2CH2C*H*3), 3.4–0.2 (8H, br s, B*H*). 13C NMR (CDCl3, ppm): *δ* 176.9 (NH=*C*), 134.0 (*o*-Ph, d, *J* = 12 Hz), 131.2 (Ph, d, *J* = 37 Hz), 130.8 (*p*-Ph), 128.9 (*m*-Ph, d, *J* = 9 Hz), 56.1 (NH*C*H2CH2CH3), 25.1 (NHCH2*C*H2CH3), 24.0 (*C*H2), 17.2 (*C*Hcarb), 11.9 (*C*H3), 10.4 (NHCH2CH2*C*H3). 31P NMR (CDCl3, ppm): 29.3 (s, *P*Ph3). 11B NMR (CDCl3, ppm): *δ* 4.0 (1B, s), −11.2 (2B, d), −13.6 (3B, d), −23.5 (2B, d), −26.7 (1B, d). IR (film, cm−1): 3437 (νN-H), 3411 (νN-H), 3308, 3245, 3057, 2966 (νC-H), 2933 (νC-H), 2875 (νC-H), 2551 (br, νB-H), 1635 (νN=C), 1557, 1503, 1480, 1436, 1380, 1325, 1310. UV/VIS (Ɯ, nm): 248, 330, 510. MALDI MS: *m*/*z* for C26H38B9N2NiP: calcd 565.299 [M]+, obsd 565.288 [M]+ (100).

#### *4.3. Synthesis of [3-PhMe2P-3-(8-PrN=C(Et)NH)-closo-3,1,2-NiC2B9H10] (3)*

The procedure was analogous to the preparation of **2** using **1** (0.13 g, 0.52 mmol), potassium *tert*-butoxide (0.30 g, 2.60 mmol) and [Ni(PMe2Ph)2Cl2] (0.25 g, 0.62 mmol) in dry tetrahydrofuran (15 mL). The column chromatography on silica gel was used for the purification of the substance with CH2Cl2 as an eluent to give maroon solid of **3** (0.19 g, 82% yield).

1H NMR (acetone-d6, ppm): *δ* 8.19 (2H, Ph), 7.54 (3H, Ph), 6.25 (1H, N*H*), 2.48 (2H, NHC*H*2CH2CH3), 2.26 (2H, C*H*2CH3), 1.65 (6H, P(C*H*3)2), 1.57 (2H, NHCH2C*H*2CH3), 1.12 (2H, s, C*H*carb), 0.98 (3H, CH2C*H*3), 0.53 (3H, NHCH2CH2C*H*3), 3.4–0.4 (8H, br s, B*H*). 13C NMR (acetone-d6, ppm): *δ* 177.7 (NH=*C*), 131.3 (Ph, d, *J* = 18 Hz), 131.2 (*o*-Ph, d, *J* = 12 Hz), 130.3 (*p*-Ph), 128.8 (*m*-Ph, d, *J* = 10 Hz), 54.3 (NH*C*H2CH2CH3), 26.5 (NHCH2*C*H2CH3), 23.4 (*C*H2), 14.7 (*C*Hcarb), 13.7 (P(*C*H3)2, d, *<sup>J</sup>* = 24 Hz), 11.7 (*C*H3), 10.3 (NHCH2CH2*C*H3). 31P NMR (acetone-d6, ppm): <sup>−</sup>2.6 (s, *<sup>P</sup>*Me2Ph). 11B NMR (acetone-d6, ppm): *<sup>δ</sup>* 4.2 (1B, s), −12.0 (2B, d, *J* = 137 Hz), −14.0 (2B, d, *J* = 156 Hz), −15.7 (1B, d, *J* = 147 Hz), −23.8 (2B, d, *<sup>J</sup>* = 143 Hz), −27.4 (1B, d, *<sup>J</sup>* = 144 Hz). IR (film, cm−1): 3410 (νN-H), 3244 (νN-H), 2962

(νC-H), 2930 (νC-H), 2874 (νC-H), 2525 (br, νB-H), 1636 (νN=C), 1568, 1493, 1437, 1377, 1300, 1292. ESI HRMS: *m*/*z* for C16H34B9N2NiP: calcd 441.2696 [M]+, obsd 441.2696 [M]+ (100).

#### *4.4. Synthesis of [3-Bu3P-3-(8-PrN=C(Et)NH)-closo-3,1,2-NiC2B9H10] (4)*

The procedure was analogous to the preparation of **2** using **1** (0.16 g, 0.64 mmol), potassium *tert*-butoxide (0.36 g, 3.20 mmol) and [Ni(PBu3)2Cl2] (0.26 g, 0.77 mmol) in dry tetrahydrofuran (15 mL). The column chromatography on silica gel was used for the purification of the substance with CH2Cl2 as an eluent to give maroon solid of **3** (0.25 g, 80% yield).

1H NMR (CDCl3, ppm): *δ* 5.04 (1H, N*H*), 2.75 (2H, NHC*H*2CH2CH3), 2.22 (2H, C*H*2CH3), 1.70 (2H, NHCH2C*H*2CH3), 1.61 (6H, P(C*H*2CH2CH2CH3)3), 1.50 (6H, P(CH2CH2C*H*2CH3)3), 1.40 (6H, P(CH2C*H*2CH2CH3)3), 1.06 (3H, CH2C*H*3), 0.93 (14H, NHCH2CH2C*H*<sup>3</sup> + P(CH2CH2CH2C*H*3)3 + C*H*carb), 3.4–0.2 (8H, br s, B*H*). 13C NMR (CDCl3, ppm): *δ* 176.6 (NH=*C*), 55.2 (NH*C*H2CH2CH3), 26.7 (NHCH2*C*H2CH3), 26.5 (P(*C*H2CH2CH2CH3)3, d, *J* = 50 Hz), 24.4 (P(CH2CH2*C*H2CH3)3, d, *J* = 12 Hz), 24.2 (*C*H2), 23.5 (P(CH2*C*H2CH2CH3)3, d, *J* = 20 Hz), 14.0 (*C*Hcarb), 13.7 (P(CH2CH2CH2*C*H3)3), 11.8 (*C*H3), 11.5 (NHCH2CH2*C*H3). 31P NMR (CDCl3, ppm): 11.9 (s, *P*Bu3). 11B NMR (CDCl3, ppm): *δ* 3.8 (1B, s), −11.8 (2B, d), −14.3 (2B, d), −15.1 (1B, d), −23.6 (2B, d), −26.8 (1B, d). IR (film, cm−1): 3447 (νN-H), 3408 (νN-H), 2957 (νC-H), 2932 (νC-H), 2872 (νC-H), 2544 (br, νB-H), 1622 (νN=C), 1557, 1497, 1464, 1379, 1283. MALDI MS: *m*/*z* for C20H50B9N2NiP: calcd 506.402 [M+H]+, obsd 506.400 [M+H]+ (100).

#### *4.5. General Procedure for the Synthesis of [3-Cl-3-R'R2P-8-PrN=C(Et)NH-closo-3,1,2-NiC2B9H10] (5–7)*

To the *N*-coordinated complexes **2**–**4** (0.40 mmol) dissolved in MeCN (10 mL), one drop (~0.1 mL) of concentrated HCl was added at room temperature. The dark red color of solution was immediately changed to amaranth. The solution was stirred for 5 min and evaporated under reduced pressure to give amaranth solid of **5**–**7.** In the case of complexes **6** and **7**, the column chromatography on silica gel was used for the purification with CH2Cl2 as an eluent.

Spectral data for [3-Cl-3-Ph3P-8-Pr*N*=C(Et)NH-*closo*-3,1,2-NiC2B9H10] (**5**)

Yield 0.18 g (75%).

1H NMR (CDCl3, ppm): *δ* 9.09 (1H, N*H*Pr), 7.62 (6H, Ph), 7.31 (9H, Ph), 5.53 (1H, N*H*), 3.22 (2H, NHC*H*2CH2CH3), 2.57 (2H, C*H*carb), 1.86 (4H, NHCH2C*H*2CH3 + C*H*2CH3), 0.93 (6H, NHCH2CH2C*H*<sup>3</sup> + CH2C*H*3), 3.8–0.3 (8H, br s, B*H*). 13C NMR (CDCl3, ppm): *δ* 167.5 (NH=*C*), 134.3 (Ph), 133.7 (Ph), 130.4 (Ph), 128.2 (Ph), 45.6 (NH*C*H2CH2CH3), 30.8 (*C*H2), 25.2 (NHCH2*C*H2CH3), 23.2 (*C*Hcarb), 11.1 (*C*H3), 9.8 (NHCH2CH2*C*H3). 31P NMR (CDCl3, ppm): 28.2 (s, *<sup>P</sup>*Ph3). 11B NMR (CDCl3, ppm): *<sup>δ</sup>* 1.9 (1B, s), −10.2 (2B, d), −12.3 (1B, d), −20.8 (4B, d), −24.2 (1B, d). IR (film, cm−1): 3397 (νN-H), 3302 (νN-H), 3233 (νN-H), 3059, 2966 (νC-H), 2967 (νC-H), 2928 (νC-H), 2878 (νC-H), 2560 (br, νB-H), 1632 (νN=C), 1553, 1501, 1481, 1437, 1385, 1259. UV/VIS (Ɯ, nm): 280, 302, 486. MALDI MS: *m*/*z* for C26H39B9N2ClNiP: calcd 601.277 [M]+, obsd 601.293 [M]+ (10), *m*/*z* for C26H39B9N2NiP: calcd 566.308 [M-Cl]+, obsd 566.321 [M-Cl]+ (90).

Spectral data for [3-Cl-3-PhMe2P-8-Pr*N*=C(Et)NH-*closo*-3,1,2-NiC2B9H10] (**6**) Yield 0.16 g (84%).

1H NMR (acetone-d6, ppm): *δ* 10.37 (1H, N*H*Pr), 7.77 (2H, Ph), 7.42 (3H, Ph), 6.15 (1H, N*H*), 3.35 (2H, NHC*H*2CH2CH3), 2.69 (2H, s, C*H*carb), 2.48 (2H, C*H*2CH3), 1.89 (2H, NHCH2C*H*2CH3), 1.53 (6H, P(C*H*3)2), 1.23 (3H, CH2C*H*3), 1.08 (3H, NHCH2CH2C*H*3), 3.0–0.5 (8H, br s, B*H*). 13C NMR (acetone-d6, ppm): *δ* 130.2 (Ph), 130.0 (Ph), 129.0 (Ph), 46.1 (NH*C*H2CH2CH3), 26.2 (*C*H2), 23.4 (NHCH2*C*H2CH3), 14.8 (P(*C*H3)2), 11.7 (NHCH2CH2*C*H3), 9.7 (*C*H3). 31P NMR (acetone-d6, ppm): −2.53 (s, *<sup>P</sup>*Me2Ph). 11B NMR (acetone-d6, ppm): *<sup>δ</sup>* 3.5 (1B, s), −10.5 (2B, d, *J* = 133 Hz), −13.4 (1B, d, *J* = 145 Hz), −19.2 (2B, d, *J* = 140 Hz), −22.9 (2B, d, *<sup>J</sup>* = 117 Hz), −26.3 (1B, d, *<sup>J</sup>* = 172 Hz). IR (film, cm−1): 3402 (νN-H), 3259 (νN-H), 3227 (νN-H), 3053, 2968 (νC-H), 2934 (νC-H), 2878 (νC-H), 2548 (br, νB-H), 1626 (νN=C), 1557, 1435, 1421, 1384, 1361, 1296. MALDI MS: *m/z* for C16H35B9N2ClNiP: calcd 477.245

[M]+, obsd 477.253 [M]+ (23), *m*/*z* for C16H35B9N2NiP: calcd 442.276 [M-Cl]+, obsd 442.275 [M-Cl]+ (77).

Spectral data for [3-Cl-3-Bu3P-8-Pr*N*=C(Et)NH-*closo*-3,1,2-NiC2B9H10] (**7**) Yield 0.19 g (88%).

1H NMR (CDCl3, ppm): *δ* 9.90 (1H, N*H*Pr), 6.22 (1H, N*H*), 3.41 (2H, q, *J* = 7.5 Hz, NHC*H*2CH2CH3), 2.80 (2H, s, C*H*carb), 2.54 (2H, q, *J* = 7.6 Hz, C*H*2CH3), 1.90 (2H, m, *J* = 7.5 Hz, NHCH2C*H*2CH3), 1.58 (12H, P(C*H*2C*H*2CH2CH3)3), 1.39 (6H, P(CH2CH2C*H*2CH3)3), 1.28 (3H, t, *J* = 7.6 Hz, CH2C*H*3), 1.07 (3H, t, *J* = 7.5 Hz, NHCH2CH2C*H*3), 0.93 (9H, P(CH2CH2CH2C*H*3)3), 3.3–0.9 (8H, br s, B*H*). 13C NMR (CDCl3, ppm): *δ* 46.2 (NH*C*H2CH2CH3), 30.8 (*C*Hcarb), 26.3 (*C*H2), 26.2 (P(CH2*C*H2CH2CH3)3), 24.5 (P(CH2CH2*C*H2CH3)3), 23.7 (P(*C*H2CH2CH2CH3)3, 23.6 (NHCH2*C*H2CH3), 11.6 (NHCH2CH2*C*H3), 10.3 (*C*H3), 13.8 (P(CH2CH2CH2*C*H3)3). 31P NMR (CDCl3, ppm): 11.7 (s, *P*Bu3). 11B NMR (CDCl3, ppm): *δ* 3.5 (1B, s), −10.6 (2B, d, *J* = 127 Hz), −13.9 (1B, d, *J* = 129 Hz), −20.3 (2B, d, *J* = 135 Hz), −23.7 (2B, d), −26.3 (1B, d). IR (film, cm−1): 3223 (νN-H), 2957 (νC-H), 2930 (νC-H), 2872 (νC-H), 2550 (br, νB-H), 1630 (νN=C), 1552, 1462, 1415, 1379, 1342, 1300. UV/VIS (Ɯ, nm): 246, 324, 517. MALDI MS: *m*/*z* for C20H51B9N2ClNiP: calcd 541.371 [M]+, obsd 541.373 [M]+ (9) *m*/*z* for C20H51B9N2NiP: calcd 506.402 [M-Cl]+, obsd 506.443 [M-Cl]+ (91).

#### *4.6. Single Crystal X-ray Diffraction Study*

X-ray experiment for compound **2** was carried out using a SMART APEX2 CCD diffractometer (λ(Mo-Kα) = 0.71073 Å, graphite monochromator, *ω*-scans) at 120 K. Collected data were processed by the SAINT and SADABS programs incorporated into the APEX2 program package [30]. The structure was solved by direct methods and refined by the full-matrix least-squares procedure against *F*<sup>2</sup> in anisotropic approximation. The refinement was carried out with the SHELXTL program [31]. The CCDC number 2065468 contains the supplementary crystallographic data (Supplementary Materials) for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data\_request/cif (accessed on 19 March 2021).

Crystallographic data for *2:* C26H38B9NiN2P·CHCl3 are triclinic, space group *P*-1: *a* = 10.0074(6) Å, *b* = 11.0974(6) Å, *c* = 16.8683(8) Å, α = 76.8900(10)◦, *β* = 81.3380(10)◦, γ = 67.4720(10)◦, *<sup>V</sup>* = 1681.05(16) Å3, *<sup>Z</sup>* = 2, *<sup>M</sup>* = 684.92, *<sup>d</sup>*cryst = 1.353 g·cm−3. *wR*2= 0.1312 calculated on *F*<sup>2</sup> *hkl* for all 8953 independent reflections with 2*θ* < 58.3◦, (*GOF* = 1.048, *R* = 0.0528 calculated on *Fhkl* for 6268 reflections with *I* > 2*σ*(*I*)).

#### *4.7. Quantum Chemical Calculation*

Optimization of the geometry of compound **2** was carried out using the Gaussian program [32] at PBE0/def2tzvp level of approximation, which was adopted in our earlier calculations [33–36]. The AIM theory [37,38] was utilized to search for bond critical points of molecular electron density. Correlation of interatomic energy and potential energy density at bond critical point (*E*=1/2*V*(*r*)) [39,40] was adopted for estimation of the energy of noncovalent intramolecular interactions taking into account its reliability for energetic analysis [41–43].

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2073-435 2/11/3/306/s1. Figure S1–S29 NMR spectra of compounds **2**–**7**.

**Author Contributions:** Synthesis and writing, NMR, IR research analysis, M.Y.S.; synthesis, S.A.E.; X-ray diffraction study, K.Y.S.; supervision, writing, I.B.S.; editing V.I.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the Russian Science Foundation (№ 19-73-00229).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available in Supplementary Materials.

**Acknowledgments:** The NMR spectroscopy and X-ray diffraction experiments were performed using equipment of the Center for Molecular Structure Studies at A.N. Nesmeyanov Institute of Organoelement Compounds, operating with financial support of the Ministry of Science and Higher Education of the Russian Federation. The authors thank Vitaliy Yu. Markov (M.V. Lomonosov Moscow State University) for collecting MALDI mass spectra.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Tuning the Liquid Crystallinity of Cholesteryl-***o***-Carborane Dyads: Synthesis, Structure, Photoluminescence, and Mesomorphic Properties**

**Albert Ferrer-Ugalde 1,†, Arántzazu González-Campo 1,†, José Giner Planas 1, Clara Viñas 1, Francesc Teixidor 1, Isabel M. Sáez 2,\* and Rosario Núñez 1,\*,‡**


**Abstract:** A set of mesomorphic materials in which the *o*-carborane cluster is covalently bonded to a cholesteryl benzoate moiety (mesogen group) through a suitably designed linker is described. The olefin cross-metathesis between appropriately functionalized styrenyl-*o*-carborane derivatives and a terminal alkenyl cholesteryl benzoate mesogen (all type I terminal olefins) leads to the desired *trans*-regioisomer, which is the best-suited configuration to obtain mesomorphic properties in the final materials. The introduction of different substituents (R = H (**M2**), Me (**M3**), or Ph (**M4**)) to one of the carbon atoms of the *o*-carborane cluster (Ccluster) enables the tailoring of liquid crystalline properties. Compounds **M2** and **M3** show the chiral nematic (N\*) phase, whereas **M4** do not show liquid crystal behavior. Weaker intermolecular interactions in the solid **M3** with respect to those in **M2** may allow the liquid crystallinity in **M3** to be expressed as enantiotropic behavior, whereas breaking the stronger intermolecular interaction in the solid state of **M2** leads directly to the isotropic state, resulting in monotropic behavior. Remarkably, **M3** also displays the blue phase, which was observed neither in the chiral nematic precursor nor in the styrenyl-cholesterol model (**M5**) without an *o*-carborane cluster, which suggests that the presence of the cluster plays a role in stabilizing this highly twisted chiral phase. In the carborane-containing mesogens (**M2** and **M3**), the *o*-carborane cluster can be incorporated without destroying the helical organization of the mesophase.

**Keywords:** boron clusters; carboranes; liquid crystals; fluorescence; cholesterol

#### **1. Introduction**

The liquid crystalline state (mesophase) is a state of matter that displays properties between those of conventional liquid and solid phases; this behavior can be subdivided into two types: thermotropic and lyotropic. Among these, thermotropic liquid crystals (LCs) are partially ordered anisotropic fluids that exhibit one or more mesophases in a given temperature range [1–3]. In most cases, the mesogenic behavior of an LC material is due to the combination of a rigid core with flexible groups to produce rod-like molecules that may show different types of mesophases, such as nematic (N) or smectic (Sm) [1]. Both the electronic structure and geometry determine the thermal and optical properties of the mesogen. Cholesteric liquid crystals (CLCs) are well-known chiral nematic materials that display the chiral nematic phase (N\*), where the chiral molecules are organized in parallel planes to the director and twisted in a perpendicular way throughout the director, describing a helical structure. The most important feature of CLCs is the selective reflection of circularly polarized light according to Bragg's law [4]. Due to their 1D photonic structure

**Citation:** Ferrer-Ugalde, A.; González-Campo, A.; Planas, J.G.; Viñas, C.; Teixidor, F.; Sáez, I.M.; Núñez, R. Tuning the Liquid Crystallinity of Cholesteryl-*o*-Carborane Dyads: Synthesis, Structure, Photoluminescence, and Mesomorphic Properties. *Crystals* **2021**, *11*, 133. https://doi.org/10.3390/ cryst11020133

Academic Editor: Marina Yu. Stogniy Received: 23 December 2020 Accepted: 24 January 2021 Published: 28 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

and ease of fabrication, cholesteric liquid crystals have also attracted much attention as optical sensors [5–7].

Icosahedral 1,2-dicarba-*closo*-dodecacarborane or *o*-carborane (1,2-C2B10H12) [8–11] have highly polarizable spherical aromaticity through σ-delocalized electron densities [12–16]. Consequently, they display characteristic electronic properties [17,18], and thermal [19–21], chemical, and photochemical stability [22,23]. All these features make them attractive and interesting systems in materials science [24–26], especially for luminescent materials [27–48], polymers [21], and Non-Linear Optical (NLO) materials, [49,50] among others [22]. The rigidity of these boron clusters promotes their use as a structural element of the rigid core in the preparation of liquid crystalline materials [51]. Boron clusters of major interest for synthesizing liquid crystals have been the 12 vertex, *p*-carborane (1,12-C2B10H12) [52,53], [1-CB11H12] − [54–56], the dianion dodecaborate [B12H12] <sup>2</sup><sup>−</sup> [57], as well as the 10 vertex *closo*-boron clusters 1,10-C2B8H10, [1-CB9H10] − [54,58]. All of them are properly functionalized with suitable organic functional groups to provide a supramolecular structure with optical properties that might be modulated by external stimuli (Chart 1) [8]. Around 200 "rod-like" mesogens containing *p*-carboranes and several examples of bent-core mesogens bearing *m*-carborane (1,7-C2B10H12) have been reported, in which the angle formed for both substituents at the carbons of the cluster (Ccluster, Cc) is 120◦ [51]. Generally, *p*-carboranes are the preferred choice to synthesize LCs, not only because of their spherocylindrical morphology, but also for the accessibility of substitution at the antipodal Cc atoms with appropriate groups, leading to an extension of the molecular shape that forms the mesogenic state (Chart 1). Most LCs containing *p*-carboranes show nematic phases [59], except those LCs with terminal alkyl chains with fluoride atoms that usually lead to smectic phases [60]. Nevertheless, the use of *o*-carborane, 1,2-C2B10H12, and its derivatives has barely been explored as LCs, but studies involving their use in boron-containing liposomes as potential agents for BNCT have been reported [61–64]. To our knowledge, only two examples on *o*-carborane-based LCs have been recently reported by Kaszynsky [65,66]. In these, *o*-carborane is used as a linear structural element with a moderate dipole moment, which is substituted at the C1 and B12 atoms to give 1,12-difunctionalized derivatives that show the smectic A (SmA) and N mesophases.

**Chart 1.** Representation of more usual twelve vertex boron clusters used to perform liquid crystalline materials.

There have been several reports of liquid crystals containing octasilsesquioxanes [67,68], or fullerenes [69–71], as scaffold to append mesogenic moieties in order to probe the effect of a rigid structurally well-defined core on mesophase behavior. In this sense, *o*carborane may also provide a unique inorganic polyhedral scaffold that can lead the way to the generation of a new family of this class of materials, and it may offer a new insight into the effects of these substituents on the mesophase behavior of the ensuing liquid crystalline materials.

Herein we describe for the first time the synthesis, structural characterization, and photophysical properties in solution of a set of new mesomorphic materials in which the *o*-carborane cluster is covalently bonded to a cholesteryl benzoate moiety (mesogen group) through a suitable methylene linker. We have assessed the influence of the *o*-carborane unit and its substituents bound to the Cc atom on their liquid crystal properties. We have also tried to establish a relationship between the 3D crystal packing in solid state and the changes of mesophase with temperature.

#### **2. Results and Discussion**

#### *2.1. Synthesis and Characterization of Mesogens*

The chemical synthesis of linear cholesteryl-*o*-carborane dyads was designed to be carried out by olefin cross-metathesis between three different styrenyl-*o*-carborane derivatives: 1-[CH2C6H4-4- -(CH=CH2)]-2-R-1,2-*closo*-C2B10H10 [R = H (**1**), R = Methyl or Me (**2**), R = Phenyl or Ph (**3**)] (see Scheme 1) [42], and a cholesterol derivative that contains a suitable alkenylene moiety. To this purpose, cholesteryl benzoate was adequately modified to introduce the terminal alkenylene group producing **M1** by adaptation of the literature method (Scheme S1) [72]. A related strategy using olefin cross-metathesis in the design of liquid crystal synthesis has been reported while the current work was in progress [73,74]. Cross-metathesis reactions of *o*-carborane derivatives **1–3** and the mesogenic cholesterol derivative **M1** were carried out using first-generation Grubbs catalyst in CH2Cl2 at reflux for 48 h under argon atmosphere (Scheme 1). The reactions were by 1H NMR spectroscopy to the total disappearance of the starting vinyl and alkenyl protons. The white solid in the remaining suspensions, which corresponds to the homo-metathesis of the **M1** side product, was filtered off, and the clear brown solutions were quenched with methanol to yield a gray precipitate that was also filtered and dried. Spectroscopic analyses proved that the latter contained a mixture of the expected final compound (**M2–M4**) and a small quantity of the **M1** homo-metathesis compound, which is partially soluble in CH2Cl2. Although we did not use excess of **1–3** to avoid the formation of compounds from their homo-metathesis, some of them were obtained and remained in the MeOH solution. The gray precipitates were finally purified by column chromatography affording the *o*-carborane-containing dyads **M2**, **M3**, and **M4** as white solids in 54, 50, and 49% yield, respectively.

**Scheme 1.** Synthesis of cholesteryl-*o*-carborane dyads **M2–M4**.

In order to understand the influence of the *o*-carborane cluster and its substituent bound to the Cc atom on the mesogenic behavior of **M2–M4**, a carborane-free compound **M5** was also synthesized by cross-metathesis of pure styrene with the mesogen **M1** (Scheme 2). We used the same conditions as those for **M2–M4** with a large excess of styrene due to the ease of the styrene to give the compound from homo-metathesis. **M5** was isolated as a white solid in 82% yield after purification by column chromatography.

**Scheme 2.** Synthesis of carborane-free cholesterol derivative **M5**.

Complete cross-metathesis between **M1** with styrenyl-*o*-carborane derivatives **1–3** or styrene afforded **M2–M5**, which was confirmed by standard spectroscopy techniques such as FT-IR, 11B, 1H, and 13C NMR, MALDI-TOF mass spectroscopy and elemental analysis. A detailed description of the characterization is included in the Supporting Information. In addition, suitable single crystals of **M2**, **M3**, and **M5** were analyzed by X-ray diffraction.

#### *2.2. X-Ray Crystal Structures of* **M2***,* **M3***, and* **M5**

Single crystals of **M2**, **M3**, and **M5** suitable for X-ray diffraction analysis were obtained by slow evaporation of CH2Cl2 (**M2–3**) or toluene (**M5**) solutions and found suitable for X-ray diffraction analysis. The molecular structures of **M2**, **M3**, and **M5** are shown in Figures 1 and 2. The main cell crystal parameters can be found in Table S1. A list of selected data and bonding parameters can also be found in the Supporting Information (Tables S2–S4).

**Figure 1.** View of independent molecules in the unit cell of **M2** (**A**) and **M3** (**B**). Intermolecular contacts are shown as dotted blue lines. Color codes: H white; C gray; O red; B orange.

Whereas both *o*-carborane-containing dyads **M2** and **M3** crystallized in the monoclinic and non-centrosymmetric *P*21 space group, the styrene derivative **M5** crystallized in the triclinic *P1* space group. As shown in Figures 1 and 2 and Figures S16 and S17, unit cells for the compounds showed one, two, or four molecules of **M2**, **M3**, and **M5**, respectively. The shape of the molecules was similar, although slightly different geometries can be observed in **M3** (Figure 1) and **M5** (Figure 2). In particular, the angles between the plane of the benzoate and the steroid ring in **M3** (i.e., for the C2–C7 plane and C24–C28 mean plane) were 69.50(14)◦ and 55.15(16)◦, respectively, and between the benzoate ring and the C6H4 ring close to the cluster (i.e., for the C2–C7 plane and C13–C18 plane), they were 25.77(14)◦ and 136.64(15), respectively.

**Figure 2.** View of independent molecules in the unit cell of **M5**. Intermolecular contacts are shown as dotted blue lines. Color codes: H white; C gray; O red; B orange.

In general, the packing motifs in the crystals were similar for all structures. Indeed, the **M2**, **M3**, and **M5** structures were built from tilted molecules to form layers that were piled giving the observed 3D structures (Figure 3). In all cases, layers were formed by self-assembled molecules oriented parallel to each other (Figure S18). The densities of the 3D structures decreased in the order **M5** (1.152 g cm−3) > **M3** (1.135 g cm−3) > **M2** (1.122 g cm<sup>−</sup>3).

**Figure 3.** Views of crystal packings of **M2** (**A**), **M3** (**B**), and **M5** (**C**). Independent molecules in the unit cells are colored. H atoms are omitted for clarity.

As it is the case in a large majority of mesogens [73–75], and also in the case of most *o*-carborane derivatives [76], the molecules presented here have no strong hydrogen bond donor groups. Therefore, in the obtained crystals, molecules in the layers as well as interactions within the layers are stabilized by weak intermolecular interactions. This is clearly confirmed by the Hirshfeld surface analysis [77] for **M2**, **M3**, and **M5**. The analysis clearly showed the presence of weak H··· H and H··· O interactions, with no long sharp spikes characteristic of strong hydrogen bonds and the absence of strong π–π interactions in the Fingerprint plots (Figure S19) [78]. The relative contributions of different interactions to the Hirshfeld surface were calculated from the Fingerprint plots (Figure 4). From this simple analysis, it can be clearly observed that H/H contacts correspond to 78–84% of the total Hirshfeld surface area for these molecules, which is consistent with the large ratio of external H to N, O, or C atoms. C/H contacts comprise around 13–15% to the total Hirshfeld surface area and C/C (i.e., π–π) interactions contribute from 0.8% in the noncarborane derivative **M5** to an almost negligible percentage in the carborane compounds **M2** and **M3** (Figure 4).

**Figure 4.** Relative contributions of various intermolecular contacts to the Hirshfeld surface area in all compounds in this study.

Analysis of the data in Figure 4 clearly indicated that the introduction of the *o*carborane moieties into the mesogen **M1** increases the percentage of H···H contacts, while it decreases the percentage of H/C and C/C contacts. The decrease of the latter two is consistent with the spherical shape of *o*-carborane disrupting the supramolecular contacts of merely flat and aromatic systems (e.g., C–H···π and π–π), and this may affect the mesogenic properties of **M2** and **M3** when compared to **M5** (vide infra). In addition to this, spatial requirements for the carborane cages in these molecules disrupt the efficient packing that, in the absence of these spherical cages, is observed in rod-like molecules such as **M5** (Figure 3). This is experimentally observed in the calculated densities of the crystals for **M2**, **M3**, and **M5** (vide supra).

#### *2.3. Photophysical Properties*

The photophysical properties of **M2–M5**, including absorption and emission maxima (*λ*abs and *λ*em), fluorescence quantum yields (*Φ*f), and Stokes shifts were assessed in CH2Cl2 (Figure S20 and Table 1). The UV-Vis spectra showed strong maximum absorption peaks at *λ*abs = 260 nm for **M2**–**M4** and *λ*abs = 257 nm for the non-carborane containing **M5**, and there were two additional very weak shoulders around 285 and 295 nm (Figure S20). The absorption peaks were attributed to the π–π\* transitions of the styrenyl group, which is the fluorophore of the molecule, and they appear in the same region as that of the starting compounds **1–3** [42]. Compounds **M2**, **M3**, and **M5** displayed emission maxima at *λ*em = 317 and 319 nm for **M2** and **M3** respectively, whereas **M5** exhibited a maximum

at 315 nm after excitation at 260 nm (Figure S20). On the contrary, compound **M4** did not show fluorescence emission in solution. These results indicated that the emission properties of **M2–M4** were essentially the same as for their precursors **1–3**. In previous works, we demonstrated that the photoluminescence properties of precursors **1–3** and their derivatives were influenced by the substituents bound to the Cc atom [42], where it was proved that the Ph-*o*-carborane moiety acts as an electron-withdrawing group, leading to an efficient quenching of the fluorescence due to a charge transfer process [42,46]. This would explain the fluorescence quenching for **M4**. Fluorescence quantum yield values (*Φ*f) confirmed that **M2** exhibited the highest efficiency (24.8%), whereas **M3** and **M5** displayed similar values to each other around 12.2%. It is noteworthy that compounds **M2–M4** followed the same trend for their emission properties than their respective precursors **1–3**. Noteworthy, the Stokes shift values did not vary significantly within the three luminescent compounds, being all of them close to the region of 7000 cm<sup>−</sup>1.

**Table 1.** Photophysical properties of **M1–M5**.


All spectra were recorded in CH2Cl2. All samples were prepared to obtain an absorbance of around 0.1 at the excitation wavelength. Quantum yields *Φ*<sup>f</sup> were calculated using quinine sulfate in a 0.5 M aqueous solution of H2SO4 (*Φ*<sup>f</sup> = 0.54) as a standard. Stokes shift = −10\*(1/*λ*em − 1/*λ*abs).

#### *2.4. Liquid Crystal Properties*

The mesomorphic properties of **M1–M5** were investigated by differential scanning calorimetry (DSC) and polarized optical microscopy (POM). The DSC thermograms of **M2– M5** are shown in Figures S21–S24, and the transition temperatures are listed in Table 2. Selected optical textures determined by POM are presented in Figure 5. The cross-metathesis reaction between the cholesteryl benzoate and the styrenyl-*o*-carborane derivatives produced only the *trans*-regioisomer, which is the preferred configuration to induce liquid crystallinity. The *o*-carborane derivative **M3** (R = Me) is the one that has the most attractive mesomorphic behavior. On cooling from the isotropic state, POM showed the presence of the rare blue phase over a very narrow temperature range (≈0.1 ◦C), which was identified by the mosaic texture (Figure 5b) with very low birefringence [79]. On further cooling, it was followed by the N\* phase, which was identified by the Grandjean planar texture typical of the N\* phase (Figure 5a). The blue phase could not be observed in the DSC thermogram, but the thermogram showed clearly the enantiotropic nature of the N\* phase. Pleasingly, mesogens **M2** (R = H) and **M3** (R = Me) also displayed the N\* phase, indicating that the N\* phase of the cholesterol substituent was preserved upon attachment to the styrenyl-*o*-carborane cluster. The N\* mesophase of **M3** is enantiotropic in nature (Figure 5a), whereas for **M2**, it is a monotropic N\* phase. It is important to note that **M2** showed considerable supercooling of the transition from the isotropic state to the N\* phase, as observed by DSC and POM. The alkenyl cholesteryl benzoate precursor **M1** showed a wide temperature range N\*, from 131.4 ◦C (crystalline state) to 238 ◦C (isotropic liquid). The model compound **M5** showed the enantiotropic N\* phase over a range ≈ 70 ◦C, with a clearing point at 224 ◦C (Figure 5d), which is at a lower temperature than **M1** (238 ◦C), indicating the slightly lower thermal stability of the mesophase in comparison with **M1** (Table 2). Interestingly, neither **M1** nor **M5** showed blue phase, suggesting that the *o*-carborane cluster plays a role in stabilizing the intermolecular interactions needed to support the blue phase. Comparison of the transition temperatures showed that the *o*-carborane-containing dyads **M2**–**M4** have lower isotropization temperatures than **M5** (Table 2). This comparison

suggested that the presence of the spherical *o*-carborane cluster disrupts significantly the intermolecular interactions of the cholesteryl benzoate cores needed to induce the mesophase. The higher isotropization temperature of **M2** in comparison with **M3** indicated stronger intermolecular interactions in **M2** (R = H) than in **M3** (R = Me). The introduction of an Me group at the *o*-carborane cluster adds steric hindrance and makes space filling more difficult, resulting in a lower clearing point. This different mesogenic behavior between **M2** and **M3** could be rationalized, taking into account the differences between the solid state at room temperature and the liquid crystalline state, if we consider that the number of H···H contacts in the 3D crystal packing of **M2** is higher than in **M3** confirmed by the Hirshfeld surface analysis (vide supra). In contrast, both mesophase range and stability of **M5**, without the *o*-carborane moiety, are higher than those for **M2** and **M3**. A similar volume effect has been observed when carbocyclic moieties are used as end-groups in certain mesogens [80]. This might be related to the disrupting effect of the *o*-carborane cluster on the N\* phase. On the other hand, *o*-carborane derivative **M4** (R = Ph) did not show liquid crystallinity, only a series of crystal forms, despite the presence of the strongly nematogen substituent. In this case, the presence of the phenyl group at the *o*-carborane cluster led to a largely bent geometry that disrupts any intermolecular interactions of the cholesterol rigid core needed to promote mesogenic behavior. This is supported by the melting point of **M4**, which was the lowest of the three cholesteryl-*o*-carborane derivatives.

**Table 2.** Phase transition temperatures (◦C) and corresponding enthalpy values (J mol<sup>−</sup>1, in square brackets) obtained from differential scanning calorimetry (DSC) second heating cycles for compounds **M2–M5**.


N\*: chiral nematic phase.

**Figure 5.** (**a**) Optical texture of **M3** in the N\* phase at 162.2 ◦C. (**b**) Polarized optical microscopy (POM) image of the blue phase of **M3**. (**c**) POM image of **M2** at 119.0 ◦C on cooling from the isotropic liquid, showing the N\* and crystal phases. (**d**) POM image of **M5** at 186.2 ◦C during the cooling cycle showing the N\* phase.

Remarkably, these results indicated that the mesomorphic properties of cholesteryl-*o*carborane dyads **M2–M4** could be modulated by modifying the substituent at the adjacent Cc atom. Non-substituted (**M2**) and Me-substituted **M3** displayed liquid crystals properties, whereas **M4** bearing a Ph group did not show liquid crystallinity. This suggested that the *o*-carborane is a suitable platform for liquid crystals and also may enable tuning of the liquid crystals properties of an attached mesogen group.

#### **3. Materials and Methods**

#### *3.1. Materials*

All reactions were performed under an atmosphere of argon employing standard Schlenk techniques. Anhydrous dichloromethane was purchased from Aldrich and used as received. 1-[CH2C6H4-4- -(CH=CH2)]-2-H-1,2-*closo*-C2B10H10 (**1**), 1-[CH2C6H4-4- -(CH=CH2)]- 2-CH3-1,2-*closo*-C2B10H10 (**2**), and 1-[CH2C6H4-4- -(CH=CH2)]-2-C6H5-1,2-*closo*-C2B10H10 (**3**) were obtained as described in the literature [42]. **M1** was prepared according to the literature process [72]. Reagent grade styrene and first generation Grubbs catalyst were purchased from Aldrich (St. Louis, MO, USA) and used as received. Solvents and all other reagents were purchased from Aldrich and used as received. Reactions were monitored by thin layer chromatography (TLC) using an appropriate solvent system. Silica-coated aluminium TLC plates used were purchased from Merck, Kenilworth, NJ, USA (Kieselgel 60 F-254) and visualized using either UV light (254 nm and 365 nm), or by oxidation with either iodine or aqueous potassium permanganate solution.

#### *3.2. Instrumentation*

Infrared spectra were measured on a Perkin-Elmer Spectrum 400 FT-IR/FT-FIR (Waltham, MA, USA) and Shimadzu IR Prestige-21 (Kyoto, Japan) with Specac Golden Gate diamond ATR IR insert (Orpington, UK). 1H NMR (300.13 MHz and 400 MHz) and 13C{1H} NMR spectra (75.47 MHz, 100.5 MHz) were obtained using Varian Unity Inova 400 MHz (Palo Alto, CA, USA), JEOL ECX 400 MHz (Akishima, Tokyo, Japan) and Bruker ARX 300 spectrometers (Billerica, MA, USA). The 11B{1H} NMR spectra (96.29 MHz) were run on a Bruker ARX 300 spectrometer. All experiments were made with concentrations between 15 and 20 mg/mL at 25 ◦C. Chemical shifts (ppm) are relative to Si(CH3)4 for 1H (of residual proton; δ 7.25 ppm) and 13C{1H} NMR (δ 77.23 ppm signal) in CDCl3. Chemical shift values for 11B{1H} NMR spectra were referenced to external BF3·OEt2. Chemical shifts are reported in units of parts per million downfield from the reference, and all coupling constants are reported in Hz. Mass spectra were recorded on a Bruker Solarix FT-ICR (MALDI TOF/TOF™ spectra). UV/Vis spectra were measured on a Perkin Elmer Lambda 900 and Shimadzu UV-1700 Pharmaspec spectrophotometers using 1.0 cm cuvettes. Fluorescence emission spectra were recorded in a Perkin-Elmer LS-45 (230 V) Fluorescence spectrometer. Samples were prepared in spectroscopic grade solvents and adjusted to a response within the linear range. No fluorescent contaminants were detected on excitation in the wavelength region of experimental interest. CHN Elemental analyses were performed using an Exeter Analytical Inc. CE440 analyzer (North Chelmsford, MA, USA).

#### *3.3. X-ray Diffraction*

Diffraction data were collected at 110 K on an Oxford Diffraction SuperNova diffractometer (Abingdon, Oxfordshire, UK) with Mo-K<sup>α</sup> radiation (λ = 0.71073 Å) using an EOS CCD camera (Canon, Tokyo, Japan). The crystal was cooled with an Oxford Instruments Cryojet. Diffractometer control, data collection, initial unit cell determination, frame integration, and unit-cell refinement was carried out with "Crysalis" [81]. Face-indexed absorption corrections were applied using spherical harmonics, implemented in a SCALE3 ABSPACK scaling algorithm [81]. OLEX2c was used for overall structure solution, refinement, and preparation of computer graphics and publication data. Within OLEX2, the algorithm used for structure solution was direct methods [82] for **M2** and **M3** and charge-flipping [83] for **M5**. Refinement by full-matrix least-squares used the SHELXL [82]

algorithm within OLEX2 [84]. All non-hydrogen atoms were refined anisotropically. Hydrogen atoms were placed using a "riding model" and included in the refinement at calculated positions.

In the case of compound **M5**, the differences among molecules in the unit cell differ in the orientation of the steroid system, the benzoate, and the terminal phenyl ring. For the four molecules, the angle between the plane of the benzoate and the steroid ring was (50.71(12)◦, 68.7(3)◦, 54.87(13)◦, and 40.88(12)◦, respectively; the angle between the benzoate ring and the phenyl ring was 112.52(12)◦, 93.8(3)◦, 67.33(12)◦, and 80.6(5)◦, respectively. Two of the molecules exhibited disorder. For one, the benzoate ring and the *para*-OCH2 was modeled in two positions with refined occupations of 0.684:0.316(4), whilst for the other, the phenyl was modeled in two positions with refined occupancies of 0.654:0.346(4). When calculating the planes for the angles described above, an average position was used.

#### *3.4. Phase Identification by Optical and Thermal Methods*

Polarized optical microscopy was performed on a Zeiss Axioskop 40Pol microscope (Oberkochen, Germany) using a Mettler FP82HT hotstage (Columbus, OH, USA) controlled by a Mettler FP90 central processor. Photomicrographs were captured via either an InfinityX-21 digital camera or a Sony NEX 5R mirrorless digital camera (Tokyo, Japan) mounted atop the microscope. Differential scanning calorimetry was performed on a Mettler DSC822e fitted with an autosampler operating with MettlerStar®software and calibrated before use against an indium standard (onset = 156.55 ± 0.2 ◦C, <sup>Δ</sup>H = 28.45 ± 0.40 Jg<sup>−</sup>1) under an atmosphere of dry nitrogen.

#### *3.5. Synthesis of Mesogens* **M2–M5**

#### 3.5.1. Synthesis of **M2**

In a dry-oven 25 mL schlenk flask, 1-[CH2C6H4-4- -(CH=CH2)]-2-H-1,2-*closo*-C2B10H10 (**1**) (68 mg, 0.26 mmol), **M1** (300 mg, 0.52 mmol), and first-generation Grubbs' catalyst (21 mg, 0.026 mmol) were dissolved in CH2Cl2 (10 mL). The solution was stirred and refluxed for 48 h. The reaction was monitored by 1H NMR to the disappearance of the alkene protons of **M1**. After that, the mixture was cooled down to room temperature, filtered, and the brown solution was quenched by precipitation into 40 mL of methanol to afford a gray solid. This is filtered and purified by silica gel column chromatography (75% CH2Cl2–25% hexane), leading to compound **M2** as a pure white solid. Yield: 114 mg, 54%. A CH2Cl2 solution of **M2** gave crystals suitable for X-ray analysis. 1H NMR (CDCl3, TMS), δ (ppm): 7.97 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.30 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.05 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.89 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.40 (d, 3J(H,H) = 16 Hz, 1H, C6H4C*H*=CH), 6.26 (dt, 3J(H,H) = 16 Hz, 3J(H,H) = 8 Hz, 1H, C6H4CH=C*H*), 5.40 (d, 3J(H,H) = 6 Hz, 1H, (Ch)C*H*=CH), 4.81 (m, 1H, OC*H* (Ch)), 4.05 (t, 2H, 3J(H,H) = 6 Hz, C*H*2O), 3.48 (s, 2H, Cc-C*H*2), 3.19 (s, 1H, Cc-*H*), 2.42 (m, 4H, OCHC*H*<sup>2</sup> (Ch) + C6H4CH=CH*CH2*), 1.05 (s, 3H, C-C*H*3), 0.91 (d, 3J(H,H) = 6 Hz, 3H, CHC*H*3), 0.85 (dd, 3J(H,H) = 6 Hz, 3J(H,H) = 3 Hz, 6H, CH(C*H*3)2), 0.67 (s, 3H, C-C*H*3). 11B{1H} NMR (CDCl3, BF3·Et2O), δ (ppm): −3.33 (1B), −6.48 (1B), −9.90 (2B), −11.54 ppm (2B), −14.05 (4B). 13C{1H} NMR (CDCl3, TMS), δ (ppm): 165.87 (s, *C*=O), 162.71 (s, *Car*-O), 139.84 (s, *C*=CH (Ch)), 137.84 (s, *C*6H4), 132.92 (s, *C*6H4), 131.62 (s, *C*6H4), 130.79 (s, *C*6H4), 130.08 (s, C6H4*C*H=CH), 129.93 (s, C6H4CH=*C*H), 126.62 (s, *C*6H4), 123.29 (s, *C*6H4), 122.78 (s, *C*6H4), 114.06 (s, C=*C*H (Ch)), 74.32 (s, *C*H-O), 67.31 (s, *C*H2-O), 56.78 (s, *C*H (Cp)), 56.21 (s, *C*H (Cp)), 50.12 (s, *C*H (Ch)), 43.22 (s, *C*H (Cp-Ch)), 42.41 (s, Cc-*C*H2), 39.83 (s, *C*H2CH(CH3)2), 39.60 (s, *C*H (Ch)), 38.38 (s, *C*H2C-O), 37.14 (s, *C*H (Ch)), 36.74 (s, *C*H (Ch)), 36.27 (s, CH3CH*C*H2), 35.89 (s, CH3*C*HCH2), 32.02 (s, *C*H (Ch)), 31.97 (s, *C*H (Ch)), 28.79 (s, CH=CH*C*H2CH2), 28.33 (s, *C*H (Ch)), 28.11 (s, *C*H (Cp)), 28.03 (s, *C*H-(CH3)2), 24.38 (s, *C*H2 (Cp)), 26.90 (s, *C*H (Ch)), 24.38 (s, *C*H2CH2CH-(CH3)2), 23.92 (s, *C*H (Ch)), 22.92 (s, CH-(*C*H3)2), 22.66 (s, CH-(*C*H3)2), 21.13 (s, *C*H (Ch)), 19.48 (s, CH-*C*H3), 18.81 (s, CH-*C*H3), 11.95 ppm (s, CH-*C*H3). ATR (cm−1): 2939 (br s, C-H st), 2576 (br s, B-H st), 1697 (s, C=O st), 1604 (m, C=C st). MALDI-TOF-MS (*m*/*z*): Calcd: 808.66. Found: 846.55 [M + K+]. Anal. Calcd. for C48H74B10O3: C, 71.42; H, 9.24. Found: C, 71.58; H, 9.15.

#### 3.5.2. Synthesis of **M3**

Compound **M3** was synthesized using the same procedure as for **M2**, but using 1- [CH2C6H4-4- -(CH=CH2)]-2-CH3-1,2-*closo*-C2B10H10 (**2**) (23 mg, 0.08 mmol), **M1** (100 mg, 0.17 mmol) and first-generation Grubbs catalyst (7 mg, 0.008 mmol) in 10 mL of CH2Cl2. After work-up, **M3** was obtained as a pure white solid. Yield: 38 mg, 50%. A CH2Cl2 solution of **M3** gave crystals suitable for X-ray analysis. 1H NMR (CDCl3, TMS), δ (ppm): 7.97 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.30 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.10 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.90 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.41 (d, 3J(H,H) = 16 Hz, 1H, C6H4C*H*=CH), 6.25 (dt, 3J(H,H) = 16 Hz, 3J(H,H) = 6 Hz, 1H, C6H4CH=C*H*), 5.40 (d, 3J(H,H) = 6 Hz, 1H, (Ch)C*H*=CH), 4.81 (m, 1H, OC*<sup>H</sup>* (Ch)), 4.05 (t, 2H, 3J(H,H) = 6 Hz, C*H*2O), 3.42 (s, 2H, Cc-C*H*2), 2.42 (m, 4H, OCHC*H*<sup>2</sup> (Ch)+C6H4CH=CH*CH2*), 2.14 (s, 3H, Cc-C*H*3), 1.06 (s, 3H, C-C*H*3), 0.91 (d, 3J(H,H) = 6 Hz, 3H, CHC*H*3), 0.85 (dd, 3J(H,H) = 6 Hz, 3J(H,H) = 3 Hz, 6H, CH(C*H*3)2), 0.67 (s, 3H, C-C*H*3). 11B{1H} NMR (CDCl3, BF3·Et2O), <sup>δ</sup> (ppm): −4.99 (1B), −6.55 (1B), −10.65 (4B), −11.29 ppm (4B). 13C{1H} NMR (CDCl3, TMS), δ (ppm): 165.84 (s, *C*=O), 162.73 (s, *Car*-O), 139.86 (s, *C*=CH (Ch)), 137.38 (s, *C*6H4), 133.75 (s, *C*6H4), 131.60 (s, *C*6H4), 130.56 (s, *C*6H4), 130.35 (s, C6H4*C*H=CH), 130.11 (s, C6H4CH=*C*H), 126.20 (s, *C*6H4), 122.71 (s, *C*6H4), 114.01 (s, C=*C*H (Ch)), 74.27 (s, *C*H-O), 67.36 (s, *C*H2-O), 56.78 (s, *C*H (Cp)), 56.21 (s, *C*H (Cp)), 50.12 (s, *C*H (Ch)), 42.40 (s, *C*H (Cp-Ch)), 41.01 (s, Cc-*C*H2), 39.82 (s, *C*H2CH(CH3)2), 39.60 (s, *C*H (Ch)), 38.38 (s, *C*H2C-O), 37.14 (s, *C*H (Ch)), 36.74 (s, *C*H (Ch)), 36.27 (s, CH3CH*C*H2), 35.89 (s, CH3*C*HCH2), 32.02 (s, *C*H (Ch)), 31.97 (s, *C*H (Ch)), 29.47 (s, CH=CH*C*H2CH2), 28.82 (s, CH=CHCH2*C*H2), 28.33 (s, *C*H (Ch)), 28.11 (s, *C*H (Cp)), 28.03 (s, *C*H-(CH3)2), 24.38 (s, *C*H2 (Cp)), 23.91 (s, *C*H2CH2CH-(CH3)2), 23.78 (s, Cc-*C*H3), 22.92 (s, CH-(*C*H3)2), 22.66 (s, CH-(*C*H3)2), 21.13 (s, *C*H (Ch)), 19.48 (s, CH-*C*H3), 18.81 (s, CH-*C*H3), 11.95 ppm (s, CH-*C*H3). ATR (cm−1): 2931 (br s, C-H st), 2584 (br s, B-H st), 1705 (s, C=O st), 1604 (m, C=C st). MALDI-TOF-MS (*m*/*z*): Calcd: 822.67. Found: 860.56 [M + K+]. Anal. Calcd. for C49H76B10O3: C, 71.66; H, 9.33. Found: C, 72.11; H, 9.32.

#### 3.5.3. Synthesis of **M4**

Compound **M4** was synthesized using the same procedure as for **M2** and **M3** but using 1-[CH2C6H4-4- -(CH=CH2)]-2-C6H5-1,2-*closo*-C2B10H10 (**3**) (50 mg, 0.15 mmol), **M1** (171 mg, 0.30 mmol) and first-generation Grubbs catalyst (12 mg, 0.015 mmol) in 6 mL of CH2Cl2. After work-up, **M4** was obtained as a pure white solid. Yield: 64 mg, 49%. 1H NMR (CDCl3, TMS), δ (ppm): 7.99 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.71 (d, 3J(H,H) = 8 Hz, 2H, C6*H*5), 7.54–7.44 (m, 3H, C6*H*5), 7.20 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.10 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.90 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.75 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.38 (d, 3J(H,H) = 16 Hz, 1H, C6H4CH=C*H*), 6.23 (dt, 3J(H,H) = 16 Hz, 3J(H,H) = 6 Hz, 1H, C6H4CH=C*H*), 5.41 (d, 3J(H,H) = 6 Hz, 1H, (Ch)C*H*=CH), 4.83 (m, 1H, OC*H* (Ch)), 4.05 (t, 2H, 3J(H,H) = 6 Hz, C*H*2O), 3.04 (s, 2H, Cc-C*H*2), 2.42 (m, 4H, OCHC*H*<sup>2</sup> (Ch) + C6H4CH=CH*CH2*), 0.91 (d, 3J(H,H) = 6 Hz, 3H, CHC*H*3), 0.86 (dd, 3J(H,H) = 6 Hz, 3J(H,H) = 3 Hz, 6H, CH(C*H*3)2), 0.68 (s, 3H, C-C*H*3). 11B{1H} NMR (CDCl3, BF3·Et2O), <sup>δ</sup> (ppm): −4.39 (2B), −10.92 (8B); 13C{1H} NMR (CDCl3, TMS), <sup>δ</sup> (ppm): 165.87 (s, *<sup>C</sup>*=O), 162.74 (s, *Car*-O), 139.86 (s, *C*=CH (Ch)), 137.15 (s, *C*6H4), 133.99 (s, *C*6H4), 131.62 (s, *C*6H4), 131.59 (s, *C*6H5), 130.93 (s, *C*6H5), 130.91 (s, *C*6H5), 130.56 (s, *C*6H4), 130.31 (s, C6H4*C*H=CH), 130.21 (s, *C*6H4), 130.10 (s, C6H4CH=*C*H), 129.13 (s, *C*6H5), 125.96 (s, *C*6H4), 123.26 (s, *C*6H4), 122.77 (s, *C*6H4), 114.07 (s, C=*C*H (Ch)), 83.76 (s, *Cc*-C6H5), 82.19 (s, *Cc*-CH2), 74.30 (s, *C*H-O), 67.34 (s, *C*H2-O), 56.79 (s, *C*H (Cp)), 56.22 (s, *C*H (Cp)), 50.13 (s, *C*H (Ch)), 42.41 (s, *C*H (Cp-Ch)), 40.72 (s, Cc-*C*H2), 39.83 (s, *C*H2CH(CH3)2), 39.61 (s, *C*H (Ch)), 38.39 (s, *C*H2C-O), 37.15 (s, *C*H (Ch)), 36.75 (s, *C*H (Ch)), 36.28 (s, CH3CH*C*H2), 35.90 (s, CH3*C*HCH2), 32.03 (s, *C*H (Ch)), 31.97 (s, *C*H (Ch)), 29.46 (s, CH=CH*C*H2CH2), 28.82 (s, CH=CHCH2*C*H2), 28.34 (s, *C*H (Ch)), 28.11 (s, *C*H (Cp)), 28.04 (s, *C*H-(CH3)2), 24.39 (s, *C*H2 (Cp)), 23.93 (s, *C*H2CH2CH-(CH3)2), 22.93 (s, CH-(*C*H3)2), 22.67 (s, CH-(*C*H3)2), 21.15 (s, *C*H (Ch)), 19.49 (s, CH-*C*H3), 18.82 (s, CH-*C*H3), 11.96 (s, CH-*C*H3). ATR (cm−1): 2939 (br s, C-H st), 2584 (br s, B-H st), 1705 (s, C=O st), 1604 (m, C=C st). MALDI-TOF-MS (*m*/*z*): Calcd: 882.64. Found: 921.57 [M + K+]. Anal. Calcd. for C54H78B10O3: C, 73.43; H, 8.90. Found: C, 72.89; H, 8.92.

#### 3.5.4. Synthesis of **M5**

Compound **M5** was synthesized using the same procedure as for **M2**, but using styrene (0.5 mL 5.3 mmol), **M1** (100 mg, 0.17 mmol) and first-generation Grubbs catalyst (20 mg, 0.024 mmol) were dissolved in CH2Cl2 (5 mL). After work-up, **M5** was obtained as a pure white solid. Yield: 93 mg, 82%. A toluene solution of **M5** gave crystals suitable for X-ray analysis. 1H NMR (CDCl3, TMS), δ (ppm): 7.97 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 7.35–7.15 (m, 5H, C6*H*5), 6.90 (d, 3J(H,H) = 8 Hz, 2H, C6*H*4), 6.43 (d, 3J(H,H) = 12 Hz, 1H, C6H4C*H*=CH), 6.23 (dt, 3J(H,H) = 12 Hz, 3J(H,H) = 4.5 Hz, 1H, C6H4CH=C*H*), 5.40 (d, 3J(H,H) = 6 Hz, 1H, (Ch)C*H*=CH), 4.81 (m, 1H, OC*<sup>H</sup>* (Ch)), 4.05 (t, 2H, 3J(H,H) = 6 Hz, C*H*2O), 2.42 (m, 2H, OCHC*H*<sup>2</sup> (Ch)), 1.06 (s, 3H, C-C*H*3), 0.91 (d, 3J(H,H) = 6 Hz, 3H, CHC*H*3), 0.85 (dd, 3J(H,H) = 6 Hz, 3J(H,H) = 3 Hz, 6H, CH(C*H*3)2), 0.67 (s, 3H, C-C*H*3). 13C {1H} NMR (CDCl3, TMS), <sup>δ</sup> (ppm): 165.89 (s, *<sup>C</sup>*=O), 162.77 (s, *Car*-O), 139.86 (s, *<sup>C</sup>*=CH (Ch)), 137.60 (s, *C*6H5), 131.60 (s, *C*6H4), 130.89 (s, *C*6H5), 129.48 (s, C6H5), 128.60 (s, C6H5), 127.14 (s, C6H5), 126.20 (s, *C*6H5), 123.22 (s, C6H4), 122.71 (s, *C*6H4), 114.01 (s, C=*C*H (Ch)), 74.27 (s, *C*H-O), 67.36 (s, *C*H2-O), 56.78 (s, *C*H (Cp)), 56.21 (s, *C*H (Cp)), 50.12 (s, *C*H (Ch)), 42.40 (s, *C*H (Cp-Ch)), 39.82 (s, *C*H2CH(CH3)2), 39.60 (s, *C*H (Ch)), 38.38 (s, *C*H2C-O), 37.14 (s, *C*H (Ch)), 36.74 (s, *C*H (Ch)), 36.27 (s, CH3CH*C*H2), 35.89 (s, CH3*C*HCH2), 32.02 (s, *C*H (Ch)), 31.97 (s, *C*H (Ch)), 29.47 (s, CH=CH*C*H2CH2), 28.82 (s, CH=CHCH2*C*H2), 28.33 (s, *C*H (Ch)), 28.11 (s, *C*H (Cp)), 28.03 (s, *C*H-(CH3)2), 24.38 (s, *C*H2 (Cp)), 23.91 (s, *C*H2CH2CH-(CH3)2), 22.92 (s, CH-(*C*H3)2), 22.66 (s, CH-(*C*H3)2), 21.13 (s, *C*H (Ch)), 19.48 (s, CH-*C*H3), 18.81 (s, CH-*C*H3), 11.95 ppm (s, CH-*C*H3). ATR (cm−1): 2931 (br s, C-H st), 1697 (s, C=O st), 1604 (m, C=C st). MALDI-TOF-MS (*m*/*z*): Calcd: 650.47. Found: 673.46 [M + Na+]. Anal. Calcd. for C45H62O3: C, 83.03; H, 9.60. Found: C, 81.92; H, 9.55. (+1/2 MeOH C45.5H64O3.5: C, 81.99; H, 9.60).

#### **4. Conclusions**

In conclusion, the olefin cross-metathesis between suitably substituted *o*-carborane derivatives and a mesogenic precursor as the modified cholesterol offers a strategy to obtain liquid crystalline materials. The reaction of styrenyl-*o*-carborane derivatives with a terminal alkenyl cholesteryl benzoate yields only the desired *trans*-regioisomer, which is the best-suited configuration to attain mesogenic behavior. The substituent linked to the Cc atom of the *o*-carborane plays an important role in tuning the liquid crystal behavior; for non-substituted *o*-carborane (**M2**, R = H) and methyl-substituted *o*-carborane (**M3**, R = Me), the N\* phase was obtained, whereas **M4** (R = Ph) was not mesogenic. In addition, **M3** exhibited blue phase, which was observed neither in the chiral nematic precursor **M1** nor in the model compound **M5**, which suggested that the presence of the *o*-carborane cluster leads to stabilizing this highly twisted chiral phase in this set of compounds. In these cases, although the thermal stability and the range of the N\* phase is substantially decreased in comparison with the precursors, the *o*-carborane cluster can be incorporated without destroying the helical organization of the mesophase.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2073-435 2/11/2/133/s1. Scheme S1. Synthesis of the mesogen **M1**, Figure S1. 1H-NMR of compound **M2**, Figure S2. 11B{1H}-NMR of compound **M2**, Figure S3. 13C{1H}-NMR of compound **M2**, Figure S4. 1H-NMR of compound **M3, Figure S5**. 11B{1H}-NMR of compound **M3, Figure S6**. 13C{1H}-NMR of compound **M3**, Figure S7. 1H-NMR of compound **M4**, Figure S8. 11B{1H}-NMR of compound **M4**, Figure S9. 13C{1H}-NMR of compound **M4**, Figure S10. 1H-NMR of compound **M5**, Figure S11. 13C{1H}-NMR of compound **M5**, Figure S12. FT-IR of compound **M2**, Figure S13. FT-IR of compound **M3**, Figure S14. FT-IR of compound **M4**, Figure S15. FT-IR of compound **M5**, Figure S16. View of independent molecules in the unit cell of **M2** (A) and **M3** (B) with numbering, Figure S17. View of independent molecules in the unit cell of **M5** with numbering, Figure S18. Views of crystal packings

of **M2** (A), **M3** (B) and **M5** (C) are shown in the left. Detail view of molecular self-assembly is shown in the right. Independent molecules in the unit cells are colored. H atoms are omitted for clarity, Figure S19. Comparison between 2D fingerprint plots for **M2** (A), **M3** (B) and **M5** (C), Figure S20. Normalized UV-Visible and fluorescence spectra of **M2-M3** and **M5**, Figure S21. DSC of compound **M2** (3 cycles at 10 ◦C/min), Figure S22. DSC of compound **M3** (3 cycles at 10 ◦C/min), Figure S23. DSC of compound **M4** (3 cycles at 10 ◦C/min), Figure S24. DSC of compound **M5** (3 cycles at 10 ◦C/min), Figure S25. Selected POM images of compound **M2**. Left side: Registered at 132 ◦C during the cooling cycle (rate 1 ◦C/min), showing N\* phase. Right side: Registered at 119 ◦C during the cooling cycle (rate 1 ◦C/min), showing the transition from N\* to crystalline phase, Figure S26. Selected POM images of compound **M5**, registered at 204.3 ◦C during the cooling cycle (rate 0.5 ◦C/min), showing N\* phase from Isotropic, Table S1. Crystal and Structure Refinement data for **M2**, **M3** and **M5**, Table S2. Selected bond lengths for **M2**, **M3** and **M5**, Table S3. Bond lengths for **M2**, **M3** and **M5**, Table S4. Bond angles for **M2**, **M3** and **M5**.

**Author Contributions:** Manuscript conception, R.N. and I.M.S.; writing and original draft preparation, R.N. and I.M.S.; synthesis and characterization of derivatives **M2–M5**, A.F.-U. and A.G.-C.; photophysical and liquid crystal, A.F.-U. and A.G.-C.; X-ray analysis, J.G.P.; editing, data analysis and interpretation, A.F.-U., A.G.-C., F.T., C.V., R.N. and I.M.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by MICINN grants (PID2019-106832RB-100 and CEX2019-000917- S) and Generalitat de Catalunya (2017/SGR/1720).

**Acknowledgments:** This work was supported by Agencia Estatal de Investigación AEI from MICINN (CTQ2016-75150-R and PID2019-106832RB-100/AEI/10.13039/501100011033 project) and Generalitat de Catalunya (2017 SGR1720 projects). The work was also supported by the MICINN through the Severo Ochoa Program for Centers of Excellence FUNFUTURE (CEX2019-000917-S). We gratefully acknowledge A. Whitwood for collecting the X-ray single crystals diffraction pattern.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


#### *Article* **Exploration of Bis(nickelation) of 1,1 -Bis(***o***-carborane)**

**Dipendu Mandal \*,† and Georgina M. Rosair \***

Institute of Chemical Sciences, School of Engineering & Physical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK

**\*** Correspondence: chmdip@nus.edu.sg (D.M.); g.m.rosair@hw.ac.uk (G.M.R.)

† Present address: Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore.

**Abstract:** The metalation of [Tl]2[1-(1- -3- ,1- ,2- -*closo*-TlC2B9H10)-3,1,2-*closo*-TlC2B9H10], with the smaller {Ni(dmpe)} fragment sourced from [Ni(dmpe)Cl2], is explored. The bis(metalated) products are obtained as a diastereoisomeric mixture. These isomers were separated, fully characterised spectroscopically and crystallographically and identified as *rac*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)- 3-(dmpe)-3,1,2-*closo*-NiC2B9H10] (**1**) and *meso*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dmpe)- 3,1,2-*closo*-NiC2B9H10] (**2**). Previously, these 3,1,2-NiC2B9-3- ,1- ,2- -NiC2B9 architectures (where both cages are not isomerised), were inaccessible, and thus new structures can be achieved during bis(nickelation) with {Ni(dmpe)}. Further, the metalation of the tetra-thallium salt with the bulky {Ni(dppe)} fragment sourced from [Ni(dppe)Cl2] was also studied. These bis(nickelated) products were also fully characterised and are afforded as the stereospecific species *rac*-[1-(1- -3- -(dppe)-3- ,1- ,2- *closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**3**) and [1-(2- -4- -(dppe)-4- ,1- ,2- -*closo*-NiC2B9H10)- 3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**4**α). In the latter metalation, compound **3** shows intramolecular dihydrogen bonding, contributing to the stereospecificity, whereas isomerisation from 3,1,2 to 4,1,2 in the **4**α is related to steric relief.

**Keywords:** 1,1- -bis(*o*-carborane); deboronation; metalation; bis(nickelation); diastereoisomers; stereospecific

#### **1. Introduction**

Since the discovery of bis(carboranes) in 1964, the chemistry of 1,1- -bis(*o*-carborane) (Scheme 1, **a**) has evolved rapidly, particularly once a high-yielding synthetic route was devised in 2003. Bis(carborane) offers a versatile building block in designing threedimensional molecules, an array of homogeneous catalyst precursors, luminescent materials and organic derivatives via Ccage-H or Bcage-H functionalisation [1–6].

Since 2010, the Welch group have established many variations of the metalation chemistry of 1,1- -bis(*o*-carborane), one approach being cage expansion chemistry via reductionmetalation of this species [1], whilst another metalation strategy explored broadly within the group is deboronation/metalation of 1,1- -bis(*o*-carborane).

The single deboronation/metalation of 1,1- -bis(*o*-carborane) has been reported for cobalt, nickel and ruthenium metal fragments and afforded a wide range of mono-metallicbis(carborane) isomers [7,8]. In further developments, double deboronation/metalation was achieved, such that both cages became metallacarboranes. The bimetallic metallacarboranes derived from the metalation of doubly deboronated 1,1- -bis(*o*-carborane) with both rhodium and ruthenium fragments are 3,1,2-MC2B9-2- ,1- ,8- -MC2B9 [M = {Rh}, {Ru}] species, in which one of the cages has isomerised (Scheme 1, **b** and **c**). There are limited examples of the parent 3,1,2-MC2B9-3- ,1- ,2- -MC2B9 form [9,10]. Variation in the isomer type with cobalt has been achieved by varying the metalation source, forming either 3,1,2- CoC2B9-3- ,1- ,2- -CoC2B9 (non-isomerised) or 3,1,2-CoC2B9-2- ,1- ,8- -CoC2B9 (isomerised) (Scheme 1, **c**) products [9]. Notably, an example of stepwise deboronation/metalationdeboronation/heterometalation is also reported [11]. Here we document an expansion

**Citation:** Mandal, D.; Rosair, G.M. Exploration of Bis(nickelation) of 1,1- - Bis(*o*-carborane). *Crystals* **2021**, *11*, 16. https://dx.doi.org/10.3390/cryst11 010016

Received: 10 December 2020 Accepted: 22 December 2020 Published: 27 December 2020

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/ licenses/by/4.0/).

of the bimetallic metallacarboranes library via the double deboronation/nickelation of 1,1- -bis(*o*-carborane).

**Scheme 1.** Line diagrams of **a**–**c**. Species **a** is 1,1- -bis(*o*-carborane), whereas **b** (α-form) and **c** (β-form) were derived via double deboronation/metalation of **a**. Unlabelled vertices are B.

#### **2. Materials and Methods**

#### *2.1. General Considerations*

Experiments were carried out under dry, oxygen-free N2, using standard Schlenk techniques, although subsequent manipulations were performed at ambient condition. Tetrahydrofuran (THF) was dried and distilled under sodium/benzophenone, whilst petrol was distilled from sodium wire before use. DCM was purified in an MBRAUN SPS-800(Dieselstr. 31, D-85748 Garching,). Degassing of solvents was performed (3 × freeze-pump-thaw cycles) before reaction. Preparative TLC used Kieselgel F254 glass plates (20 <sup>×</sup> 20 cm). 1H (400.1 MHz), 31P (162.0 MHz) or 11B (128.4 MHz) NMR spectra and 1H-31P Heteronuclear Multiple Bond Correlation (HMBC) experiment (in Supplementary Materials) were run on a Bruker DPX-400 spectrometer (Bruker BioSpin AG, Fallenden, Switzerland). The precursors 1,1- -bis(*o*-carborane) [12], its deboronated derivative [Tl]2[1-(1- -3- ,1- ,2- *closo*-TlC2B9H10)-3,1,2-*closo*-TlC2B9H10] (Tl4-salt) (WARNING: Thallium is extremely toxic, appropriate precautions are required when handling thallium compounds) [9] and [NiCl2 (dmpe)] (dmpe = 1,2-bis(dimethylphosphino)ethane) [13] were prepared by modified literature methods. [NiCl2(dppe)] (dppe = 1,2-bis(diphenylphosphino)ethane) and the remaining reagents were purchased commercially.

2.1.1. Synthesis and Characterisation of *rac*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3- (dmpe)-3,1,2-*closo*-NiC2B9H10] (**1**) and *meso*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3- (dmpe)-3,1,2-*closo*-NiC2B9H10] (**2**)

The Tl4-salt (0.60 g, 0.56 mmol) was taken into THF (15 mL). The yellow suspension was degassed by freeze-pump-thaw (three cycles). [NiCl2(dmpe)] (0.31 g, 1.1 mmol) was transferred at −196 ◦C. The reaction suspension was allowed to warm and was stirred at room temperature. After overnight stirring, the mixture turned dark green. All volatiles were removed *in vacuo*. The residue was dissolved in DCM and passed through a small pad of silica. The filtrate was reduced in volume under low pressure and purified by preparative TLC using DCM and petrol (80:20) to afford two dark green bands, which were collected as solids. The upper green band with Rf = 0.84 afforded *rac*-[1-(1- -3- -(dmpe)- 3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dmpe)-3,1,2-*closo*-NiC2B9H10] (**1**) (80 mg, 21%) and a lower green band with Rf = 0.63 gave *meso*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dmpe)- 3,1,2-*closo*-NiC2B9H10] (**2**) (76 mg, 20%).

Compound **1**: 1H NMR (CD2Cl2): δ 2.43 (d, <sup>3</sup>*J*PH = 14.0 Hz, 2H, C*H*cage), 2.14–1.90 (m, 8H, P{C*H*2}2P), 1.83 (d, <sup>2</sup>*J*PH = 10.0 Hz, 6H, C*H*3), 1.67 (d, <sup>2</sup>*J*PH = 10.0 Hz, 6H, C*H*3), 1.55 (d, <sup>2</sup>*J*PH = 10.0 Hz, 12H, C*H*3). 1H{31P} NMR (CD2Cl2): δ 2.43 (s, 2H, C*H*cage), 2.12–1.89 (m, 8H, P{C*H*2}2P), 1.83 (s, 6H, C*H*3), 1.67 (s, 6H, C*H*3), 1.56 (s, 12H, C*H*3). 31P{1H} NMR

(CD2Cl2): δ 43.1 (d, <sup>2</sup>*J*PP = 29.2 Hz, 2P), 33.0 (d, <sup>2</sup>*J*PP = 29.2 Hz, 2P). 11B{1H} NMR (CD2Cl2): δ −2.7 (2B), −4.1 (2B), −7.3 (2B), −11.8 (2B), −13.6 (2B), −15.8 (4B), −21.1 (4B).

Compound **2**: 1H NMR (CD2Cl2): δ 2.48 (d, <sup>3</sup>*J*PH = 10.0 Hz, 2H, C*H*cage), 2.16–1.79 (m, 8H, P{C*H*2}2P), 1.73 (d, <sup>2</sup>*J*PH = 10.0 Hz, 6H, C*H*3), 1.70 (d, <sup>2</sup>*J*PH = 8.0 Hz, 6H, C*H*3), 1.57 (d, <sup>2</sup>*J*PH = 10.0 Hz, 6H, C*H*3), 1.55 (d, <sup>2</sup>*J*PH = 8.0 Hz, 6H, C*H*3). 1H{31P} NMR (CD2Cl2): δ 2.49 (s, 2H, C*H*cage), 2.14–1.78 (m, 8H, P{C*H*2}2P), 1.74 (s, 6H, C*H*3), 1.70 (s, 6H, C*H*3), 1.57 (s, 6H, C*H*3), 1.55 (s, 6H, C*H*3). 31P{1H} NMR (CD2Cl2): δ 43.7 (d, <sup>2</sup>*J*PP = 28.4 Hz, 2P), 33.4 (d, <sup>2</sup>*J*PP = 28.4 Hz, 2P). 11B{1H} NMR (CD2Cl2): <sup>δ</sup>−1.5 (2B), −4.0 (2B), −10.4 (2B), −11.5 (2B), −14.5 (3B), −15.9 (5B), −20.4 (2B).

2.1.2. Synthesis and Characterisation of *rac*-[1-(1- -3- -(dppe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3- (dppe)-3,1,2-*closo*-NiC2B9H10] (**3**) and [1-(2- -4- -(dppe)-4- ,1- ,2- -*closo*–NiC2B9H10)-3-(dppe)- 3,1,2-*closo*-NiC2B9H10] (**4**α)

A yellow suspension of Tl4-salt (0.60 g, 0.56 mmol) in THF (20 mL) was frozen at −196 ◦C. [NiCl2(dppe)] (0.59 g, 1.1 mmol) added to the frozen mixture. After overnight stirring, the reaction mixture gave a green suspension. All volatiles were evaporated off under reduced pressure. The mixture was taken into DCM and filtered through silica. The filtrate was reduced in a minimum amount under low pressure. The mixture was purified by preparative TLC with DCM and petrol (60:40). This yielded, along with a trace amount of purple band with Rf = 0.39, two major mobile components. The lower army-green band with Rf = 0.28 collected as a solid, and after crystallisations, yielded *rac*-[1-(1- -3- -(dppe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**3**) (130 mg, 20%). The upper army green-band with Rf = 0.35 afforded [1-(2- -4- -(dppe)-4- ,1- ,2- -*closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**4**α) (122 mg, 19%).

Compound **3**: 1H NMR (CD2Cl2): δ 7.80–7.25 (m, 40H, C6*H*5), 3.29–1.96 (m, 8H, P{C*H*2}2P), 1.83 (d, <sup>3</sup>*J*PH = 11.6 Hz, 2H, C*H*cage). 1H{31P} NMR (CD2Cl2): δ 7.79–7.25 (m, 40H, C6*H*5), 3.29–1.97 (m, 8H, P{C*H*2}2P), 1.84 (s, 2H, C*H*cage). 31P{1H} NMR (CD2Cl2): δ 51.6 (d, <sup>2</sup>*J*PP = 19.9 Hz, 2P), 39.2 (d, <sup>2</sup>*J*PP = 19.9 Hz, 2P). 11B{1H} NMR (CD2Cl2): δ 4.9 (2B), −2.5 (4B), −9.0 (2B), −14.7 (5B), −18.0 (5B).

Compound **4**α: 1H NMR (CD2Cl2): δ 8.01–7.02 (m, 40H, C6*H*5), 3.67–2.26 (m, 8H, P{C*H*2}2P), 1.94 (s, 1H, C*H*cage), 1.66 (d, <sup>3</sup>*J*PH = 11.2 Hz, 1H, C*H*cage). 1H{31P} NMR (CD2Cl2): δ 8.02–7.01 (m, 40H, C6*H*5), 3.68–2.25 (m, 8H, P{C*H*2}2P), 1.95 (s, 1H, C*H*cage), 1.67 (s, 1H, <sup>C</sup>*H*cage). 31P{1H} NMR (CD2Cl2): <sup>δ</sup> 62.9 (s, 2P), 49.4 (d, <sup>2</sup>*J*PP = 16.6 Hz, 1P), 47.5 (d, <sup>2</sup>*J*PP = 16.6 Hz, 1P). 11B{1H} NMR (CD2Cl2): <sup>δ</sup> 5.9 (1B), 1.7 (1B), <sup>−</sup>0.2 (1B), <sup>−</sup>4.6 (3B), <sup>−</sup>9.8 to −20.4 multiple overlapping resonances with maxima at −9.8, −13.5, −15.6, −16.8, −20.4 (total integral of last five resonances 12B).

#### *2.2. Crystallographic Studies*

X-ray diffraction quality crystals of **1**–**4**α were obtained by solvent diffusion at 5 ◦C using DCM and petrol 40–60 as antisolvent. Intensity data for compounds **1**–**3** were collected on a Bruker X8 APEXII diffractometer, whereas for compound **4**α on a Rigaku FRE+ equipped with VHF Varimax confocal mirrors and an AFC10 goniometer and HG Saturn 724+ detector using Mo-K*<sup>α</sup> X*-radiation at the UK National Crystallography Service. All crystals were mounted in inert oil on a cryoloop and cooled to 100 K by an Oxford Cryosystems Cryostream. Indexing, data collection and absorption correction were performed using the APEXII suite of programmes [14]. Structures were solved with the SHELXS programme [15] and refined by full-matrix least-squares (SHELXL), using OLEX2 [16]. Table 1 summarises the crystallographic parameters. The location of the CH vertices in all cases was established by the Vertex to Centroid (VCD) method developed by the Welch group [17]. The VCD method is based on the fact that, in a carborane cage, the C vertices are closer to the cage centroid than the B vertices. The CH vertex's location was corroborated by the Boron-to-Hydrogen distance method ("B"–H bond lengths for actual C atoms refined to ca. <0.8Å) in all cases except **4**α, where the positions of the H atoms could not be freely refined against the weak and twinned diffraction data. In this case, B-H distances were restrained to 1.10(2) Å.


**Table 1.** Crystallographic data for compounds **1**–**4**α.

**1** was treated as a two-component crystal and refined with hklf 5 data. In **4**α, disordered phenyl carbon atoms were constrained to have equal displacement parameters. The crystal of **4**α was not single and refined as a two-component twin with the twin law 1 0 0 0 –1 0 0 0 –1 against HKLF 4 data, which were themselves treated with the solvent mask procedure implemented in OLEX2. HKLF 5 data were not available from the National crystallography service for this structure. A refinement model of the solvent showed approximately three molecules of highly disordered CH2Cl2 per bis carborane.

#### **3. Results and Discussion**

#### *3.1. Characterisation of Compounds 1 and 2*

Previously, the double deboronated [7-(7- -7- ,8- -*nido*-C2B9H10)-7,8-*nido*-C2B9H10] 4− tetraanion derived from [1-(1- -1- ,2- -*closo*-C2B10H11)-1,2-*closo*-C2B10H11], commonly called 1,1- -bis(*o*-carborane), was isolated as [HNMe3] <sup>+</sup> or [BTMA]+ [BTMA = benzyltrimethylammonium] salts. Later, [Tl]<sup>+</sup> salts were chosen for metalation because of their generally better yields [9]. We have reacted a suspension of [Tl]2[1-(1- -3- ,1- ,2- -*closo*-TlC2B9H10)-3,1,2-*closo*-TlC2B9H10] in THF with [Ni(dmpe)Cl2] at room temperature. Workup and purification of the dark-green suspension involving preparative thin-layer chromatography (TLC) re-

sulted in two dark green bands, compounds **1** and **2**, in moderate yield (Scheme 2). These were fully characterised by a variety of spectroscopic and crystallographic analyses.

**Scheme 2.** Synthesis of **1** and **2** from the nickelation of [Tl]2[1-(1- -3- ,1- ,2- -*closo*-TlC2B9H10)-3,1,2-*closo*-TlC2B9H10]. NiPP = {Ni(dmpe)}.

An interesting feature of the double deboronation-metalation of 1,1- -bis(*o*-carborane) is that it gives diastereomeric metallcarborane products. Indeed, compound **1** turns out to be *racemic*, whilst compound **2** is in the *meso* form. These observations are also consistent with the NMR spectroscopic analysis. The 1H NMR spectrum of compound **1** reveals resonances arising from the methylene bridge of the two dmpe ligands, whilst the methyl groups of the dmpe ligands produce three doublet resonances. The first two are of integral-6 at δ 1.83, 1.67 ppm with coupling <sup>2</sup>*J*PH = 10.0 Hz, and the last one is of integral-12 at δ 1.55, <sup>2</sup>*J*PH = 10.0 Hz. On the contrary, the 1H NMR spectrum of compound **2** reveals, in addition to the signals for methylene bridge of the dmpe ligands, four doublets assigned to the methyl protons of the dmpe fragments at δ 1.73 (2*J*PH = 10.0 Hz), 1.70 (2*J*PH = 8.0 Hz), 1.57 (2*J*PH = 10.0 Hz) and 1.55 (2*J*PH = 8.0 Hz) ppm, each of integral-6. These resonances collapse to the corresponding singlets in the 1H{31P} spectrum. In the 1H NMR spectra there is also a single C*H*cage resonance of integral two for compound **1** appearing as a doublet δ 2.43 (3*J*PH = 14.0 Hz) ppm, confirmed as arising from coupling to phosphorus, since it collapses to a singlet on broad-band 31P decoupling, whilst the C*H*cage signal of integral two for compound **2** appears at a higher frequency, δ 2.48 (3*J*PH = 10.0 Hz) ppm, and collapses to a singlet on broad-band 31P decoupling. Therefore, these two isomers show slightly different characteristics in their proton NMR spectra. Moving to the 31P{1H} NMR spectrum of compound **1**, this shows two mutual doublets with the integral ratio of 1:1 at δ 43.1 and 33.0 ppm and a coupling constant <sup>2</sup>*J*PP = 29.2 Hz. This indicates that in each cage the two phosphorus atoms are magnetically inequivalent. Moreover, the two pairs of phosphorus atoms in different cages are also magnetically equivalent. This clearly shows that the metallacarborane cages are asymmetric. Similarly, the 31P{1H} NMR spectrum of compound **2** reveals two mutually coupled doublets of integral two at δ 43.7

and 33.4 ppm with coupling <sup>2</sup>*J*PP = 28.4 Hz, not very different from that of compound **1**. Notably, from the 1H-31P HMBC experiment, the splitting of C*H*cage arises from the *trans* phosphorus at δ 33.0 ppm for compound **1** and the *trans* phosphorus at δ 33.4 ppm for compound **2**. Therefore, the resonances δ 43.1 ppm and δ 43.7 ppm correspond to the phosphorus atoms *cis* to the C*H*cage for compound **<sup>1</sup>** and compound **<sup>2</sup>**, respectively. The 11B{1H} NMR spectrum of compound **<sup>1</sup>** consists of seven resonances with relative integrals 2:2:2:2:2:4:4 from high frequency to low frequency, whereas compound **2** shows a distinctly different pattern comprising seven resonances in the integral ratio 2:2:2:2:3:5:2 from high frequency to low frequency. However, this spectroscopic information is not conclusive in determining the exact structures of the *racemic* and *meso* isomers.

A crystallographic study was carried out for both compounds **1** and **2**. It is envisaged that the double deboronation-metalation of 1,1- -bis(*o*-carborane) with {Ni(dmpe)} fragments generates *racemic* and *meso* mixtures with 3,1,2-NiC2B9-3- ,1- ,2- -NiC2B9 architectures. Indeed, both cages of compounds **1** and **2** are singly-metalated and in the 3,1,2-NiC2B9 form. As regards the identification of the diastereoisomer, a crystallographic *C*<sup>2</sup> axis passes perpendicular to the C1-C1 bond, leading to the same chirality for both cages and meaning that compound **1** is a racemic isomer, whereas a non-crystallographic inversion centre *i* can be imagined at the mid-point of the C1-C1 bond in **2**, meaning different chirality for each cage and showing that compound **2** is a meso isomer. Figure 1 shows a perspective view of a single molecule of *racemic*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)- 3-(dmpe)-3,1,2-*closo*-NiC2B9H10] (**1**), whereas a perspective view of a single molecule of *meso*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dmpe)-3,1,2-*closo*-NiC2B9H10] (**2**) is presented in Figure 2. These diastereoisomers are the first such examples in which both the metallacarborane moieties remain unisomerised, i.e.,**1** and **2** are 3,1,2-NiC2B9-3- ,1- ,2- - NiC2B9 in racemic and meso forms, previously inaccessible. Both compounds **1** and **2** show clear evidence of internal crowding, since both cages have 3,1,2-NiC2B9 architectures. The presence of the {(dmpe)NiC2B9} substituents on C1 or C1 inhibit the free rotation of the {Ni(dmpe)} fragment on both the primed and non-primed cages. It is confirmed that in each cage the phosphorus atoms are inequivalent, giving rise to two doublets observed in the 31P NMR spectra (previously discussed). Since both cages are of the 3,1,2-NiC2B9 form, the two metallacarborane units at C1 and C1 push each other away, so that the {Ni(dmpe)} fragment is bent away from its ideal orientation within the cages. In principle, the ideal orientation of the {NiP1P2} fragment in a NiC2B9 icosahedron is perpendicular to the vertical mirror plane through the C2B9 unit, conveniently defined by the interplane dihedral angle (θ) of 90◦. [8] For compound **1**, θ is found to be 55.36(8)◦ for both the primed cage and non-primed cage, whilst for compound **2**, θ is 57.59(15)◦ and 49.58(18)◦ for the primed cage and the non-primed cage, respectively. These were calculated using the planes through Ni3P1P2 and Ni3- P1- P2 and through B6B8B10 and B6- B8- B10- for the non-primed and primed cages, respectively. Thus, the dihedral angles are clearly twisted away from the idealised 90◦. We also note that in compound **1** the {NiPP} plane is bent away from the perpendicular to the bottom pentagonal B5B6B11B12B9 plane by.6(7)◦, whereas in compound **2** the corresponding angles are 6.1(12)◦ for the non-primed cage and 21.5(15)◦ for the primed cage. The internal steric crowding between two (dmpe)NiC2B9 units is also evidenced by the elongated Ni3-C1 distances compared to Ni3–C2 [compound **1** primed cage: 2.2977(19) *versus* 2.0657(18) Å]. This is also evidenced in compound **2**, Ni3-C1: 2.326(4) [Ni2-C2: 2.066(4) Å] and Ni3- -C1- : 2.299(4) [Ni2- -C2- : 2.103(3) Å].

**Figure 1.** Molecular structure of *rac*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dmpe)-3,1,2-*closo*-NiC2B9H10] (**1**). Atoms with dashed suffixes are generated by the symmetry operation 1.5-x, y, 0.5-z.

**Figure 2.** Molecular structure of *meso*-[1-(1- -3- -(dmpe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dmpe)-3,1,2-*closo*-NiC2B9H10] (**2**).

#### *3.2. Characterisation of Compounds 3 and 4α*

The treatment of [Tl]2[1-(1- -3- ,1- ,2- -*closo*-TlC2B9H10)-3,1,2-*closo*-TlC2B9H10] in THF with [Ni(dppe)Cl2] at room temperature followed by work up and purification involving preparative TLC afforded two army-green bands, compounds **3** and **4**α (Scheme 3). The compounds were characterised spectroscopically as well as by single crystal XRD.

**Scheme 3.** Synthesis of **3** and **4**α from the nickelation of [Tl]2[1-(1- -3- ,1- ,2- -*closo*-TlC2B9H10)-3,1,2 *closo*-TlC2B9H10]. NiPP = {Ni(dppe)}.

Earlier we noted a diastereoisomeric mixture resulting from the deboronationmetalation of 1,1- -bis(*o*-carborane) with the {Ni(dmpe)} fragment. However, with the {Ni(dppe)} fragment, additionally, isomerisation occurred in the case of a mono-metalated [3,1,2-(dppe)-NiC2B9-1,2-C2B10] species which transformed to [4,1,2-(dppe)-NiC2B9-1,2- C2B10] due to the stereo-electronic nature of the dppe. [8] Therefore, with double metalation using the{Ni(dppe)} fragment, the products could be diastereoisomers, stereospecific products and other isomers thereof. Although the 1H NMR spectra of **3** and **4**α are more complex than those of compounds **1** and **2**, there are resonances from phenyl protons and methylene bridge protons of the dppe. In the proton spectrum of **3**, there is also a C*H*cage resonance of integral two which appears at δ 1.84 ppm as a doublet *J* = 11.6 Hz, whilst in the proton spectrum of **4**α, there are two C*H*cage resonances, each of integral-1, one at δ 1.94 ppm appearing as a singlet and the other at δ 1.67 ppm appearing as a doublet *J* = 11.2 Hz. The doublets collapse to singlets in the 1H{31P} spectra. Notably the C*H*cage resonance of **3** appears at a lower frequency than that of **1** (δ 2.43 ppm) or **2** (δ 2.48 ppm). The two doublet C*H*cage resonances clearly indicate that both cages of compound **3** are nickellated and could be of 3,1,2-NiC2B9-3- ,1- ,2- -NiC2B9 architecture with reference to the proton spectra of compounds **1** and **2**. The doublet and singlet C*H*cage resonances indicate that nickelation occurred for both cages of compound **4**α where one cage could be unisomerised, i.e., the {3,1,2-NiC2B9} form and another cage could be isomerised to either a {4,1,2-NiC2B9} or a {2,1,8-NiC2B9} form. These inferences are further supported by the 31P NMR spectra

of compounds **3** and **4**α. The 31P{1H} NMR spectrum of compound **3** consists of two mutual doublets with integral ratio 1:1 at δ 51.6 and 39.2 ppm with coupling *J* = 19.1 Hz; on the contrary, the 31P{1H} NMR spectrum of **4**α consists of a singlet of integral-2 at δ 62.9 ppm and two mutual doublets, each of integral-1 at δ 49.4 and 47.5 ppm, with a coupling *J* = 16.6 Hz. This signifies the asymmetric nature of each metallacarborane cage but the symmetric nature of the whole molecule. The 11B{1H} spectrum of compound **3** reveals five resonances with the relative integrals 2:4:2:5:5 from high frequency to low frequency, a different pattern to that for compounds **1** or **2**, whilst the 11B{1H} NMR spectrum of **4**α consists of multiple overlapping resonances with a total integral of 18B and is different in pattern to that of compounds **1**, **2** and **3**, thus preventing the identification of the exact isomer present and thereby requiring single crystal XRD analysis.

The precise natures of **3** and **4**α were confirmed by crystallographic analysis. The racemic form of compound **3** is confirmed by a crystallographic *C*<sup>2</sup> axis passing through the mid-point of the C1-C1 bond, and thus requiring both cages to be of the same chirality. Figure 3 shows a perspective view of a single molecule of *rac*-[1-(1- -3- -(dppe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**3**). However, it is clear that for compound **4**α one of the cages has isomerised. A perspective view of a single molecule of [1-(2- -4- -(dppe)- 4- ,1- ,2- -*closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**4**α) is shown in Figure 4. For compound **3**, both cages are nickellated by {Ni(dppe)} fragments and are of 3,1,2-NiC2B9 architecture. As noted, the midpoint of the C1-C1 bond lies on a crystallographic 2-fold axis. This leads to two bulky {(dppe)NiC2B9} units, connected at C1 and C1- , pushing each other apart. There is a clear indication of internal crowding which is obvious from the orientation of the {NiP1P2} fragment in the NiC2B9 icosahedra. In **3** (for the non-primed cage), the θ (previously discussed) between the plane containing Ni3P1P2 and the plane through B6B8B10 is 60.72(9)◦, which deviates significantly from 90◦. Further evidence of steric congestion is demonstrated by the fact that the plane of the {NiP1P2} fragment deviates from perpendicularity to the plane through B5B6B11B12B9 vertices by 6.6(8)◦. Additionally, the longer Ni3–C1 distance compared to Ni3–C2 [2.336(2) *versus* 2.124(2) Å] again supports significant steric crowding in the molecule. The steric congestion in **3** would be slightly relieved if the 3,1,2-NiC2B9 cage would isomerise to either the 4,1,2-NiC2B9 form or the 2,1,8-NiC2B9 form. Indeed, this has been experimentally observed in **4**α, which is of 3,1,2-NiC2B9-4- ,1- ,2- -NiC2B9 form, where one of the cages has isomerised. The molecule of **4**α is partly disordered (two orientations for a phenyl ring), but the atomic connectivity is nonetheless clear. The internal crowding slightly reduces as, for compound **4**α (for 3,1,2-NiC2B9 cage), the θ between the Ni3P1P2 and B6B8B10 planes is 64.8(4)◦, closer to the idealised 90◦. However, in **4**α, the non-primed (3,1,2-NiC2B9 cage) still has significant internal crowding, as indicated by the longer distances Ni3-C1: 2.327(9) Å *versus* Ni2-C2: 2.129(9) Å. The relaxation of steric congestion in the (4- ,1- ,2- -NiC2B9 cage) of **4**α is reflected in the slightly shorter distance Ni4- -C1 is 2.167(8) Å. Furthermore the dihedral angle between the Ni4- P1- P2 and C1- B11- B12 planes for the primed cage is 86.1(4)◦. Thus, the metalation of doubly deboronated species from 1,1- -bis(*o*-carborane) with {Ni(dppe)} results in bis-nickelated products, and although in principle both *racemic* and *meso* products were anticipated, only stereospecific racemic-**3** and **4**α were observed. The meso form was not found. The rationale for the formation of the *rac* form is discussed below.

**Figure 3.** A perspective view of *rac*-[1-(1- -3- -(dppe)-3- ,1- ,2- -*closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**3**) (all phenyls and the –CH2–CH2– bridge of dppe are in wireframe for clarity). Atoms with dashed suffixes are generated by the symmetry operation 1 − x, y, 0.5 − z.

**Figure 4.** Molecular structure of [1-(2- -4- -(dppe)-4- ,1- ,2- -*closo*-NiC2B9H10)-3-(dppe)-3,1,2-*closo*-NiC2B9H10] (**4**α) (all phenyls and the –CH2-CH2- bridge of dppe are in wireframe for clarity).

#### *3.3. Dihydrogen Interaction in 3 and Isomerisation in 4α*

The reaction of the [7-(7- -7- ,8- -*nido*-C2B9H10)-7,8-*nido*-C2B9H10] <sup>4</sup><sup>−</sup> tetraanion with {Ni(dmpe)} fragments affords diastereomeric mixture products, i.e., *rac* (**1**) and *meso* (**2**) forms of 3,1,2-NiC2B9-3- ,1- ,2- -NiC2B9, whereas the same reaction with the more bulky {Ni(dppe)} fragment results in two isolable stereospecific products, i.e., *rac* 3,1,2-NiC2B9- 3- ,1- ,2- -NiC2B9 (**3**) and 3,1,2-NiC2B9-4- ,1- ,2- -NiC2B9, in compound **4**α. The isolated *rac*-**3** displays intramolecular dihydrogen bonding, as was found previously during the rationalisation of the stereospecific {CoCp} fragment metalation reaction with the [7-(7- - 7- ,8- -*nido*-C2B9H10)-7,8-*nido*-C2B9H10] <sup>4</sup><sup>−</sup> tetranion to form the only racemic form of the product 3,1,2-CoC2B9-3- ,1- ,2- -CoC2B9 [9]. The formation of intramolecular dihydrogen bonding associated with C*H* atoms of one cage and B*H* atoms of the other cage results from the relatively protonic and hydridic nature of the C*H* and B*H* atoms, respectively. 11B NMR analysis of individual vertices for 3,1,2-MC2B9 compounds established that the most hydridic B*H* atoms are H5 and H6 [18]. In the case of *rac*-**3**, the orientation of two {(dppe)NiC2B9} cages enables two sets of intramolecular dihydrogen bonds, CH2··· BH6i 2.07(3) Å (where i = 1 − x, y, 0.5 − z) and its symmetry equivalent, CH2i ··· BH6. Notably, the hypothetical analogous meso form could only allow for one set of such intramolecular dihydrogen bonding, whatever the rotameric arrangement, thereby rendering the *meso* isomer less favourable.

Crowding between the bulky {(dppe)(3,1,2-NiC2B9)} substituent on C1 and the dppe ligand on Ni3 of an unisomerised 3,1,2-NiC2B9-3- ,1- ,2- -NiC2B9 species is likely to be the major cause of the isomerisation observed in compound **4**α. Isomerisation moves the {(dppe)(3,1,2-NiC2B9)} substituent down to the lower pentagonal belt further away from the {Ni(dppe)2} on the other cage. Indeed, we have already observed steric crowding in the unisomerised compound **3**. In regards to the {3- ,1- ,2- -NiC2B9} to {4- ,1- ,2- -NiC2B9} isomerisation, this can likely be related to electronic factors, specifically the electron withdrawing nature of the dppe ligand. A similar 3,1,2- to 4,1,2- isomerisation of mono-metalated nickelacarboranes with dppe and PPh2Me ligands has been previously established [8].

#### **4. Conclusions**

Four new bis(nickelated) species are documented from the metalation of doubly deboronated 1,1- -bis(*o*-carborane), and their identity was confirmed by both spectroscopic and crystallographic means. The metalation of the {Ni(dmpe)} fragment with the Tl4-salt gives diastereoisomeric products with the unusual crowded architecture of 3,1,2-NiC2B9- 3- ,1- ,2- -NiC2B9 as racemic and meso forms. In contrast, metalation with the {Ni(dppe)} fragment results in the 3,1,2-NiC2B9-4- ,1- ,2- -NiC2B9 species, a stereospecific racemic product. The racemic product in the latter case shows intramolecular dihydrogen bonding, hence explaining the stereospecific reaction, whereas the stereo-electronic nature of the bis(phosphine) ligand influences the formation of the isomerised 3,1,2-NiC2B9-4- ,1- ,2- - NiC2B9 species.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/2073-4 352/11/1/16/s1, NMR spectra of all new compounds are available online along with the crystallographic data, free of charge. Crystallographic information for all compounds here has been deposited in the Cambridge Crystallographic Data Centre as supplementary publications nos. CCDC 2048463, 2048465, 2,048,466 and 2,048,464 (compounds **1**, **2**, **3** and **4**α).

**Author Contributions:** Syntheses and original draft preparation, and editing, D.M.; final structure analysis and manuscript editing, G.M.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** D.M. is grateful to the Heriot-Watt University for financial assistance; a James-Watt Ph.D. studentship was awarded to D.M. from 2013 to 2016. We thank EPSRC for funding the Bruker X8Apex2 diffractometer.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** This work is taken from the Ph.D. thesis (2016) of D.M., supervised by Alan J. Welch at Heriot-Watt University. We thank EPSRC for funding the Bruker X8Apex2 diffractometer and the U.K. National Crystallography Service for data collection of compound **4**α.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **"Free of Base" Sulfa-Michael Addition for Novel** *o-***Carboranyl-DL-Cysteine Synthesis** †

**Julia Laskova 1,\*, Irina Kosenko 1, Ivan Ananyev 1, Marina Stogniy 1,2, Igor Sivaev <sup>1</sup> and Vladimir Bregadze <sup>1</sup>**


Received: 2 November 2020; Accepted: 7 December 2020; Published: 11 December 2020

**Abstract:** The sulfa-Michael addition reaction was applied for the two-step synthesis of *o-*carboranyl cysteine 1-HOOCCH(NH2)CH2S-1,2-C2B10H11 from the trimethylammonium salt of 1-mercapto-*o-*carborane and methyl 2-acetamidoacrylate. To avoid the decapitation of o-carborane into its *nido*-form, the "free of base" method under mild conditions in a system of two immiscible solvents toluene-H2O was developed. The replacement of H2O by 2H2O resulted in carboranyl-cysteine containing a deuterium label at the α-position of the amino acid 1-HOOCCD(NH2)CH2S-1,2-C2B10H11. The structure of the protected *o-*carboranyl cysteine was determined by single-crystal X-ray diffraction. The obtained compounds can be considered as potential agents for the Boron Neutron Capture Therapy of cancer.

**Keywords:** boron chemistry; *o-*carborane; sulfa-Michael addition reaction; cysteine; boron neutron capture therapy; *o-*carborane decapitation; labeled compound

#### **1. Introduction**

Cancer remains one of the major health issues and the second leading cause of death worldwide [1]. Radiation therapy has a central role in the management of malignant brain tumors, especially for the most aggressive ones [2–4]. First introduced in 1936 by Locher [5], the Boron Neutron Capture Therapy model (BNCT) is a promising type of radiation therapy for cancer that has the potential to be an important treatment for numerous types of tumors, including for those lying in areas that are difficult to access for surgery intervention, such as high-grade gliomas and metastatic brain tumors [6–8] Currently, only two low molecular weight boron-containing compounds, sodium mercapto-undecahydro-*closo*-dodecaborate (BSH, developed in about 1965) and a water-soluble fructose complex of borylphenylalanine (BPA, discovered in 1958), have been approved and found clinical use in BNCT [9–11].

Recent advances in the development of potential agents for BNCT treatment include various boron-containing bioconjugates [12]; there are many with natural and unnatural amino acids among them [13]. It is believed that amino acid conjugates are preferentially taken up by rapidly growing tumor cells [14]. Polyhedral carboranes as a boron-rich compounds are considered to have potential for usage in BNCT [15]. In this regard, the chemistry of dicarbaboranes [C2B10H12] and their derivatives has been well investigated [16]. The synthesis of *o-*carboranes containing unnatural ω-mercapto-amino acids with side-chain lengths ranging from 4 to 6 methylene units **1(a–c)** [17], L-*o-*carboranyl-alanine **2** [18] as well as *o-*carboranyl derivatives of (S)-Asparagine **3a** and (S)-Glutamine **3b** [19], phenylalanine **4** [20] and (S)-*m-*carboranyl-homocysteine sulfoxide **5 [21]**, have been published. Our current interest was inspired by the recent synthesis [22] of *m-*carboranyl-cysteine **6** and its biological evaluation as an agent for Boron Neutron Capture Therapy [23,24] (Figure 1). Therefore, we decided to synthesize its analogue based on readily available 1-mercapto-*o-*carborane [25,26].

**Figure 1.** *o-* and *m-* carborane–amino acid conjugates.

Here, we report the synthesis of *o-*carboranyl-cysteine via the sulfa-Michael addition reaction of an easily available salt of 1-mercapto-*o-*carborane and methyl 2-acetamidoacrylate (MAA) using a two-phase system.

#### **2. Materials and Methods**

#### *2.1. General*

[1-S-1,2-C2B10H11](Me3NH) (**7**) was prepared according to the literature procedure [27]. Methyl 2-acetamidoacrylate (Sigma Aldrich Chemie GmbH, NJ, USA) was used without purification. 2H2O was received from Carl Roth GmbH. The reaction proceedings were monitored via thin-layer chromatograms (Merck F254 silica gel on aluminums plates). Boron compounds were visualized with PdCl2 stain solution, which, upon heating, gave dark brown spots. Purifications were carried out using column chromatography with Silica gel 60 0.060–0.200 mm (Acros Organics, NJ, USA). 1H, 13C, and 11B NMR spectra were recorded at 400.13, 100.61, and 128.38 MHz, respectively, on a BRUKER-Avance-400 spectrometer. Tetramethylsilane and BF3 <sup>×</sup> Et2O were used as standards for 1H, 13C NMR, and 11B NMR, respectively. All the chemical shifts are reported in ppm (δ) relative to external standards. IR spectra were recorded on an IR Prestige-21 (SHIMADZU, Kyoto, Japan) instrument. High-resolution mass spectra (HRMS) were measured on a Bruker micrOTOF II instrument using electrospray ionization (ESI), with a mass range from *m*/*z* 50 to *m*/*z* 3000; external or internal calibration was carried out with ESI Tuning Mix, Agilent. Spectra can be found in the Supplementary Materials.

#### *2.2. Synthesis of 1-CH3O(O)C(CH3(O)CHN)CHCH2S-1,2-C2B10H11 (***8***)*

To a solution of **7** (0.5 g, 2.1 mmol) in toluene (30 mL) methyl 2-acetamidoacrylate (0.3 g, 2.1 mmol) and H2O (30 mL) were added; the resulting system was vigorously stirred under reflux for 24 h. Then, the mixture was cooled to r.t. and the toluene layer was separated, washed with H2O (2 × 20), dried over Na2SO4, and evaporated. The product was purified by silica gel column chromatography using Et2O as an eluent and vacuum-dried to give a light yellow solid. Yield: 0.46 g (68%). 1H NMR (Chloroform-*d*) δ = 6.29 (d, *J* = 7.7 Hz, 1H, NH), 4.89 (td, *J* = 7.2, 4.5 Hz, 1H, α-CH), 3.99 (broad s, 1H, carb-CH), 3.82 (s, 3H, COOCH3), 3.51 (dd, *J* = 13.4, 4.6 Hz, 1H, C*H*2CH), 3.18 (dd, *J* = 13.4,

6.9 Hz, 1H, C*H*2CH), 2.06 (s, 3H, NHCOC*H*3), 3.0–1.5 (broad, 10H, BH); 11B NMR (Chloroform-*d*) δ = −1.5 (d, *J* = 150 Hz, 1B), −4.7 (d, *J* = 146 Hz, 1B), −8.8 (d, *J* = 154 Hz, 4B), −12.2 (d, *J* = 162 Hz, 4B); 13C NMR (Chloroform-*d*) δ = 170.21, 170.17 (CO), 73.9, 67.9 (C-carb), 53.2 (OCH3), 51.2 (α-CH), 39.4 (*C*H2CH), 23.1 (CO*C*H3). ESI-MS, *m*/*z*, C8H21B10NO3S calcd. 320.2322, found 320.2322 ([M+H]+). IR-FT (ν, cm<sup>−</sup>1): 3371(NH), 3060 (CH-carb); 2948, 2926, 2852 (broad CH); 2607, 2584, 2557 (BH); 1736, 16177 (CO).

#### *2.3. Synthesis of 1-HOOC(NH2)CHCH2S-1,2-C2B10H11* × *HCl (***9***)*

Water (2 mL) and concentrated HCl (10 mL) were added to a solution of **8** (0.1 g, 0.3 mmol) in glacial acetic acid (10 mL). The resulting mixture was heated at 70 ◦C for 40 h., cooled to r.t., and evaporated. The residue was suspended in 5 mL of water, formed solid was filtered, washed with water (2 × 5 mL) and vacuum dried to give **9**. Light yellow solid (0.093 mg, 70%). 1H NMR (Methanol-*d*4) δ = 4.89 (s, 1H, carb-CH), 4.09 (m, 1H, α-CH), 3.64 (m, 1H, C*H*2CH), 3.43 (m, 1H, C*H*2CH), 3.0–1.5 (broad, 10H, BH); 11B NMR (Methanol-*d*4) <sup>δ</sup> <sup>=</sup> <sup>−</sup>1.8 (d, *<sup>J</sup>* <sup>=</sup> 151 Hz, 1B), <sup>−</sup>5.1 (d, *<sup>J</sup>* <sup>=</sup> 148 Hz, 1B), <sup>−</sup>9.2 (d, *<sup>J</sup>* <sup>=</sup> 148 Hz, 4B), <sup>−</sup>12.3 (d, *J* = 172 Hz, 4B). 13C NMR (Methanol-*d*4) <sup>δ</sup> 172.0 (COOH), 74.8, 68.4 (C-carb), 51.60 (α-CH), 38.3 (*C*H2CH). ESI-MS, *m*/z, C5H17B10NO2S calcd. 264.2059, found 264.2061. IR-FT (ν, cm<sup>−</sup>1): 3399 (broad NH+), 3058 (CH- carb); 2923, 2854 (CH); 2580 (broad BH), 1728 (CO).

#### *2.4. Synthesis of 1-CH3O(O)C(CH3(O)CHN)CDCH2S-1,2-C2B10H11 (***10***)*

Under an argon atmosphere, methyl 2-acetamidoacrylate (0.06 g, 0.4 mmol) and 2H2O (5 mL) were added to a solution of **7** (0.1 g, 0.4 mmol) in toluene (15 mL); the resulting two-phase system was vigorously stirred under reflux for 30 h. Purification was held in the same manner as for compound **8**. Light yellow solid (90 mg, 66%). 1H NMR (Chloroform-*d*) δ = 6.31 (s, 1H, NH), 4.00 (broad s, 1H, CH-carb), 3.82 (s, 3H, COOCH3), 3.50 (d, *J* = 13.4 Hz, 1H, CH2CD ), 3.17 (d, *J* = 13.4 Hz, 1H, C*H*2CD), 2.06 (s, 3H, NHCOC*H*3), 3.0–1.5 (broad, 10H, BH); 11B NMR (Chloroform-*d*) δ = −1.5 (d, *J* = 151 Hz, 1B), −4.8 (d, *J* = 150 Hz, 1B), −8.9 (d, *J* = 151 Hz, 4B), −12.4 (d, *J* = 176 Hz, 4B); 13C NMR (Chloroform-*d*) δ = 170.23, 170.17 (CO), 73.9, 67.9 (C-carb), 53.2 (OCH3), 51.0 (t, α-CD), 39.3 (CH2), 23.1 (CO*C*H3). ESI-MS, *m*/*z*, C8H20DB10NO3S calcd. 321.2385, found 321.2399. IR-FT (ν, cm<sup>−</sup>1): 3370 (NH), 3061 (CH- carb); 2953 (CH); 2608, 2586, 2557 (BH), 1733, 1674 (CO).

#### *2.5. Synthesis of 1-HOOC(NH2)CDCH2S-1,2-C2B10H11* × *HCl (***11***)*

Water (2 mL) and concentrated HCl (5 mL) were added to a solution of **10** (0.05 g, 0.16 mmol) in glacial acetic acid (5 mL). The resulting mixture was heated at 70 ◦C for 40 h., cooled to r.t., and evaporated. The residue was suspended in 5 mL of water, formed solid was filtered, washed with water (2 mL) and vacuum dried to yield **11**: 38 mg (81%). 1H NMR (Methanol-*d*4) δ = 4.88 (broad s, 1H, CH-carb), 4.27–4.20 (α-CH, m, 0,1H), 3.64 (d, *J* = 13.9 Hz, 1H, C*H*2CD), 3.44 (d, *J* = 13.9 Hz, 1H, C*H*2CD), 3.0–1.5 (broad, 10H, BH). 11B NMR (Methanol-*d*4) <sup>δ</sup> <sup>=</sup> -1.9 (d, *<sup>J</sup>* <sup>=</sup> 149 Hz, 1B), <sup>−</sup>4.9 (d, *<sup>J</sup>* <sup>=</sup> 140 Hz, 1B), <sup>−</sup>9.0 (d, *<sup>J</sup>* <sup>=</sup> 143 Hz, 4B), <sup>−</sup>12.3 (d, *<sup>J</sup>* <sup>=</sup> 175 Hz, 4B). 13C NMR (Methanol-*d*4) <sup>δ</sup> <sup>=</sup> 168.0 (CO), 73.6 (C-carb), 68.4 (C-carb), 51.8 (α-C), 36.0 (CH2). ESI-MS, *m*/z, C5H17B10NO2S calcd. 265.2122, found 265.2127.

#### *2.6. Synthesis of 1-CH3O(O)C(CH3(O)CHN)CHCH2S-2-D-1,2-C2B10H10 (***12***)*

Under an argon atmosphere, methyl 2-acetamidoacrylate (0.06 g, 0.4 mmol), 2H2O (5 mL), and anhydrous K2CO3 (0.055 g, 0.4 mmol) were added to a solution of **7** (0.1 g, 0.4 mmol) in toluene (15 mL); the resulting system was vigorously stirred under reflux for 10 h. The purification was held in the same manner as for compound **8**. Light yellow solid (13 mg, 10%). 1H NMR (Chloroform-*d*) δ = 6.20 (d, *J* = 7.7 Hz, 1H, NH), 4.90 (td, *J* = 7.2, 4.7 Hz, 1H, α-CH), 4.00 (undetectable, s, 0H, carb-CD), 3.83 (s, 3H, COOCH3), 3.53 (dd, *J* = 13.4, 4.7 Hz, 1H, C*H*2CH), 3.19 (dd, *J* = 13.4, 6.9 Hz, 1H, C*H*2CH), 2.07 (s, 3H, NHCOC*H*3), 3.0–1.5 (broad, 10H, BH); 11B NMR (Chloroform-*d*) <sup>δ</sup> = <sup>−</sup>1.5 (d, *<sup>J</sup>* <sup>=</sup> 152 Hz, 1B), <sup>−</sup>4.8 (d, *J* = 147 Hz, 1B), <sup>−</sup>8.9 (d, *J* = 152 Hz, 4B), <sup>−</sup>12.5 (d, *J* = 173 Hz, 4B); 13C NMR (Chloroform-*d*) δ = 170.16,170.14 (CO), 73.8 (CH-carb), 68.5–66.8 (t, CD-carb), 53.2 (OCH3), 51.2

(α-CH), 39.4 (CH2), 23.1 (CO*C*H3). ESI-MS, *m*/*z*, C8H20DB10NO3S calcd. 321.2385, found 321.2384. IR-FT (ν, cm<sup>−</sup>1): 3372(NH), 2923, 2851 (CH); 2607, 2584, 2557 (BH); 2289 (CD-carb); 1736, 1677 (CO).

#### *2.7. X-ray Di*ff*raction*

Crystals of **8** are triclinic, space group P-1, at 120K: a = 7.6513(5), b = 10.5669(7), c = 11.5392(7), <sup>α</sup>= 68.378(1)◦, <sup>β</sup> = 85.686(1)◦, <sup>γ</sup> = 72.663(1)◦, V = 827.27(9) Å3, Z = 2 (Z' = 1), dcalc = 1.282 g·cm−3, F(000) = 332. Intensities of 11,058 reflections were measured with a Bruker SMART APEX 2 Duo CCD diffractometer [λ(MoKα) = 0.71072Å, ω-scans, 2θ < 60◦] and 4824 independent reflections [Rint = 0.0324] were used in further refinement. The structure was solved by the direct method and refined by the full-matrix least-squares technique against F2 in the anisotropic-isotropic approximation. The hydrogen atoms were found in the difference Fourier synthesis and refined in the isotropic approximation within the riding model. For **8**, the refinement converged to wR2 = 0.1167 and GOF = 0.781 for all independent reflections (R1 = 0.0371 was calculated for 3661 observed reflections with I > 2σ(I)). All the calculations were performed using SHELX2018 [28]. The CCDC 2034871 contains the supplementary crystallographic data for **8**. These data can be obtained free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html (or from the CCDC, 12 Union Road, Cambridge, CB21EZ, UK; or deposit@ccdc.cam.ac.uk).

Computational details: All the calculations were conducted in the Gaussian09 program (rev. D01) [28]. The geometry optimization procedures were performed using standard criteria on displacements and forces. The DFT optimization were performed using the PBE0 functional [29,30] and the 6-311++G(d,p) basis set (ultrafine integration grids were used). The non-specific solvation was modelled using the self-consistent reaction field approach (PCM model, ε = 72). The influence of specific solvation on the geometry of **8** was accounted for by the optimization of a central molecule in clusters; two models were used: (1) the trimer of molecules from crystal of **8** with fixed coordinates and normalized X-H bond lengths for lateral molecules and (2) the shell of molecules from crystal of **8** with fixed coordinates and normalized X-H bond lengths. The Quantum theory of Atoms in Molecules surface integrals for the trimer were calculated using the AIMAll program [31]. The shell in the second model was generated using the criteria of at least one geometrical contact (within the sum of van der Waals radii plus 5 Å) between the surrounding molecules and the central molecule. The shell was described by the ONIOM approach (PBE0/6-311++G(d,p):PBE0/3-21G), and only the internal layer (the central molecule) was optimized. The Hessian calculations for all the optimized structures revealed their correspondence to energy minimums.

#### **3. Results**

#### *3.1. o-Carboranyl-Cysteine Synthesis*

The reaction conditions for the sulfa-Michael addition of thiophenols to methyl 2-acetamidoacrylate are well-studied [32]. It has been shown that reactions proceed in toluene or THF with catalytic quantity of the basic salt K2CO3 in the presence of solid-liquid phase-transfer catalyst. On the one hand, it was reasonable to apply conditions studied for thiophenols to the mercapto derivative of *o-*carborane; however, prolonged reaction with a base may cause of a side reaction due to the well-known property of *o-*carborane to undergo deboronation in the presence of bases, even mild ones, to yield *nido-*[7,9-C2B9H12] [33–39]. The deboronation of *o*-C2B10H12 proceeds fairly easily, whereas *m*-C2B10H12, which was used for the synthesis of **6**, is much more stable under the same conditions. A detailed study of such differences has been presented [40]. On the other hand, more common work up of *o-*carboranyl thiols includes the conversion of the formed SH-derivative to the more stable triethylammonium [41,42] or trimethylammonium salt [27] **7** (Scheme 1). The usage of the trimethylammonium salt **7** let us to avoid the deprotonation stage during the reaction process to keep the *o-*carborane structure from degradation under the action of a base. The synthetic route developed

for the preparation of the cysteine derivative of *o-*carborane conjugate with cysteine is outlined in Scheme 1.

**Scheme 1.** Synthesis of *o-*carboranyl cysteine.

Compared to mercapto-*o*-carborane, *o*-carboranyl thiolate **7** lacks sufficient acidic hydrogen, which is necessary for the reaction to proceed. We found that reaction of **7** with methyl 2-acetamidoacrylate in a biphasic toluene-H2O system under reflux for 24 h without base or any additional catalyst results in the formation of sulfa-Michael addition reaction product **8**.

The structure of **8** was confirmed by 1H, 11B, 13C NMR, and IR spectroscopy, high-resolution mass spectrometry, and single-crystal X-ray diffraction study. The 1H NMR spectrum of **8** in CDCl3 shows two singlets at 2.06 and 3.83 due to the acetamide and the ester groups, a doublet at 6.17 ppm from the NH-group, a multiplet at 4.90 ppm attributed to the proton bonded to the α-carbon, two doublet of doublets at 3.51 and 3.18 ppm assigned to the diastereotopic methylene protons, and a broad singlet of carborane C-H group at 3.99 ppm. The 13C-NMR spectrum contains signals of two carbonyl groups at 170.21 and 170.17 ppm., two cluster carbons at 73.9 and 67.9 ppm, two methyls of ester at 53.2 and acetamide at 23.1 ppm., as well as the CH and CH2 groups at 51.2 and 39.4 ppm. The 11B and 11B{1H} NMR spectra are in agreement with the C-mono-substituted o-carborane structure.

The solid-state structure of **8** was determined by single-crystal X-ray diffraction. The molecule of **8** is crystallized as the racemate in the centrosymmetric P-1 space group and contains a stereocenter at the C4 atom (Figure 2).

**Figure 2.** General view of the molecule **8** in crystal. Non-hydrogen atoms are given as probability ellipsoids of atomic displacements (*p* = 0.5).

While the main structural features of **8** are expected for this class of compounds (the C1-C2, S1-C1, and S1-C3 bond lengths equal 1.669(2), 1.791(1), and 1.826(1) Å, respectively; the sum of valence angles at the amide-type N1 atom equals 358.9◦), the rotation of the substituent at the C1 atom with respect to the carborane cage can, however, hardly be rationalized by common intramolecular structural effects. For instance, the lone electron pairs of the S1 atom are nearly periplanar to the C3-H and C3-C4 bonds that contradicts the preferred geometry of expected LP(S)→σ\* stereoelectronic interactions (Figure 3a). Indeed, the Cambridge Structural Database search reveals the predominance of staggered conformations of the C-Ssp3-Csp3-Csp3 fragments (Figure 3b), whereas the corresponding torsion angle equals 118.1(2)◦ in **8**.

**Figure 3.** The conformation of the C1-S1-C3-C4 fragment in **8** (LP–positions of lone electronic pairs, (**a**)) and the distribution of corresponding torsion angles in the C-Ssp3-Csp3-Csp3 fragments retrieved from the Cambridge Structural Database (**b**).

Based on the geometrical analysis of the crystal packing of **8**, it has been found that the substituent conformation could be stabilized by the environment effects. Indeed, there are several shortened intermolecular contacts formed by atoms of the C1 substituent which can be attributed to rather strong interactions – NH ... O H-bonds (N1 ... O1 2.933(2) Å, NHO 161.3◦ with normalized N-H bond length) and O ... π interactions (O1 ... O3 3.020(2) Å), both bounding molecules into infinite chains (Figure 4). The weak BH ... HC and CH ... O interactions are the only meaningful interactions between these chains.

In order to reveal the influence of environment effects on the conformation of **8**, the DFT calculations were additionally performed. Indeed, both the isolated molecule optimization and the optimization of **8** accounting for non-specific solvation effects (the SCRF-PCM model, relative dielectric permittivity of 72) produced structures being significantly different from those observed in the crystal: the root-mean-square deviations of the best overlap for non-hydrogen atoms are 0.495 and 0.446 Å, respectively. Surprisingly, the explicit DFT treatment of three neighboring molecules from a chain (see Figure 4) forming the mentioned shortened intramolecular contacts did not lead to any pronounced changes: the rmsd value for the central molecule in the calculated trimer equals 0.320 Å. Note that the total energy of mentioned N-H ... O H-bonds and O ... <sup>π</sup> interactions exceeds 24 kcal·mol−<sup>1</sup> according to the QTAIM electron density analysis [43] carried out for the trimer. Our best result was achieved by the ONIOM calculation, which considers all neighboring molecules at the DFT level (the rmsd value equals to 0.168 Å, Figure 5). Thus, despite the relatively large strength of the H-bonds and O ... π interactions, the conformation of **8** in crystal depends heavily on the peculiarities of other, weaker intermolecular interactions such as BH ... HC and CH ... O.

**Figure 4.** A fragment of the infinite chain formed in crystal of **8** along the *a* axis. Non-hydrogen atoms are given as probability ellipsoids of atomic displacements (*p* = 0.5).

**Figure 5.** The best root-mean-square overlap for non-hydrogen atoms between crystal (solid lines) and ONIOM-modelled (dashed lines) conformations of **8**.

The treatment of **8** with a mixture of glacial acetic and hydrochloric acids under heating gave the title S-substituted cysteine **9** in the form of hydrochloride with a good yield (Scheme 1). The synthesized amino acid was characterized by 1H, 11B, 13C NMR spectroscopy, IR spectroscopy, and high-resolution mass spectrometry. A singlet at 4.89 ppm corresponding to the C-H from the o-carborane was observed in the 1H NMR; other chemical shifts agreeing with amino acid structure were also observed.

#### *3.2. The Sulfa-Michel Addition Mechanism Investigation for o-Carboranyl Cysteine and Synthesis of Deuterium Labeled Compounds*

The investigation of the reaction conditions for the synthesis of **8** revealed a crucial role of the choice of solvent. Thus, the formation of only an insignificant amount of **8** was observed when the reaction proceeded in homogenous systems EtOH-H2O or THF-H2O, whereas in absolute EtOH no product was detected at all. The best result was achieved when the toluene-H2O system was used without any base. Since the initial salt **7** did not have an acid proton, as 1-mercapto-o-carborane has, this is why we assumed that α-proton comes into intermediate **8** from water. To gain some insight into the process, we carried out the "free of base" reaction with 2H2O, which required strictly anhydrous conditions (Scheme 2).

**Scheme 2.** Synthesis of the labeled with deuterium *o-*carboranyl conjugate.

We found that the two-phase system reaction proceeds in the way we proposed. Compound **10** with deuterium at the α-position of amino acid was isolated and characterized by NMR, IR spectroscopy, and mass spectrometry. The reaction time for the synthesis of **10** in the toluene-2H2O system increased as compared with those in the toluene-H2O system. This difference in reaction time may be attributed to the presence of a deuterium isotope effect. The yield of reaction, which was detected, remains over 65%. The comparison of the 1H NMR and 13C spectra of **8** and **10** is displayed below (Figure 6).

The 1H NMR spectrum of **10** in CDCl3 does not contain the multiplet at 4.90 ppm, which indicates the absence of a proton at the α-position of the amino acid, whereas in the 13C-NMR spectrum the triplet due to the C2H group appears at 51.0 ppm, indicating the substitution of H for 2H. In addition, the doublet of the NH group (6.17 ppm) and two doublets of doublets from the methylene CH2 protons, which were detected in the 1H NMR spectrum of **8**, collapsed to a singlet and two doublets in the spectrum of [α-2H]carboranyl-cysteine **10,** respectively, due to the absence of an adjacent proton.

The subsequent acid hydrolysis of **10** resulted in the o-carboranyl-cysteine **11,** labeled at the α-position of the amino acid (Scheme 3).

The presence of 2H and its position were determined by high-resolution mass spectrometry and 1H NMR spectroscopy. The 1H NMR spectrum showed the presence of an unlabeled compound in an amount of less than 10% (Figure 7).

**Figure 6.** Comparison of 1H NMR and fragments of the 13C NMR spectra of **8** and **10** in CDCl3.

**Scheme 3.** *O*-carboranyl-DL-[α-2H]-cysteine synthesis.

**Figure 7.** 1H NMR spectrum of **11** in methanol-d4.

The method proposed above can be considered as a preparative one for the synthesis of labeled o-carboranyl-DL-[α-2H]-cysteine, which could be useful for study of the metabolism of o-carboranyl-DL-cysteine itself, as well as the metabolism of boron-containing proteins on its base as potential BNCT agents [44–46].

It was mentioned that the o-carborane system is sensitive to the presence of a base. To investigate the effect of water-soluble basic salt toward the reaction that proceeds in a two-phase system, we added K2CO3 to the reaction system. We found that the addition of a basic salt accelerates the reaction, which is complete in 8 h, but resulted in an only 17% yield of **8**, while the main product is *nido*-carborane derivatives, as found by 11B NMR. The solvent change from H2O to 2H2O along with the use of anhydrous K2CO3 and a Schlenk technique under an atmosphere of argon increases the reaction time from 8 h to 10 h due to the isotope effect, and the yield of product falls from 17% to 10%. Interestingly, the structure of the resulting product was found to be different from **8** and **10**. According to the NMR spectroscopy data, deuterium from 2H2O took up a place of the 1-CH-*o-*carborane proton, whereas proton appeared at the α-position of the amino acid (Scheme 4, Figure 8). Compound **12** was isolated and characterized, and its structure and composition were confirmed by 1H, 11B, 13C NMR, and IR spectroscopy and high-resolution mass spectrometry.

**Scheme 4.** The sulfa-Michael addition reaction in the presence of the basic salt.

**Figure 8.** 1H and a fragment of 13C NMR spectra of **12** in CDCl3.

#### **4. Discussion**

To summarize the obtained results, we assumed the mechanism of reaction proceeding (Scheme 5).

**Scheme 5.** Proposed mechanism of the reaction of trimethylammonium salt of 1-mercapto-*o*-carborane with methyl 2-acetamidoacrylate.

The reaction realizes according to a classical Michael addition reaction, where thiolate **7** acts as a nucleophile without additional deprotonation. In the first step, nucleophile reacts with the electrophilic alkene to form **A** in a conjugate addition reaction. In the next stage, the deuterium abstraction from solvent by the enolate **A** forms the final conjugate and a 2HO<sup>−</sup> anion from a water molecule. 2HO<sup>−</sup> anion, being a base, may cause a deboronation process, however its cooperation with a cation forms a water-soluble trimethylammonium hydroxide. When the reaction proceeds in the toluene-2H2O system, trimethylammonium hydroxide eliminates from the toluene medium and under the reaction conditions it decomposes in an aqueous medium with the formation of trimethylamine and water. Thus, the final stage process may be described as proton abstraction from cation that results in the deactivation of a base. The proposed mechanism is also supported by the fact that an increase in the reaction time does not have a significant effect on the yield of **8**. The reaction between **7** and methyl 2-acetamidoacrylate in EtOH does not give the desired product. According to the proposed mechanism, in this case the proton abstraction from solvent by the enolate **A** forms an EtO− anion, which is not eliminated from the reaction system. The EtO− anion being formed may undergo the Michael addition reaction, as well as **7** and/or deboronate *o-*carborane structure. The explanation

above can be applied to a reaction in a system of two miscible liquids, such as THF-H2O. The formed hydroxide anion remains in one system with the reactants to yield mainly a *nido-*carborane product.

#### **5. Conclusions**

Two possibilities of the sulfa-Michael addition reaction in the synthesis of *o*-carboranyl-DL-cysteine were investigated. The reaction proceeding between the trimethylammonium salt of 1-mercapto-*o-*carborane and methyl 2-acetamidoacrylate in the two-phase system, toluene-H2O with base, as expected, demonstrates a low yield. Using the "free of base" method under the mild conditions in the two-phase system, the side processes were minimized and after the deprotection 1-HOOCCH(NH2)CH2S-1,2-C2B10H11 was obtained in a good total yield. The developed "free of base" method was applied for the preparation of *o-*carboranyl-DL-cysteine labeled with deuterium at the α-position of amino acid using cheap and easily available 2H2O as a deuterium source. The obtained compounds can be considered as potential agents for the Boron Neutron Capture Therapy of cancer as well as for protein metabolism studies.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4352/10/12/1133/s1.

**Author Contributions:** Synthesis and writing—original draft preparation, J.L.; synthesis, M.S.; NMR research analysis, I.K.; X-ray diffraction study, I.A.; supervision, V.B. and I.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the Russian Science Foundation (№ 19-72-30005).

**Acknowledgments:** The NMR spectroscopy studies and X-ray diffraction experiment were performed using equipment of the Center for Molecular Structure Studies at A.N. Nesmeyanov Institute of Organoelement Compounds operating with financial support of the Ministry of Science and Higher Education of the Russian Federation.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Reactions of Experimentally Known** *Closo***-C2B8H10 with Bases. A Computational Study**

#### **Josef Holub 1, Jindˇrich Fanfrlík 2, Michael L. McKee <sup>3</sup> and Drahomír Hnyk 1,\***


Received: 19 August 2020; Accepted: 29 September 2020; Published: 3 October 2020

**Abstract:** On the basis of the direct transformations of *closo*-1,2-C2B8H10 with OH(−) and NH3 to *arachno*-1,6,9-OC2B8H13(−) and *arachno*-1,6,9-NC2B8H13, respectively, which were experimentally observed, the DFT computational protocol was used to examine the corresponding reaction pathways. This work is thus a computational attempt to describe the formations of 11-vertex *arachno* clusters that are formally derived from the hypothetical *closo*-B13H13(2−) . Moreover, such a protocol successfully described the formation of *arachno*-4,5-C2B6H11(−) as the very final product of the first reaction. Analogous experimental transformations of *closo*-1,6-C2B8H10 and *closo*-1,10-C2B8H10, although attempted, were not successful. However, their transformations were explored through computations.

**Keywords:** carboranes; DFT; reaction pathways

#### **1. Introduction**

Polyhedral borane and heteroborane clusters are known for the presence of delocalized electron-deficient bonding [1,2] and characterized by forming three-center, two-electron (3c-2e) bonds. This bonding is quite different from organic chemistry that is dominated by classical two-center two-electron (2c-2e) bonds. The trigonal faces of boranes and carboranes are assembled to create three-dimensional shapes such as icosahedron and bicapped-square antiprism [2] appearing in *closo* systems. The preparation and subsequent reactivity of these clusters have been extensively studied by experiments [1,2]. In particular, the 12-vertex icosahedral *closo* clusters have been one of the main targets. In contrast to well-understood reaction mechanisms in organic chemistry, those in boron cluster chemistry can be very complex because there are very small energy differences between many intermediates and transition states. On that basis, the reaction of boron hydrides may involve many competing pathways [3]. For that reason, relatively little progress has so far been made in the understanding of the reaction mechanisms of boron hydrides and carboranes of various molecular shapes [4–6]. To our knowledge, the reaction pathways associated with ten-vertex *closo* carboranes, for instance *closo*-1,2-C2B8H10 (see Scheme 1), have not yet been explored.

**Scheme 1.** Molecular diagrams of *closo*-1,2-C2B8H10, *arachno*-1,6,9-OC2B8H13(−) (**oP1-O**), and *arachno*-1,6,9-NC2B8H13 (**oP1-N**) and the corresponding atomic numberings.

The C2B8H10 molecular shape of a bicapped-square antiprism exists in seven positional isomers, only three of which (1,2-, 1,6- and 1,10-isomers) are known experimentally [2]. The most stable one is the 1,10-isomer and the least stable one is the 2,3-isomer [7]. The 1,6-isomer has recently been examined experimentally [8]. In addition, a new synthetic pathway for the preparation of the 1,2-isomer has also been outlined, including some mechanistic considerations [9]. Mutual isomerizations of all the possible *closo*-C2B8H10 have been studied computationally [10]. As to the observed reactivity of the 1,2-isomer, its direct *closo* to *arachno* transformations have been experimentally observed [11,12], with the resulting *arachno* structural motif being based on the hypothetical *closo*-B13H13(2−) as also seen from the corresponding atomic numberings (see also Scheme 1) [13]. Since no computational work has been reported in the area of the reactions of 10-vertex *closo* carboranes, we have undertaken a computational study of the experimentally known isomers of *closo*-C2B8H10 with the Lewis bases OH(−) and NH3. Both bases are hard in terms of the HSAB theory of acids and bases [14] with OH(−) being harder than NH3 on the HSAB scale.

#### **2. Methods**

All of the stationary points in the reactions of *closo*-1,2-, 1,6-, and 1,10-C2B8H10 with OH(−) were optimized and frequencies were calculated at the SMD(water) [15,16]/B3LYP/6-311+G(2d,p) level, a model chemistry well-established for this class of materials [4–6]. The entries in Table 1 for B3LYP/6-311+G(2d,p) are single-point energies at SMD(water)/B3LYP/6-311+G(2d,p)-optimized geometries. The reactions of the same carborane with neutral NH3 were optimized at the B3LYP/6- 311+G(2d,p) level without taking solvation effects into account, which is quite reasonable for neutral species. Moreover, unlike the investigation of organic reaction mechanisms where the bonding is 2c–2e and bond-breaking reaction steps can cause the wave function to become unstable with respect to symmetry-breaking, in the investigation of electron deficient reactions, the bonding scheme fluidly changes between different patterns of multi-center bonding. While dynamic electron correlation is important (and described by the B3LYP approach used in this study), non-dynamic (i.e., static) electron correlation (caused by near degeneracy effects) is not important. All of the transition states were relaxed in terms of the application of the intrinsic reaction coordinate (IRC) approach at the SMD(water)/B3LYP/6-31+G(d) level for the reactions with OH(−) and at the B3LYP/6-31G(d) level for the reactions with NH3, and the structures obtained are very similar to those obtained with the 6-311+G(2d,p) basis set. In order to check the possible influence of dispersion corrections, the wB97XD/6-311+G(2d,p), model chemistry was also employed for some stationary points. All the computations were performed using Gaussian09, in which the above model chemistries and basis sets are incorporated [17].

**Table 1.** Solvation free energies (kcal·mol<sup>−</sup>1), and free energies relative to the appropriate reference. The small "o" refers to ortho (1,2-) C2B8H10. The capital letters "**A**", "**B**", "**C**", etc. and "TS#" are related to individual intermediates and transition states, respectively. If there were two conformations possible, two letters are used to differentiate between them (e.g., **oG**/**G**- -**O**, **oK**/**K**- -**O** and **oM**/**M**- -**O**). The capital "-**O**" and "-**N**" distinguish between the reactions of OH(−) and NH3, respectively. Wherever the intermediate is relatively stable (and already isolated or possibly trappable), "**P1**", "**P2**", "**P3**", etc. are used instead of "**A**", "**B**", "**C**".


1. Solvation free energies (the energy difference between B3LYP/6-311+G(2d,p) and SMD/B3LYP/6-311+G(2d,p) at 298.15 K (kcal·mol−1). 2. Relative free energies in kcal·mol−<sup>1</sup> (where the references are 1,2-C2B8H10+OH(−) for OH(−) and 1,2-C2B8H10+NH3 for NH3 reactions). The ΔG(aq,298K) and ΔG(solv) for 1,2-C2B8H10 are 0.0 kcal·mol−<sup>1</sup> and −0.5 kcal·mol−1, respectively. The solvation free energy of H2O is taken from the experiment (−2.05 kcal·mol<sup>−</sup>1). In some cases, H2NBH2 is added for mass balance. In the formation of C2B6H11(−) , the comparison is made with respect to 1,2-C2B8H10 + OH(−) +3H2O. 3. Geometry optimizations and frequencies calculated at the SMD(water)/B3LYP/6-311+G(2d,p) level of theory. 4. Geometry optimizations and frequencies calculated at the B3LYP/6-311+G(2d,p) level of theory.

#### **3. Results and Discussion**

#### *3.1. The Reaction with Hydroxides*

The first part of the reaction of 1,2-C2B8H10 with OH(−) is a quite straightforward process. This experimentally verified reaction, with the **oP1**-**O** final product, arachno-1,6,9-OC2B8H13(−) (Scheme 1), proceeds through three transition states (TSs) and two intermediates, with the latter still bearing OH(−) (see Figure 1). The B-H-B bridge is a result of H migration to the B(7)–B(8) position in the final step of this reaction cascade. However, when one molecule of H2O is added to **oP1**-**O**, the cluster further degrades and through seven TSs an intermediate **oI**-**O** is obtained, which is prone to further degradation, when another water molecule is added to the O–B–O–H chain through two hydrogen bonds. This initiation results, via selective degradation of B(3,6)-cage atoms, in the formation of arachno-C2B6H11(−) through a number of TSs and intermediates, this arachno system being also isolated experimentally (see Figures 1 and 2 and Table 1) [18]. Note that the first barrier associated with **oTS1-O** was also examined with wB97XD/6-311+G(2d,p) and no significant difference from the B3LYP/6-311+G(2d,p) value was found, i.e., 24.9 kcal·mol−<sup>1</sup> vs. 27.2 kcal·mol−<sup>1</sup> as seen from Table 1. The potential energy surface attributed to the reaction of 1,6-C2B8H10 + OH(−) was rather difficult to follow. Since this reaction was not observed experimentally, we moved this computational effort to Supplementary Materials (see Figures S1 and S2). The geometrical shape of the final product **mP3**-**O** bears a slight resemblance to the nido-11 vertex geometry, but not with the open pentagonal belt because the OH(−) group migrates through the entire process without any indication of the insertion of oxygen into the cage boron atoms (see Figure S1). The mechanism of the reaction of the 1,10 isomer with OH(−) , also not observed experimentally, is even more complex. Interestingly, the final product (**pP1**-**O**, see Figures S3 and S4) of the latter reaction is of the same molecular shape, i.e., with a B–O–B bridge, as in the case of the reaction of 1,2-C2B8H10 with OH(−) (**oP1**-**O**), but it originates through five TSs instead of three.

**Figure 1.** *Cont*.

**Figure 1.** Individual stationary points as determined in the reaction pathway of the reaction of *closo*-1,2-C2B8H10 with OH(−) .

**Figure 2.** *Cont*.

**Figure 2.** Relative free energies (kcal·mol<sup>−</sup>1) of the individual stationary points on the Potential Energy Surface (PES) of the reaction of *closo*-1,2-C2B8H10 with OH(−) (see Table 1 and Methods for details).

#### *3.2. The Reaction with Amines*

The reaction of *closo*-1,2-C2B8H10 with NH3 has been experimentally known to provide *arachno*-1,6,9-NC2B8H13 (Scheme 1) [11,17]. The initiation of this reaction is based on the attack of NH3 on the most positive boron within the cage, i.e., B(3), which forms a triangle with both C vertices. The highest free energy barrier (50.3 kcal·mol−1) is **TS2**-**N**, through which the NH3 group becomes NH2 in the 1,6,9 isomer denoted as **oP1**-**N** in the reaction pathway (see Figures 3 and 4 and Table 1). However, this experimentally known part of the entire reaction (see above) occurs without any large intervening barriers and, consequently, the **oP1**-**N** isomer is obtained through two transition states and one intermediate. The same is apparently true for various amines of the R1R2NH type [19]. This reaction proceeds through a series of intermediate steps to the known 1,8,11-isomer (**oP2**-**N**), experimentally available by another procedure [20]. To our knowledge, there is no experimental evidence of the conversion of **oP2**-**N** to **oP1**-**N**, although this process is computed to be exothermic (ΔG = <sup>−</sup>8.6 kcal·mol−1). Note that the experimentally detected *arachno*-1,6,9-NC2B8H13 originates under less exothermic conditions than its oxygen analog. In analogy with the reaction of 1,2-C2B8H10 with OH(−) , we also examined the first barrier using wB97XD/6-311+G(2d,p) and again no significant difference from the B3LYP/6-311+G(2d,p) value was computed, i.e., 28.3 kcal·mol−<sup>1</sup> vs. 32.1 kcal·mol−<sup>1</sup> as provided by Table 1. When the 1,6,9-isomer (**oP1**-**N**) was examined computationally in terms of searching for another TS, i.e., **oTS7**-**N**, it isomerizes to a new isomer, **oP3**-**N**, through two TSs and one intermediate, **oE**-**N**, with the C–C bond remains intact as judged by a separation of 1.565 Å. When the

C–C separation was increased, another TS and intermediate, i.e., **oTS9**-**N** and **oF**-**N**, respectively, were located with much longer C ... C separations of 2.409 and 2.774 Å, respectively. This significant geometrical change initiates a continuation of the reaction through other four subsequent TSs to the next known isomer, i.e., **oP4**-**N**. The reaction of the 1,6-isomers (see Figures S5 and S6) was a result of the simultaneous initial attack of NH3 on B(3) and C(6); the N atom forms a cap above the B(2)B(3)C(6) triangle in the transition state with a free energy barrier of 34.6 kcal·mol−1. The second free energy barrier was even higher, 52.0 kcal·mol−<sup>1</sup> and might account for the fact that this reaction does not occur experimentally. Interestingly, when the common product of the reactions of *closo*-1,2-C2B8H10 and *closo*-1,6-C2B8H10 with NH3, i.e., **oP4**-**N**, is reached (see also Figure S5), both reactions proceed in the same way and *closo*-C2B7H9 is obtained, where the two carbon atoms are separated from each other. Basically, there are five isomers of *closo*-C2B7H9. The isomerization barrier between the most stable *C*2v (C ... C separation) and the third most stable *C*<sup>1</sup> (C–C bond) forms is quite high (36 kcal·mol<sup>−</sup>1). The high barrier is not unusual as isomerizations involving the C–C bond in carboranes often have rather high barriers. This type of *closo*/*closo* isomerization can also be described [21] in a more detailed way in terms of the consecutive double-Diamond-Square-Diamond (DSD) mechanism [10,22]. These two *closo* isomers have already been discussed by Schleyer [7] favoring the *C*2v-symmetrical 4,5-isomer as well.

**Figure 3.** *Cont*.

**Figure 3.** Individual stationary points on the PES of the reaction of *closo*-1,2-C2B8H10 with NH3.

**Figure 4.** Relative free energies (kcal·mol<sup>−</sup>1) of the individual stationary points on the PES of the reaction of *closo*-1,2-C2B8H10 with NH3 (see Table 1 and Methods for details).

The reaction of the 1,10 isomer with NH3 is rather complex and proceeds through a cascade of TSs. The initial attack of NH3 simultaneously takes place at the boron atoms of both CB4 hemispheres (**pTS1**-**N**) with an initial free energy barrier of more than 60 kcal·mol−<sup>1</sup> and *closo*-C2B7H9 originating as the final product in the endothermic reaction, which is entirely the same as above (see Figures S7 and S8). The experimental reaction of neither the 1,6- nor the 1,10-isomer of C2B8H10 with NH3 have been successfully carried out.

#### **4. Conclusions**

The reaction pathways of the experimentally known reactions of *closo*-1,2-C2B8H10 with both OH(−) and NH3 were computed using the DFT protocol. The final predicted products from the extensive search of the potential energy surfaces correspond to the same products detected experimentally. Both the *closo*-1,6-C2B8H10 and the *closo*-1,10-C2B8H10 isomers were allowed to react with the OH(−) and NH3 bases, without any defined products being observed. Finally, this work represents a computational attempt to study the debor reaction, in contrast to the debor principle [23] (the successive elimination of vertices), where boron vertices are removed in the course of the reaction as illustrated by obtaining *arachno*-C2B6H11(−) as the very final product from the reaction of *closo*-1,2-C2B8H10 with OH(−) .

*Crystals* **2020**, *10*, 896

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4352/10/10/896/s1: Figure S1: Individual stationary points as detected in the reaction pathway of the reaction of *closo*-1,6-C2B8H10 with OH(−) , Figure S2: Relative free energies (kcal·mol<sup>−</sup>1) of the individual stationary points on the PES of the reaction of *closo*-1,6-C2B8H10 with OH(−) , Figure S3: Individual stationary points as detected in the reaction pathway of the reaction of *closo*-1,10-C2B8H10 with OH(−) , Figure S4: Relative free energies (kcal·mol<sup>−</sup>1) of the individual stationary points on the PES of the reaction of *closo*-1,10-C2B8H10 with OH(−) , Figure S5: Individual stationary points as detected in the reaction pathway of the reaction of *closo*-1,6-C2B8H10 with NH3, Figure S6: Relative free energies (kcal·mol<sup>−</sup>1) of the individual stationary points on the PES of the reaction of *closo*-1,6-C2B8H10 with NH3, Figure S7: Individual stationary points as detected in the reaction pathway of the reaction of 1,10-C2B8H10 with NH3, Figure S8: Relative free energies (kcal·mol<sup>−</sup>1) of the individual stationary points on the PES of the reaction of *closo*-1,10-C2B8H10 with NH3, Table S1: Number of imaginary frequencies, zero-point energies (kcal·mol<sup>−</sup>1), heat capacity correction (kcal·mol<sup>−</sup>1), entropies (cal.mol−1·K<sup>−</sup>1), solvation free energies (kcal·mol<sup>−</sup>1), and free energies relative to the appropriate reference. Small o, m, or p refer to ortho (1,2-), meta (1,6-), or para (1,10-)C2B8H10, respectively. Capital letters "A", "B", "C", etc. or "TS#" are related to individual intermediates or transition states, respectively. If there were two conformations possible, two letters are used to discern them (e.g., oG/G -O, oK/K -O, or oM/M -O). Capital "-O" or "-N" distinguish between the reactions of OH(-) or NH3, respectively. In cases where the intermediate is relatively stable (and already isolated or possibly trappable), "P1", "P2", "P3", etc. is used instead of "A", "B", "C". Table S2: Cartesian coordinates of all species in Table S1

**Author Contributions:** Computations, M.L.M., J.F., and D.H.; synthesis, J.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** We thank the Czech Science Foundation for financial support (project no. 19-17156S). MLM thanks Auburn University for access to the HOPPER computer.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Crystals* Editorial Office E-mail: crystals@mdpi.com www.mdpi.com/journal/crystals

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com

ISBN 978-3-0365-3620-0