Rhodapentalenes: Pincer Complexes with Internal Aromaticity

Article

Rhodapentalenes: Pincer Complexes with Internal Aromaticity Qingde Zhuo, Hong Zhang, Linting Ding, ..., Yuhui Hua, Jun Zhu, Haiping Xia [email protected]

HIGHLIGHTS A concept for pincer chemistry: internal aromaticity involving metal d-orbitals Stable cyclic rhodium complexes with obvious rhodium carbene character Dual stabilization of rigid CCC pincer architectures and intrinsic aromaticity Rhodium-containing aromatics with a single metal entity

DATA AND CODE AVAILABILITY www.ccdc.cam.ac.uk/ data_request/cif

Zhuo et al., iScience 19, 1214– 1224 September 27, 2019 ª 2019 The Author(s). https://doi.org/10.1016/ j.isci.2019.08.027

Article

Rhodapentalenes: Pincer Complexes with Internal Aromaticity Qingde Zhuo,1,3 Hong Zhang,1,3 Linting Ding,1 Jianfeng Lin,1 Xiaoxi Zhou,1 Yuhui Hua,1 Jun Zhu,1 and Haiping Xia1,2,4,* SUMMARY Pincer complexes are a remarkably versatile family benefited from their stability, diversity, and tunability. Many of them contain aromatic organic rings at the periphery, and aromaticity plays an important role in their stability and properties, whereas their metallacyclic cores are not aromatic. Herein, we report rhodapentalenes, which can be viewed as pincer complexes in which the metallacyclic cores exhibit considerable aromatic character. Rhodapentalenes show good thermal stability, although the rhodium-carbon bonds in such compounds are fragile. Experimental and computational studies suggest that the stabilization of rigid CCC pincer architectures together with an intrinsic aromaticity is vital for these metallacyclic rhodium species. Dearomatization-aromatization reactions, corresponding to metal-ligand cooperation of classical aromatic pincer complexes, were observed in this system. These findings suggest a new concept for pincer chemistry, the internal aromaticity involving metal d-orbitals, which would be useful for exploiting the nature of construction motif and inspire further applications.

INTRODUCTION Since the first reports in the late 1970s (Kelly et al., 1971; Moulton and Shaw, 1976; van Koten et al., 1978), pincer complexes have made great progresses and have been widely used in the fields of synthesis (Martinez et al., 2016; Pell et al., 2017; Rafiq et al., 2019), catalysis (Gao et al., 2018; Luque-Urrutia et al., 2019; Nielsen et al., 2013; Niu et al., 2019; Zhang et al., 2018), materials science (Albrecht et al., 2000; To et al., 2017; Zhang et al., 2017a), and biological systems (Desguin et al., 2015; Fellner et al., 2017). This has had an effect on the development of inorganic chemistry, materials chemistry, supramolecular chemistry, and bioorganometallics chemistry and on bond-making and bond-breaking processes (Albrecht and Lindner, 2011; Albrecht and van Koten, 2001; Gunanathan and Milstein, 2014; Kumar et al., 2017; Leis et al., 2008; Li et al., 2019; Morales-Morales, 2018, O’Reilly and Veige, 2014; Szabo´ and Wendt, 2014). Pincer complexes typically refer to tridentate chelates where the tridentate ligands bind to the metal centers in a meridional fashion (Albrecht and van Koten, 2001; Peris and Crabtree, 2018). The advantages of pincer complexes lie in their good thermal stability and tunable chemical properties. The electronic and steric properties of the metal centers can be easily modulated by introducing donor atoms (N, O, P, S, and C, etc.), substituents (alkyl or aryl groups), and frameworks (neutral, anionic, and cationic frameworks, etc.) and by the size of the metallic rings (five- or six-membered rings) of the pincer complexes. The thermal stability of pincer complexes can be enhanced by increasing the rigidity of pincer ligands, and a common way to achieve rigid pincer ligands is to incorporate the aromatic groups into the ligand backbones (Peris and Crabtree, 2018). Thus, although both aliphatic and aromatic backbones, represented by I and II in Scheme 1, were introduced in the early work of pincer complexes (Empsall et al., 1977; Kelly et al., 1971; Moulton and Shaw, 1976; van Koten et al., 1978), the latter, exemplified by the aromatic NCN palladium pincer complex (van Koten et al., 1978) shown in Scheme 1 have dominated the subsequent literature.

1State

Key Laboratory of Physical Chemistry of Solid Surfaces and Collaborative Innovation Center of Chemistry for Energy Materials (iChEM), College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China

2Department

Aromaticity is one of the most fundamental concepts in chemistry, and has played an important role in the development of pincer chemistry. Aromatic pincer ligands provide a rigid coordination environment for the metal center, thus resulting in pincer complexes with high thermal stability (Albrecht and van Koten, 2001). New mode of metal-ligand cooperation was developed based on the dearomatization-aromatization of pincer ligands, which has led to unusual bond activation processes for many novel environmentally benign catalysis (Gunanathan and Milstein, 2011, 2013; 2014; Khusnutdinova and Milstein, 2015; Li et al., 2019; Zell and Milstein, 2015). A general feature of these pincer complexes is the aromatic organic rings fused to their central metallacycles, whereas the metallacycles of these pincer complexes are not aromatic. Herein we refer to these complexes as ‘‘pincer complexes with peripheral aromaticity’’ (Scheme 1, middle).

1214 iScience 19, 1214–1224, September 27, 2019 ª 2019 The Author(s). This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

of Chemistry, Shenzhen Grubbs Institute, Southern University of Science and Technology, Shenzhen 518055, China

3These

authors contributed

equally 4Lead

Contact

*Correspondence: [email protected] https://doi.org/10.1016/j.isci. 2019.08.027

Scheme 1. Primary Coordination Environment and New Concept of Pincer Complexes

In this article, we describe rhodapentalenes, a class of compounds that can be considered as pincer complexes, and present a concept of pincer complex with internal aromaticity (Scheme 1, right). Unlike traditional pincer complexes in which the metal centers are not involved in the aromatic rings, the metal centers of pincer complexes with internal aromaticity participate in the construction of the aromatic systems and the metallacyclic core is aromatic. The rhodium pincer complexes with internal aromaticity, rhodapentalenes, were synthesized by efficient one-pot reactions of triyne chains with commercially available RhCl(PPh3)3 and acid. Remarkably, although rhodapentalenes display an evident rhodium carbene character and readily undergo ring-opening and ring-expansion reactions with nucleophiles and oxidants, respectively, they exhibit good thermal stability. Experimental and theoretical studies show that the intrinsic aromaticity together with the rigid pincer frameworks stabilize the cyclic rhodium species. Notably, dearomatization-aromatization processes related to classical aromatic pincer complexes were also developed in the rhodapentalene system.

RESULTS Synthesis and Characterization of CCC Rhodium Pincer Complexes: Rhodapentalenes As shown in Scheme 2, treatment of the triyne chain 1a or 1b (Zhuo et al., 2017) with RhCl(PPh3)3 and HBF4 at room temperature (rt) led to the rhodapentalenes 2a or 2b, CCC rhodium pincer complexes, in isolated yields of 88% and 81%, respectively. The structures of 2a and 2b were confirmed by NMR spectroscopy (please see Figures S18–S37 for the NMR spectra of the reported compounds in this work). Their NMR spectra are similar; in the 1H NMR spectrum of 2a, C1H, C3H, and C7H were observed at 11.76, 7.91, and 12.83 ppm and in the spectrum of 2b, at 11.50, 8.07, and 13.08 ppm, respectively. These proton chemical shifts are consistent with those of the reported osmapentalenes (Zhu et al., 2014) and osmapentalynes (Zhu et al., 2013; Zhu and Xia, 2018; Zhuo et al., 2017) and are located in the metalla-aromatic region (Bleeke, 2001; Cao et al., 2014; Chen and Jia, 2013; Ferna´ndez et al., 2015; Frogley and Wright, 2018; Landorf and Haley, 2006; Saito, 2012; Wei et al., 2018; Wright, 2017). In the 13C NMR spectra, the resonances of the metal-bound carbon atoms appeared at 239.66 (C1), 188.66 (C4), and 265.29 ppm (C7) for 2a and at 239.86 (C1), 187.62 (C4), and 261.20 ppm (C7) for 2b. The chemical shifts of C4 are slightly down-field from those of Rh–Cvinyl carbons and fall within the range of Rh–Caryl carbons (Kubo et al., 2005), whereas the chemical shifts of C1 and C7 are consistent with those reported for rhodium carbenes (Kornecki et al., 2013; Werle´ et al., 2016). X-ray crystallographic analysis confirmed the pincer structure of 2a. As shown in Figure 1 (see Figure S1 for details), the tridentate CCC pincer ligand is attached to the rhodium center. The core metallabicycle is coplanar, as indicated by the small mean deviation (0.018 A˚) from the least-squares plane through Rh and C1 to C7. The carbon-carbon bond distances (1.362(5)–1.425(5) A˚) in the metallabicycle core show somewhat alternation, but are intermediate between typical single and double bonds, which is comparable with those of typical metalla-aromatics, such as the first metallabenzene (1.36(2)–1.42(2) A˚) (Elliott et al., 1982), metallabenzyne (1.376(5)–1.420(5) A˚) (Wen et al., 2001), and osmapentalene (1.365(5)–1.414(9) A˚) (Zhu et al., 2014). The bond lengths of Rh–C1 (2.048(3) A˚) and Rh–C7 (2.056(3) A˚) are identical and at the high end of the range for Rh=C double bonds (1.743–2.059 A˚, ranges based on data from the Cambridge Structural Database, CSD version 5.40, in November 2018) (Werle´ et al., 2016). The Rh–C4 bond length is

iScience 19, 1214–1224, September 27, 2019 1215

Scheme 2. Synthesis of Rhodapentalenes

1.987(3) A˚, which is slightly shorter than those of Rh–C1 and Rh–C7. This is possibly a combined result of the weak trans influence of the Cl ligand (Housecroft and Sharpe, 2012) and the delocalized structure of 2a around the peripheral skeleton. The structural features of 2a are similar to those of osmapentalenes (Zhu et al., 2014), suggesting that it can be represented by the resonance structures of rhodapentalene A and rhodapentalene B (Scheme 2). Note that although carbon is ubiquitous in coordination chemistry, carbon as binding atom is relatively rare and CCC rhodium pincer complexes are uncommon (Kubo et al., 2005). To our knowledge, rhodapentalenes 2 represent the first examples of all-carbon-ligated rhodium pincer complexes, in which the pincer skeletons are composed entirely of carbon atoms. Remarkably, rhodapentalenes 2 exhibit high thermal stability. For example, 2a can survive in air at 120
C for at least 3 h in the solid state, and subsequent thermal gravimetric analysis shows that weight changes occur only over 150
C (see Figures S7 and S8). To understand the mechanism for the formation of rhodapentalenes 2, we studied the reaction of 1a with RhCl(PPh3)3 in the absence of HBF4, which led to a new CCC pincer complex (3), isolated in 93% yield (Scheme 2). The structure of 3 was determined by X-ray crystallographic analysis (Figures 2 and S2). The crystal structure of 3 shows that C3 is sp3-hybridized and is attached to a hydroxyl group. Distinct from the delocalized structure of 2a, the core structure in 3 is localized within the metallabicycle, as reflected by the bond lengths in the structure. The C1–C2 (1.358(6) A˚), C4–C5 (1.375(6) A˚), and C6–C7 (1.337(6) A˚) bond lengths are consistent with those of double bonds, whereas the C2–C3 (1.503(6) A˚), C3–C4 (1.479(6) A˚), and C5–C6 (1.461(6) A˚) bond lengths are appropriate for single bonds. The bond lengths of the three Rh–C bonds are 2.065(4) A˚ (Rh–C1), 1.999(4) A˚ (Rh–C4), and 2.118(4) A˚ (Rh–C7), which are in accordance with the bond lengths of Rh–Cvinyl bonds (Wu et al., 2002). The Rh–C1 and Rh–C7 bond lengths are obviously longer than the corresponding Rh–C bonds lengths in 2a. In addition, 3 is less thermally stable than 2a. A solid sample of 3 begins to convert into unidentified material at 60
C in air. In view of the fact that the hydroxyl group is labile in the presence of acids, we conjectured that the rhodabicycle (3) may be a key intermediate in the formation of rhodapentalene (2a). As expected, when HBF4 was added to a green solution of 3, the reaction mixture immediately turned red and 2a was isolated in 91% yield after workup (Scheme 2). Based on the experimental observations, a plausible mechanism was postulated for the formation of 2 in Scheme S1.

DFT Computations on the Internal Aromaticity of Pincer Complexes 2 The good thermal stability, ring planarity, and low-field proton chemical shifts indicate that rhodapentalenes 2 are aromatic. To gain more insight into the aromaticity and electronic structure of the rhodapentalenes 2, we performed density functional theory ([DFT] B3LYP/6-311++G(d,p)) calculations based on a simplified unsubstituted model complex 20 , in which the PPh3 ligands were replaced by PH3 groups (see the cartesian coordinate in Data S2). We first investigated the aromatic stabilization energy of 20 by the isomerization stabilization energy (ISE) method outlined by Schleyer and Pu¨hlhofer (Schleyer and

1216 iScience 19, 1214–1224, September 27, 2019

Figure 1. X-Ray Molecular Structure of the Cation of Rhodapentalene 2a Ellipsoids are displayed at the 50% probability. The phenyl groups of PPh3 and the ester groups on C9 are omitted for clarity. Selected bond distances (A˚) and bond angles (deg): Rh–C1 2.048(3), Rh–C4 1.987(3), Rh–C7 2.056(3), C1–C2 1.372(5), C2–C3 1.425(5), C3–C4 1.362(5), C4–C5 1.405(5), C5–C6 1.387(5), C6–C7 1.399(5); Rh–C1–C2 113.9(3), C1–C2–C3 115.3(3), C2–C3–C4 114.5(3), C3–C4–Rh 116.8(2), C1–Rh–C4 79.39(13), Rh–C4–C5 114.9(2), C4–C5–C6 116.2(3), C5–C6–C7 115.1(3), C6–C7–Rh 113.5(2), C7–Rh–C4 80.14(13).

Pu¨hlhofer, 2002; Wannere et al., 2003). As shown in Scheme 3A, the ISE values of 20 (26.7 and 25.6 kcal mol1) are close to those of benzene (33.2 and 29.0 kcal mol1) (Schleyer and Pu¨hlhofer, 2002; Wannere et al., 2003), indicating global aromaticity in 2. The nucleus-independent chemical shift (NICS) is another computational method commonly exploited to diagnose aromaticity (Schleyer et al., 1996). The NICS(1)zz value for each ring of 20 was calculated to be 12.97 ppm (Scheme 3B, also see the values of the model complex 2-PH3 with the charged phosphonium group in Figure S13), which is comparable with the value (19.6 ppm) calculated by Mauksch and Tsogoeva (Mauksch and Tsogoeva, 2018), and indicates the aromaticity of 2. The aromaticity of 2 is further supported by the anisotropy of the current-induced density (Herges and Geuenich, 2001), which can simulate the density and direction of the induced ring current in a molecular system under an external magnetic field. As shown in Scheme 3C and Figure S17, an obvious clockwise diatropic ring current can be observed in the p-system of 20 , indicating p-aromaticity in the two fused five-membered rings. Therefore, experimental observations and the results of DFT calculations both demonstrate that rhodapentalenes (2) exhibit considerable aromatic character. The facile synthesis of 2 could be attributed to the internal aromaticity of 2 and could be considered to be an aromaticity-driven process. Notably, although rhodium has been demonstrated to form the metalla-aromatic model complexes in the pioneering theoretical work of metalla-aromatics by Hoffmann (Thorn and Hoffmann, 1979), and is among the metals that have been most investigated in the theoretical studies of metalla-aromatic chemistry (Iron et al., 2003; Islas et al., 2014; Ferna´ndez and Frenking, 2007; Mauksch and Tsogoeva, 2018), well-defined rhoda-aromatics remain scarce (Wei et al., 2015; Zhang et al., 2017b). All the known examples require extra stabilization by coordination to additional metal fragments (Wei et al., 2015; Zhang et al., 2017b). Rhodapentalenes 2 represent the first free rhoda-aromatics. Selected molecular orbitals (MOs) for the model complex 20 are shown in Figure 3. The five occupied pMOs (highest occupied molecular orbital [HOMO]-1, HOMO-2, HOMO-5, HOMO-10, and HOMO-12) are derived mainly from the orbital interactions between the d-orbitals of the Rh atom (5dxz and 5dyz) and the pzp orbitals of the C7H5 unit, indicating the involvement of the d-orbitals of the metal center in the p-delocalization along the perimeter of the pincer skeleton. These results are similar to those computed for the osmapentalenes (Zhu et al., 2014) and ruthenapentalenes (Zhuo et al., 2018). However, p-overlaps between the d-orbitals on metal centers and the p-orbital on the carbon atoms of rhodapentalene 20 are less effective when compared with osmapentalenes (see Figures S9 and S10) and ruthenapentalenes (see Figures S11 and S12). To reveal the nature of the bonding in rhodapentalenes, natural bond orbital analysis was performed. The Wiberg bond indices ([WBIs] bond orders, which are a measure of bond strength) of C–C bonds for 20 are between 1.33 and 1.56 and are in accordance with the unsaturated character of the rhodapentalene ring. Notably, the WBIs of Rh–C bonds (Rh–C1/C7: 0.76) are significantly smaller than those of Os–C (Os–C1/C7: 1.02) and Ru–C (Ru–C1/C7: 0.91) bonds in the corresponding metallapentalenes (see Table S3), which also confirms the diminished orbital hybridization of the metal center and the carbon atoms in rhodapentalenes. The WBIs of Rh–C bonds in rhodapentalenes are in accordance with those of Rh–C bonds in reported polarized rhodium-carbene complexes, which were critical

iScience 19, 1214–1224, September 27, 2019 1217

Figure 2. X-Ray Molecular Structure of Compound 3 Ellipsoids are displayed at the 50% probability. The phenyl groups of PPh3 and the ester groups on C9 are omitted for clarity. Selected bond distances (A˚) and bond angles (deg): Rh–C1 2.065(4), Rh–C4 1.999(4), Rh–C7 2.118(4), C1–C2 1.358(6), C2–C3 1.503(6), C3–C4 1.479(6), C3–O1 1.395(6), C4–C5 1.375(6), C5–C6 1.461(6), C6–C7 1.337(6); Rh–C1–C2 115.4(3), C1–C2–C3 117.4(4), C2–C3–C4 108.9(4), C2–C3–O1 115.6(4), C3–C4–Rh 117.5(3), C1–Rh–C4 80.29(17), Rh–C4–C5 114.7(3), C4–C5–C6 115.8(4), C5–C6–C7 117.1(4), C6–C7–Rh 111.6(3), C7–Rh–C4 80.76(17).

intermediates in rhodium-catalyzed reactions (Padwa et al., 2000; Sheehan et al., 1988). We expect that the relatively weaker rhodium-carbon bonds would result in rhodapentalenes having high reactivity.

Reactivity of Pincer Complex 2a Encouraged by the computational results, we studied the reaction chemistry of rhodapentalenes. We chose 8-hydroxyquinoline to test the reactivity of rhodapentalene (2a) because of its versatile function as a bidentate ligand and a nucleophile. As shown in Scheme 4A, the reaction of 2a with 8-hydroxyquinoline immediately produced a ring-opened product 4, in which the Rh–C7 bond had been cleaved (see Scheme 4B and Figure S3 for its crystal structure). A possible mechanism is proposed in Scheme S2 for the formation of 4. Judging by the reaction of 2a with 8-hydroxyquinoline, rhodapentalenes should be sensitive toward PPh3. Indeed, trace amounts of PPh3 can trigger the transformation of 2a into a mixture, from which ring-opening product 5 can be isolated in low yield and whose X-ray crystal structure is shown in Figure S4. When the reaction was performed in the presence of HBF4 and NaCl, 5 can be obtained cleanly (Scheme 4A). We deduce that the formation of 5 (see Scheme S3) is also driven by a nucleophilic addition step, similar to that in the formation of 4. It should be noted that nucleophilic addition-induced ring-opening reactions are rare in metalla-aromatic species (Zhang et al., 2010), and the reported examples usually require reaction temperatures above rt or a long reaction time (for example, more than 4 days) (Zhang et al., 2010). The mild reaction conditions in the case of 2a could be attributed to the weakness of the rhodium-carbon bonds in rhodapentalenes, in agreement with the computational results. We also probed the reactions of rhodapentalene 2a with a number of oxidants, such as pyridine N-oxide, dimethyl sulfoxide, and 3-chloroperoxybenzoic acid, which all lead to the same product. As demonstrated by the reaction with pyridine N-oxide in Scheme 4A, 6 was isolated as a blue powder in 82% yield. The X-ray diffraction study revealed that 6 contains an oxygen-containing six-membered ring arising from the insertion of an oxygen atom into one of the five-membered metallacycles (see Scheme 4B and Figure S5). To our knowledge, there are no reports of oxygen-insertion reactions of metalla-aromatics. The observed ring expansion of rhodapentalene 2a may be induced by the nucleophilic attack of pyridine N-oxide at the C7 site (see Scheme S4). The reactions of 2a with 8-hydroxyquinoline, PPh3, and pyridine N-oxide indicate the pronounced electrophilicity and lability of the rhodium-carbon bonds in rhodapentalenes and are consistent with the chemical behavior of rhodium carbene moieties (Berry, 2012; Tindall et al., 2018; Zhang et al., 2019). The steric crowding of the bulky phosphonium group adjacent to the C1-position may account for the C7-position being the sole active site in the reactions of 2a. Dearomatization-aromatization reactions, similar to those of classical pincer ligand systems (Gunanathan and Milstein, 2011, 2013, 2014; Khusnutdinova and Milstein, 2015; Li et al., 2019; Zell and Milstein, 2015),

1218 iScience 19, 1214–1224, September 27, 2019

Scheme 3. Evaluation of the Internal Aromaticity of Model Pincer Complex 20 (A) ISE (isomerization stabilization energy) evaluations of the aromaticity of 20 . (B) NICS(1)zz (nucleus-independent chemical shift along the z axis at 1 A˚ above the ring critical point) evaluations of the aromaticity of 20 . (C) ACID (anisotropy of the current-induced density) plot of 20 from p contribution with an isosurface value of 0.025. The magnetic field vector is orthogonal to the ring plane and points upward, i.e., outward direction of the z axis (aromatic species exhibit clockwise diatropic circulations).

can also be achieved in rhodapentalenes. As shown in Scheme 4A, one of the protons attached to C8 in 2a can be readily removed when operated in column chromatography with neutral alumina, leading to compound 7. Both structural characterization data (Scheme 4B and Figure S6) and DFT calculations (Figures S14–S16) demonstrate the nonaromaticity of 7. Interestingly, 7 could be re-aromatized upon treatment with HBF4, regenerating the original rhodapentalene 2a. The interconversion between aromatic complex 2a and nonaromatic complex 7 corresponds to the intriguing dearomatization-aromatization process of classical aromatic pincer complexes. Such dearomatization-aromatization reactions of pincer complexes based on pyridine and acridine have been regarded as promising activation mode via metal-ligand cooperation and led to extraordinary applications in various dehydrogenative/hydrogenative and bond activation reactions (Gunanathan and Milstein, 2011, 2013; 2014; Khusnutdinova and Milstein, 2015; Li et al., 2019; Zell and Milstein, 2015).

DISCUSSION The discovery of rhodapentalenes (2) featuring both rhodium carbene and aromatic character leads to interesting questions regarding the enhancement of the stability of the compounds. The high thermal stability of rhodapentalenes 2 is somewhat surprising, as complexes with obvious rhodium carbene property are generally labile. Rhodium carbene complexes have often been proposed as key intermediates of numerous rhodium-catalyzed reactions but have rarely been isolated (Berry, 2012; Davies and Manning, 2008; DeAngelis et al., 2016). Besides, such cyclic rhodium carbene complexes have not been reported to date, although rhodium was predicted to form rhodabenzenes in a pioneering theoretical work of metalla-aromatic chemistry (Thorn and Hoffmann, 1979). Previous studies indicate that rhodabenzenes decompose readily through carbene migration reactions (Wu et al., 2002; Haley, 2017), and we reasoned that the exceptional thermal stability of rhodapentalenes (2) can be attributed to the considerable aromatic character and rigid chelating properties of CCC pincer frameworks. Aromaticity has long been regarded as the property that can efficiently stabilize a wide range of reactive compounds. Besides that, the high thermal stabilities associated with pincer complexes have been ascribed to the chelating and multidentate nature of pincer ligands. The CCC-pincer ligand-based

iScience 19, 1214–1224, September 27, 2019 1219

Figure 3. Selected Molecular Orbitals of 20 The eigenvalues of the molecular orbitals are given in parentheses.

rhodapentalene (2) is significant in that it represents the first example of rhodium pincer complexes with aromatic metallacycles. The fact that the rhodium carbene structures formed through the chelation of aliphatic carbon chains with the metal center were very stable prompted us to re-examine the structure of pincer complexes. We envision that pincer complexes with aromatic metallacycle cores should be classified as pincer complexes with internal aromaticity (demonstration in Scheme 1). The significant difference between the pincer complexes with internal aromaticity (III) and common aromatic pincer complexes (II) is the participation in the aromaticity of the d-orbitals of the metal centers. We have recently reported metallapentalenes and metallapentalynes with group 8 metal centers. However, in these complexes the metalcarbon multiple bonds are rather robust (Zhu et al., 2013; Zhu and Xia, 2018; Zhuo et al., 2018). In the case of rhodapentalenes 2, the metallapentalene chemistry has been extended for the first time from group 8 to group 9 elements. The phenomenon of both high reactivity and good thermal stability of rhodapentalenes 2 made us aware of the dual stabilization of pincer framework and aromaticity in this system for the first time. More interestingly, dearomatization/aromatization reactions can also be realized in rhodapentalene system. In this context, previous reported metallapentalynes (Zhu et al., 2013; Zhu and Xia, 2018; Zhuo et al., 2018), metallapentalenes (Zhu et al., 2014), and other bridge-headed fused metalla-aromatic compounds (Frogley et al., 2018; Frogley and Wright, 2014, 2017; Wang et al., 2012, 2013) could also be regarded as pincer complexes with internal aromaticity. We believe that these conceptually new pincer complexes, which combine the structural features of both aromatic compounds and rigid pincer complexes would provide new opportunities for pincer chemistry. In summary, we have discovered a new class of metalla-aromatic compounds, the rhodapentalenes, which can be viewed as CCC pincer complexes with internal aromaticity. Rhodapentalenes exhibit good thermal stability, although electronic structure analysis and reactivity studies reveal the high reactivity of the Rh–C(1) and Rh–C(7) bonds toward nucleophiles. These results demonstrate the promising stabilization of internal aromatic pincer frameworks, which should be attributed to dual stabilization deriving from both rigid polydentate chelation and aromatic stabilization energy. Given the above-mentioned findings the pincer complex platform with internal aromaticity can serve as a new candidate for the stabilization of vulnerable species.

Limitations of the Study The reactions of rhodapentalenes with unsaturated species such as alkenes and alkynes were also tested, but the anticipated carbene insertion products were not observed.

1220 iScience 19, 1214–1224, September 27, 2019

Scheme 4. Reactions of Rhodapentalene 2a (A) Ring-opening, ring-expansion, and dearomatization-aromatization reactions of rhodapentalene 2a (B) X-ray molecular structure of complex 7 and the cations of complexes 4 and 6. Ellipsoids are displayed at the 50% probability and the phenyl groups of PPh3, and the ester groups on C9 are omitted for clarity.

METHODS All methods can be found in the accompanying Transparent Methods supplemental file.

DATA AND CODE AVAILABILITY Crystallographic data were deposited in the Cambridge Crystallographic Data Centre (CCDC) under accession numbers CCDC: 1506335 (2a), 1506333 (3), 1848362 (4), 1848363 (5), CCDC-1848364 (6), and CCDC-1848365 (7). These data can be obtained free of charge from the Cambridge Crystallographic Date Centre via www.ccdc.cam.ac.uk/data_request/cif. CIF files for 2a and 3–7 are also provided in Data S1.

iScience 19, 1214–1224, September 27, 2019 1221

SUPPLEMENTAL INFORMATION Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2019.08.027.

ACKNOWLEDGMENTS This research was supported by the National Key R&D Program of China (2017YFA0204902) and the National Natural Science Foundation of China (Nos. U1705254, 21572185, and 21490573).

AUTHOR CONTRIBUTIONS Conceptualization, H.X.; Methodology, Q.Z.; Investigation, Q.Z., L.D., J.L., X.Z., and Y.H.; Writing – Original Draft, Q.Z. and H.Z.; Writing – Review & Editing, Q.Z., H.Z., and H.X.; Funding Acquisition, H.Z. and H.X.; Resources, H.Z., J.Z., and H.X.; Supervision, H.X.

DECLARATION OF INTERESTS Q.Z., H.Z., J.L., X.Z., and H.X. are inventors on three patents related to this work filed by Xiamen University (no. 201510241958.4, licensed November 22, 2018; no. 201510606889.2, licensed September 12, 2018; no. 15/811,277, licensed January 24, 2018). The authors declare no other competing interests.

Received: July 4, 2019 Revised: August 8, 2019 Accepted: August 16, 2019 Published: September 27, 2019 REFERENCES Albrecht, M., and Lindner, M.M. (2011). Cleavage of unreactive bonds with pincer metal complexes. Dalton Trans. 40, 8733–8744. Albrecht, M., Lutz, M., Spek, A.L., and van Koten, G. (2000). Organoplatinum crystals for gastriggered switches. Nature 406, 970–974. Albrecht, M., and van Koten, G. (2001). Platinum group organometallics based on ‘‘pincer’’ complexes: sensors, switches, and catalysts. Angew. Chem. Int. Ed. 40, 3750–3781. Berry, J.F. (2012). The role of three-center/fourelectron bonds in superelectrophilic dirhodium carbene and nitrene catalytic intermediates. Dalton Trans. 41, 700–713. Bleeke, J.R. (2001). Metallabenzenes. Chem. Rev. 101, 1205–1227. Cao, X.-Y., Zhao, Q., Lin, Z., and Xia, H. (2014). The chemistry of aromatic osmacycles. Acc. Chem. Res. 47, 341–354. Chen, J., and Jia, G. (2013). Recent development in the chemistry of transition metal-containing metallabenzenes and metallabenzynes. Coord. Chem. Rev. 257, 2491–2521. Davies, H.M.L., and Manning, J.R. (2008). Catalytic C–H functionalization by metal carbenoid and nitrenoid insertion. Nature 451, 417–424. DeAngelis, A., Panish, R., and Fox, J.M. (2016). Rh-Catalyzed intermolecular reactions of a-alkyla-diazo carbonyl compounds with selectivity over b-hydride migration. Acc. Chem. Res. 49, 115–127. Desguin, B., Zhang, T., Soumillion, P., Hols, P., Hu, J., and Hausinger, R.P. (2015). A tethered niacin-derived pincer complex with a nickel-

1222 iScience 19, 1214–1224, September 27, 2019

carbon bond in lactate racemase. Science 349, 66–69. Elliott, G.P., Roper, W.R., and Waters, J.M. (1982). Metallacyclohexatrienes or ’metallabenzenes.’ Synthesis of osmabenzene derivatives and X-ray crystal structure of [Os(CSCHCHCHCH)(CO)(PPh3)2]. J. Chem. Soc. Chem. Commun. 811–813. Empsall, H.D., Hyde, E.M., Markham, R., McDonald, W.S., Norton, M.C., Shaw, B.L., and Weeks, B. (1977). Synthesis and X-ray structure of an unusual iridium ylide or carbene complex. J. Chem. Soc. Chem. Commun. 589–590. Fellner, M., Desguin, B., Hausinger, R.P., and Hu, J. (2017). Structural insights into the catalytic mechanism of a sacrificial sulfur insertase of the N-type ATP pyrophosphatase family, LarE. Proc. Natl. Acad. Sci. U S A 114, 9074–9079. Ferna´ndez, I., and Frenking, G. (2007). Aromaticity in metallabenzenes. Chem. Eur. J. 13, 5873–5884. Ferna´ndez, I., Frenking, G., and Merino, G. (2015). Aromaticity of metallabenzenes and related compounds. Chem. Soc. Rev. 44, 6452–6463.

Frogley, B.J., and Wright, L.J. (2018). Recent advances in metallaaromatic chemistry. Chem. Eur. J. 24, 2025–2038. Gao, Y., Emge, T.J., Krogh-Jespersen, K., and Goldman, A.S. (2018). Selective dehydrogenative coupling of ethylene to butadiene via an iridacyclopentane complex. J. Am. Chem. Soc. 140, 2260–2264. Gunanathan, C., and Milstein, D. (2011). Metal– ligand cooperation by aromatization– dearomatization: a new paradigm in bond activation and ‘‘green’’ catalysis. Acc. Chem. Res. 44, 588–602. Gunanathan, C., and Milstein, D. (2013). Applications of acceptorless dehydrogenation and related transformations in chemical synthesis. Science 341, 1229712. Gunanathan, C., and Milstein, D. (2014). Bond activation and catalysis by ruthenium pincer complexes. Chem. Rev. 114, 12024– 12087. Haley, M.M. (2017). Metallabenzenes and valence isomers via the nucleophilic 3vinylcyclopropene route. In Metallabenzenes: An Expert View, L.J. Wright, ed. (John Wiley & Sons Ltd), pp. 109–143.

Frogley, B.J., Perera, L.C., and Wright, L.J. (2018). Syntheses of amino-substituted iridabenzofurans and subsequent selective N-functionalisation. Chem. Eur. J. 24, 4304–4309.

Herges, R., and Geuenich, D. (2001). Delocalization of electrons in molecules. J. Phys. Chem. A 105, 3214–3220.

Frogley, B.J., and Wright, L.J. (2014). Fused-ring metallabenzenes. Coord. Chem. Rev. 270, 151–166.

Housecroft, C.E., and Sharpe, A.G. (2012). Inorganic Chemistry, Fourth Edition (Prentice Hall), p. 831.

Frogley, B.J., and Wright, L.J. (2017). A metallaanthracene and derived metallaanthraquinone. Angew. Chem. Int. Ed. 56, 143–147.

Iron, M.A., Martin, J.M.L., and van der Boom, M.E. (2003). Metallabenzene versus Cp complex formation: a DFT investigation. J. Am. Chem. Soc. 125, 13020–13021.

Islas, R., Poater, J., and Sola`, M. (2014). Analysis of the aromaticity of five-membered heterometallacycles containing Os, Ru, Rh, and Ir. Organometallics 33, 1762–1773.

Niu, W., Gonsales, S.A., Kubo, T., Bentz, K.C., Pal, D., Savin, D.A., Sumerlin, B.S., and Veige, A.S. (2019). Polypropylene: now available without chain ends. Chem 5, 237–244.

Kelly, W.S.J., Ford, G.H., and Nelson, S.M. (1971). Studies on the magnetic crossover in five-coordinate complexes of iron(II), cobalt(II), and nickel(II). Part I. J. Chem. Soc. A, 388–396.

O’Reilly, M.E., and Veige, A.S. (2014). Trianionic pincer and pincer-type metal complexes and catalysts. Chem. Soc. Rev. 43, 6325–6369.

Khusnutdinova, J.R., and Milstein, D. (2015). Metal–ligand cooperation. Angew. Chem. Int. Ed. 54, 12236–12273. Kornecki, K.P., Briones, J.F., Boyarskikh, V., Fullilove, F., Autschbach, J., Schrote, K.E., Lancaster, K.M., Davies, H.M.L., and Berry, J.F. (2013). Direct spectroscopic characterization of a transitory dirhodium donor-acceptor carbene complex. Science 342, 351–354. Kubo, K., Jones, N.D., Ferguson, M.J., McDonald, R., and Cavell, R.G. (2005). Chelate and pincer carbene complexes of rhodium and platinum derived from hexaphenylcarbodiphosphorane, Ph3P=C=PPh3. J. Am. Chem. Soc. 127, 5314– 5315. Kumar, A., Bhatti, T.M., and Goldman, A.S. (2017). Dehydrogenation of alkanes and aliphatic groups by pincer-ligated metal complexes. Chem. Rev. 117, 12357–12384. Landorf, C.W., and Haley, M.M. (2006). Recent advances in metallabenzene chemistry. Angew. Chem. Int. Ed. 45, 3914–3936. Leis, W., Mayer, H.A., and Kaska, W.C. (2008). Cycloheptatrienyl, alkyl and aryl PCP-pincer complexes: ligand backbone effects and metal reactivity. Coord. Chem. Rev. 252, 1787–1797. Li, H., Gonc¸alves, T.P., Lupp, D., and Huang, K.-W. (2019). PN3(P)-pincer complexes: cooperative catalysis and beyond. ACS Catal. 9, 1619–1629. Luque-Urrutia, J.A., Sola`, M., Milstein, D., and Poater, A. (2019). Mechanism of the manganesepincer-catalyzed acceptorless dehydrogenative coupling of nitriles and alcohols. J. Am. Chem. Soc. 141, 2398–2403. Martinez, G.E., Ocampo, C., Park, Y.J., and Fout, A.R. (2016). Accessing pincer bis(carbene) Ni(IV) complexes from Ni(II) via halogen and halogen surrogates. J. Am. Chem. Soc. 138, 4290–4293. Mauksch, M., and Tsogoeva, S.B. (2018). Strict correlation of HOMO topology and magnetic aromaticity indices in d-block metalloaromatics. Chem. Eur. J. 24, 10059–10063.

Padwa, A., Snyder, J.P., Curtis, E.A., Sheehan, S.M., Worsencroft, K.J., and Kappe, C.O. (2000). Rhodium(II)-catalyzed equilibration of push-pull carbonyl and ammonium ylides. A computationally based understanding of the reaction pathway. J. Am. Chem. Soc. 122, 8155– 8167. Pell, C.J., Zhu, Y., Huacuja, R., Herbert, D.E., Hughes, R.P., and Ozerov, O.V. (2017). Fluorocarbene, fluoroolefin, and fluorocarbyne complexes of Rh. Chem. Sci. 8, 3178–3186. Peris, E., and Crabtree, R.H. (2018). Key factors in pincer ligand design. Chem. Soc. Rev. 47, 1959– 1968. Rafiq, S., Bezdek, M.J., Chirik, P.J., and Scholes, G.D. (2019). Dinitrogen coupling to a terpyridinemolybdenum chromophore is switched on by Fermi resonance. Chem 5, 402–416. Saito, M. (2012). Challenge to expand the concept of aromaticity to tin- and lead-containing carbocyclic compounds: synthesis, structures and reactions of dilithiostannoles and dilithioplumbole. Coord. Chem. Rev. 256, 627–636. Schleyer, P.v.R., Maerker, C., Dransfeld, A., Jiao, H., and Hommes, N.J.R.v.E. (1996). Nucleusindependent chemical shifts: a simple and efficient aromaticity probe. J. Am. Chem. Soc. 118, 6317–6318. Schleyer, P.v.R., and Pu¨hlhofer, F. (2002). Recommendations for the evaluation of aromatic stabilization energies. Org. Lett. 4, 2873–2876. Sheehan, S.M., Padwa, A., and Snyder, J.P. (1988). Dirhodium(II) tetracarboxylate carbenoids as catalytic intermediates. Tetrahedron Lett. 39, 949–952. Szabo´, K.J., and Wendt, O.F. (2014). Pincer and Pincer-type Complexes: Applications in Organic Synthesis and Catalysis (Wiley-VCH). Thorn, D.L., and Hoffmann, R. (1979). Delocalization in metallocycles. Nouv. J. Chim. 3, 39–45.

Morales-Morales, D. (2018). Pincer Compounds: Chemistry and Applications (Elsevier).

Tindall, D.J., Werle´, C., Goddard, R., Philipps, P., Fare`s, C., and Fu¨rstner, A. (2018). Structure and reactivity of half-sandwich Rh(+3) and Ir(+3) carbene complexes. Catalytic metathesis of azobenzene derivatives. J. Am. Chem. Soc. 140, 1884–1893.

Moulton, C.J., and Shaw, B.L. (1976). Transition metal–carbon bonds. Part XLII. Complexes of nickel, palladium, platinum, rhodium and iridium with the tridentate ligand 2,6-bis[(di-tbutylphosphino)methyl]phenyl. J. Chem. Soc. Dalton Trans. 1020–1024.

To, W.-P., Zhou, D., Tong, G.S.M., Cheng, G., Yang, C., and Che, C.-M. (2017). Highly luminescent pincer gold(III) aryl emitters: thermally activated delayed fluorescence and solution-processed OLEDs. Angew. Chem. Int. Ed. 56, 14036–14041.

Nielsen, M., Alberico, E., Baumann, W., Drexler, H.-J., Junge, H., Gladiali, S., and Beller, M. (2013). Low-temperature aqueous-phase methanol dehydrogenation to hydrogen and carbon dioxide. Nature 495, 85–89.

van Koten, G., Timmer, K., Noltes, J.G., and Spek, A.L. (1978). A novel type of Pt–C interaction and a model for the final stage in reductive elimination processes involving C–C coupling at Pt; synthesis and molecular geometry of [1,N,N0 -h-2,6-bis

{(dimethylamino)methyl}-toluene] iodoplatinum(II) tetrafluoroborate. J. Chem. Soc. Chem. Commun. 250–252. Wang, T., Zhang, H., Han, F., Lin, R., Lin, Z., and Xia, H. (2012). Synthesis and characterization of a metallapyridyne complex. Angew. Chem. Int. Ed. 51, 9838–9841. Wang, T., Zhang, H., Han, F., Long, L., Lin, Z., and Xia, H. (2013). Key intermediates of iodinemediated electrophilic cyclization: isolation and characterization in an osmabenzene system. Angew. Chem. Int. Ed. 52, 9251–9255. Wannere, C.S., Moran, D., Allinger, N.L., Hess, B.A., Schaad, L.J., and Schleyer, P.V.R. (2003). On the stability of large [4n]annulenes. Org. Lett. 5, 2983–2986. Wei, J., Zhang, W.-X., and Xi, Z. (2018). The aromatic dianion metalloles. Chem. Sci. 9, 560–568. Wei, J., Zhang, Y., Zhang, W.-X., and Xi, Z. (2015). 1,3-Butadienyl dianions as non-innocent ligands: synthesis and characterization of aromatic dilithio rhodacycles. Angew. Chem. Int. Ed. 54, 9986– 9990. Wen, T.B., Zhou, Z.Y., and Jia, G.C. (2001). Synthesis and characterization of a metallabenzyne. Angew. Chem. Int. Ed. 40, 1951–1954. Werle´, C., Goddard, R., Philipps, P., Fare`s, C., and Fu¨rstner, A. (2016). Structures of reactive donor/ acceptor and donor/donor rhodium carbenes in the solid state and their implications for catalysis. J. Am. Chem. Soc. 138, 3797–3805. Wright, L.J. (2017). Metallabenzenes: An Expert View (John Wiley & Sons Ltd). Wu, H.-P., Weakley, T.J.R., and Haley, M.M. (2002). Metallabenzenes and valence isomers. 5. Synthesis and structural characterization of a rhodabenzvalene: a rare h2-cyclopropene/svinylrhodium(I) complex. Organometallics 21, 4320–4322. Zell, T., and Milstein, D. (2015). Hydrogenation and dehydrogenation iron pincer catalysts capable of metal–ligand cooperation by aromatization/dearomatization. Acc. Chem. Res. 48, 1979–1994. Zhang, D., Kang, Z., Liu, J., and Hu, W. (2019). Metal-dependent umpolung reactivity of carbenes derived from cyclopropenes. iScience 14, 292–300. Zhang, H., Lin, R., Hong, G., Wang, T., Wen, T.B., and Xia, H. (2010). Nucleophilic aromatic addition reactions of the metallabenzenes and metallapyridinium: attacking aromatic metallacycles with bis(diphenylphosphino) methane to form metallacyclohexadienes and cyclic h2-allene-coordinated complexes. Chem. Eur. J. 16, 6999–7007. Zhang, K., Yeung, M.C.-L., Leung, S.Y.-L., and Yam, V.W.-W. (2017a). Manipulation of nanostructures in the Co-assembly of platinum(II) complexes and block copolymers. Chem 2, 825–839. Zhang, Y., Wei, J., Chi, Y., Zhang, X., Zhang, W.-X., and Xi, Z. (2017b). Spiro metalla-aromatics of Pd, Pt, and Rh: synthesis and characterization. J. Am. Chem. Soc. 139, 5039–5042.

iScience 19, 1214–1224, September 27, 2019 1223

Zhang, Y., Chen, X., Zheng, B., Guo, X., Pan, Y., Chen, H., Li, H., Min, S., Guan, C., Huang, K.-W., and Zheng, J. (2018). Structural analysis of transient reaction intermediate in formic acid dehydrogenation catalysis using two-dimensional IR spectroscopy. Proc. Natl. Acad. Sci. U S A 115, 12395–12400.

membered rings with a transition metal. Nat. Chem. 5, 698–703.

Zhu, C., Li, S., Luo, M., Zhou, X., Niu, Y., Lin, M., Zhu, J., Cao, Z., Lu, X., Wen, T.B., et al. (2013). Stabilization of anti-aromatic and strained five-

Zhu, C., and Xia, H. (2018). Carbolong chemistry: a story of carbon chain ligands and transition metals. Acc. Chem. Res. 51, 1691–1700.

1224 iScience 19, 1214–1224, September 27, 2019

Zhu, C., Luo, M., Zhu, Q., Zhu, J., Schleyer, P.v.R., Wu, J.I.-C., Lu, X., and Xia, H. (2014). Planar Mo¨bius aromatic pentalenes incorporating 16 and 18 valence electron osmiums. Nat. Commun. 5, 3265.

Zhuo, Q., Lin, J., Hua, Y., Zhou, X., Shao, Y., Chen, S., Chen, Z., Zhu, J., Zhang, H., and Xia, H. (2017). Multiyne chains chelating osmium via three metal-carbon s bonds. Nat. Commun. 8, 1912. Zhuo, Q., Zhang, H., Hua, Y., Kang, H., Zhou, X., Lin, X., Chen, Z., Lin, J., Zhuo, K., and Xia, H. (2018). Constraint of a rutheniumcarbon triple bond to a five-membered ring. Sci. Adv. 4, eaat0336.

ISCI, Volume 19

Supplemental Information

Rhodapentalenes: Pincer Complexes with Internal Aromaticity Qingde Zhuo, Hong Zhang, Linting Ding, Jianfeng Lin, Xiaoxi Zhou, Yuhui Hua, Jun Zhu, and Haiping Xia

Proposed Mechanisms

Scheme S1. Plausible mechanism for the formation of rhodapentalenes 2, related to Scheme 2.

Scheme S2. Plausible mechanism for the formation of 4, related to Scheme 4.

Scheme S3. Plausible mechanism for the formation of 5, related to Scheme 4.

Scheme S4. Plausible mechanism for the formation of 6, related to Scheme 4.

Crystal Structures

Figure S1. X-ray molecular structure of the cation of complex 2a (ellipsoids are drawn at the 50% probability level). The hydrogen atoms of PPh3 are omitted for clarity, related to Figure 1.

Figure S2. X-ray molecular structure of complex 3 (ellipsoids are drawn at the 50% probability level). The hydrogen atoms of PPh3 are omitted for clarity, related to Figure 2.

Figure S3. X-ray molecular structure of the cation of complex 4 (ellipsoids are drawn at the 50% probability level), related to Scheme 4B. The hydrogen atoms in the aromatic rings are omitted for clarity. Selected bond distances (Å) and angles (deg): Rh–C1 1.989(3), Rh–C4 2.038(3), C1–C2 1.358(4), C2–C3 1.465(4), C2–P2 1.781(3), C3–C4 1.346(5), C4–C5 1.470(4), C5–C6 1.347(5), C6–C7 1.499(5), C7–P3 1.826(3); Rh–C1–C2 115.0(2), C1–C2–C3 115.4(3), C2–C3–C4 114.2(3), C3–C4–Rh 114.4(2), C1–Rh–C4 80.45(12), Rh–C4–C5 118.5(2), C4–C5–C6 127.0(3), C5–C6–C7 125.2(3).

Figure S4. X-ray molecular structure of the cation of complex 5 (ellipsoids are drawn at the 50% probability level), related to Scheme 4. The hydrogen atoms of PPh3 are omitted for clarity. Selected bond distances (Å) and angles (deg): Rh–C1 1.977(5), Rh–C4 2.022(5), C1–C2 1.351(7), C2–C3 1.469(7), C2–P2 1.776(5), C3–C4 1.348(7), C4–C5 1.454(7), C5–C6 1.348(7), C6–C7 1.492(7), C7–P3 1.822(5); Rh–C1–C2 115.4(4), C1–C2–C3 114.8(4), C2–C3–C4 114.2(4), C3–C4–Rh 114.2(4), C1–Rh–C4 80.7(2), Rh–C4–C5 121.0(3), C4–C5–C6 127.5(5), C5–C6–C7 125.3(5).

Figure S5. X-ray molecular structure of the cation of complex 6 (ellipsoids are drawn at the 50% probability level), related to Scheme 4B. The hydrogen atoms of PPh3 are omitted for clarity. Selected bond distances (Å) and angles (deg): Rh–C1 1.982(4), Rh–C4 2.011(4), Rh–O1 2.182(3), C1–C2 1.350(6), C2–C3 1.449(6), C3–C4 1.350(6), C4–C5 1.437(6), C5–C6 1.352(6), C6–C7 1.437(7), C7–O1 1.225(5); Rh–C1–C2 115.7(3), C1–C2–C3 114.5(4), C2–C3–C4 114.9(4), C3–C4–Rh 114.4(3), C1–Rh–C4 80.41(16), Rh–C4–C5 122.8(3), C4–C5–C6 127.0(4), C5–C6–C7 127.3(4), C6–C7–O1 126.6(4), C7–O1–Rh1 125.1(3), O1–Rh–C4 90.96(14).

Figure S6. X-ray molecular structure of complex 7 (ellipsoids are drawn at the 50% probability level), related to Scheme 4B. The hydrogen atoms of PPh3 are omitted for clarity. Selected bond distances (Å) and angles (deg): Rh–C1 2.056(3), Rh–C4 2.001(3), Rh–C7 2.114(3), C1–C2 1.365(5), C2–C3 1.459(5), C3–C4 1.338(5), C4–C5 1.472(5), C5–C6 1.451(5), C6–C7 1.342(5), C5–C8 1.346(5), C8–C9 1.521(5), C9–C10 1.573(5), C6–C10 1.516(5); Rh–C1–C2 112.9(2), C1– C2–C3 115.9(3), C2–C3–C4 114.5(3), C3–C4–Rh 116.6(2), C1–Rh–C4 80.10(13), Rh–C4–C5 113.3(2), C4–C5–C6 114.1(3), C5–C6–C7 118.5(3), C6–C7–Rh 112.3(2), C7–Rh–C4 81.83(13), C6–C5–C8 112.4(3), C5–C8–C9 110.9(3), C8–C9–C10 104.0(3), C9–C10–C6 104.5(3).

TGA Data for Complex 2a

120

152 ° C

100

Weight (%)

80

60

40

20

0 0

200

400

600

800

o

Temperature ( C)

Figure S7. TGA data for complex 2a at a constant heating rate of 10 °C min−1 under a flowing air atmosphere, related to Scheme 2.

120

195 °C

Weight (%)

100

80

60

40

0

200

400

600

800

o

Temperature ( C)

Figure S8. TGA data for complex 2a at a constant heating rate of 10 °C min−1 under a flowing N2 atmosphere, related to Scheme 2.

Selected Molecular Orbitals of Osmapentalenes and Ruthenapentalenes

Figure S9. Selected frontier molecular orbitals of osmapentalene Os-1 calculated at the B3LYP/[LanL2DZ for Os, Cl and P; 6-311++G(d,p) for C and H]. The eigenvalues of the molecular orbitals are given in parentheses, related to Figure 3.

Figure S10. Selected frontier molecular orbitals of osmapentalene Os-2 calculated at the B3LYP/[LanL2DZ for Os and P; 6-311++G(d,p) for O, C and H]. The eigenvalues of the molecular orbitals are given in parentheses, related to Figure 3.

Figure S11. Selected frontier molecular orbitals of ruthenapentalene Ru-1 calculated at the B3LYP/[LanL2DZ for Ru, Cl and P; 6-311++G(d,p) for C and H]. The eigenvalues of the molecular orbitals are given in parentheses, related to Figure 3.

Figure S12. Selected frontier molecular orbitals of ruthenapentalene Ru-2 calculated at the B3LYP/[LanL2DZ for Ru and P; 6-311++G(d,p) for O, C and H]. The eigenvalues of the molecular orbitals are given in parentheses, related to Figure 3.

Aromaticity Evaluations

Figure S13. NICS evaluations of model complex 2-PH3, related to Scheme 3.

Figure S14. ISE evaluation of model complex 7-1', related to Scheme 4.

Figure S15. NICS evaluations of model complex 7', related to Scheme 4.

Figure S16. ACID plot of model complex 7′ from π contributions with an isosurface value of 0.025. No clear diatropic or paratropic ring current could be found in the metallacycles, related to Scheme 4.

Figure S17. ACID plot of model complex 2′ from π contributions with an isosurface value of 0.025. The magnetic field vector is orthogonal to the ring plane and points upward (aromatic species exhibit clockwise diatropic circulations), related to Scheme 3C.

1.67 1.20 0.82 0.81 0.80

2.33

7.91 7.82 7.81 7.80 7.56 7.55 7.54 7.54 7.53 7.52 7.34 7.33 7.28 7.21 7.20 7.19 6.92 6.91 6.90 6.89 5.27 5.26 3.69 3.64 3.53 3.04

11.77 11.74

12.83

NMR Spectra

COOCH3

C8 H C10H C3 H

12

11 1.00

13 1.00

14

10

9

8

n-hexane

CH2Cl2

7

6

5

4

3

2

1

0

-1

ppm

2.03

CH

1.06

1

2.01

CH

6.03

7

Figure S18. 1H NMR spectrum (600.1 MHz) of complex 2a in CD2Cl2 at room temperature, related to Scheme 2.

9.72

27.04 26.60

Rh(PPh3)2 CPPh3

100

90

Figure S19.

80 31P

70

60

50

40

30

20

10

0

-10

-20

-30

-40

-50

ppm

NMR spectrum (242.9 MHz) of complex 2a in CD2Cl2 at room temperature,

related to Scheme 2.

C

C7

260

Figure S20.

C5

4

ppm

240

13C

220

related to Scheme 2. 132.08 132.05 132.02 131.71 131.68 131.65

157.46 157.29

188.82 188.77 188.66 188.63 188.46

157.5 ppm

C6

200 180 160

54.36 54.18 54.00 53.82 53.68 53.64

COOCH3

C3 C5

C3 132.0

C1

140 120 100 80

COOCH3

C9

C2

60

C10 C8 54.0

40 20

ppm

ppm

n-hexane

CHCl3

0 ppm

NMR spectrum (150.9 MHz) of complex 2a in CD2Cl2 at room temperature,

188.86 188.82 188.77 188.66 188.63 188.46 172.54 171.10 157.46 157.29 135.39 134.89 134.87 134.65 134.21 134.14 132.08 132.05 132.02 131.71 131.68 131.65 131.03 130.82 130.74 128.40 119.48 118.90 78.04 63.73 54.36 54.18 54.00 53.82 53.76 53.68 53.64 39.83 38.60 32.00 23.09 14.37

239.66

265.29

13 12 11

related to Scheme 2. 10

CH2Cl2 2.3

9 8 7 6 5

2.2 2.1

4 2.0 1.9

3

1

CH

CH 1.8

3 2.05 2.01

C1 H

2.02

CH ppm

1.01

14 1.02

1.00

7 11.51 11.48

C10H C9 H 8 H2O C H

1.7 1.6

2

1.5

1

1.4 ppm

C9 H 8 C10H C H

0 ppm

Figure S21. 1H NMR spectrum (600.1 MHz) of complex 2b in CD2Cl2 at room temperature, 1.57 1.49 1.48 1.47 1.45 1.44 1.38 1.36 1.35

2.22 2.21 2.20 2.20 2.19

2.22 2.21 2.20 2.20 2.19 1.57 1.49 1.48 1.47 1.45 1.44 1.38 1.36 1.35

8.07 7.80 7.79 7.77 7.57 7.56 7.56 7.55 7.54 7.54 7.30 7.29 7.28 7.15 6.91 6.90 6.89 6.88 5.24 5.23

11.51 11.48

13.08

9.17 9.14 9.11

28.93 28.91 28.49 28.46

Rh(PPh3)2

CPPh3

90

80

Figure S22.

70 31P

60

50

40

30

20

10

0

-10

-20

-30

-40

-50

ppm

NMR spectrum (242.9 MHz) of complex 2b in CD2Cl2 at room temperature,

related to Scheme 2.

C1

240

260

C4

ppm

C7

240

Figure S23. 13C

220

related to Scheme 2.

C3

187.5 ppm

C5

C1

200

C6

180

C2

ppm

C2

160 140 120

129.5

100 80

9

C C8

60 40

C10

ppm

C3

C4

20 0 ppm

NMR spectrum (150.9 MHz) of complex 2b in CD2Cl2 at room temperature, 129.26 129.23 129.21 129.18

129.65 129.62 129.59 129.57

157.41 157.23

187.83 187.80 187.77 187.67 187.64 187.61 187.58 187.48 187.44 187.41

240.04 239.98 239.91 239.81 239.74 239.67

261.20 240.04 239.98 239.91 239.81 239.74 239.67 195.42 187.83 187.80 187.77 187.67 187.64 187.61 187.58 187.48 187.44 187.41 178.70 157.41 157.23 134.95 134.93 134.39 133.92 133.85 131.12 130.41 130.27 130.20 129.65 129.62 129.59 129.57 129.26 129.23 129.21 129.18 128.45 127.79 119.29 118.71 53.84 53.66 53.48 53.30 53.12 32.50 31.46 28.42

10

to Scheme 2. 9 8 7 6 5

C10H

ppm 2.4

4 3

2.2

n-hexane

2.0

C3 H

2 1.00

C3 H

1.05 1.01 1.03 1.02

11

C7H CH2Cl2

3.00 3.01 1.01

C1 H

1.02

1.00

0.18 0.16

1.79 1.76

2.02 1.99

2.36 2.34 2.28 2.25

3.46 3.44

COOCH3

C8 H OH

ppm 0.2

1 0

ppm

OH

ppm

Figure S24. 1H NMR spectrum (600.1 MHz) of complex 3 in CD2Cl2 at room temperature, related

3.57 3.54 3.46 3.44 2.36 2.34 2.28 2.25 2.02 1.99 1.79 1.76 1.25 1.23 1.22 0.84 0.83 0.82 0.18 0.16

8.20 8.14 7.64 7.63 7.62 7.39 7.39 7.38 7.37 7.36 7.36 7.25 7.10 7.08 7.08 7.06 6.94 6.92 6.92 6.90 6.87 6.53 5.26 5.25

10.29 10.24

34.46 33.98 33.96 32.70 32.68 32.20 32.19 31.69 31.67 30.42 30.40 29.91 8.01 7.99 7.97 7.94 34.46 33.98 33.96 32.70 32.68 32.20 32.19 31.69 31.67 30.42 30.40

Rh(PPh3)2 CPPh3 35 34 33 32 31

160

140

Figure S25.

120 31P

to Scheme 2.

ppm

100

80

60

40

20

0

-20

-40

-60

-80

-100

ppm

NMR spectrum (242.9 MHz) of complex 3 in CD2Cl2 at room temperature, related

C1

ppm

220

Figure S26. 200

13C

to Scheme 2. 167.65 167.60 167.45 167.40 167.34 167.25

221.39 221.31 221.22 221.14

C7

C4 153.5

167.5 ppm

C1 C4

180 160

C2

ppm

140 120

C3

ppm

C6 C5 C7

100

COOCH3

ppm

80

C9

COOCH3

60

C8

C 10

40

ppm

n-hexane

C3

20 0 ppm

NMR spectrum (150.9 MHz) of complex 3 in CD2Cl2 at room temperature, related

52.70

52.90

80.40 80.26 80.23

122.76

123.06

153.60 153.52 153.49 153.45 153.42 153.39 153.35

221.31 221.22 221.14 174.08 173.27 167.65 167.60 167.54 167.51 167.45 167.40 167.34 167.30 167.25 167.21 154.85 153.60 153.52 153.49 153.45 153.42 153.39 153.35 145.57 136.42 134.58 134.51 133.63 129.82 129.74 129.20 128.02 127.32 123.17 123.06 122.76 122.61 80.40 80.26 80.23 64.39 54.36 54.18 54.00 53.82 53.64 52.90 52.70 39.06 36.01 32.14 23.21 14.44

to Scheme 4A.

C10H C3 H

8 7 6 5 4 3 ppm

C7 H

2 1.01

9 ppm

1.12

C1 H

2.07

9.97 9.94

2.11 2.09 2.07 1.28 1.26 1.23 0.57 0.55 0.54 0.52

CH2Cl2

1.15

10 ppm

1.06 3.15 3.02

C1 H

1.04

1.00

COOCH3

C7 H C7 H C10H

ppm

C8H H2O C7H C10H 1 ppm

Figure S27. 1H NMR spectrum (600.1 MHz) of complex 4 in CD2Cl2 at room temperature, related

0.00

2.11 2.09 2.07 1.63 1.58 1.28 1.26 1.23 0.57 0.55 0.54 0.52

4.21 3.86 3.72 3.67 3.60 3.51 3.45 3.41 3.35

8.07 8.06 8.00 7.67 7.65 7.64 7.42 7.41 7.39 7.35 7.34 7.19 7.18 6.99 6.98 6.95 6.94 6.69 6.68 6.63 6.62 5.25 5.24 5.23 5.23

9.97 9.94

14.31 6.78 6.71

25.17 24.14

C7PPh3

C2PPh3 Rh(PPh3)2

140

120

100

80

60

40

20

0

-20

-40

-60

-80

-100

ppm

Figure S28. 31P NMR spectrum (161.9 MHz) of complex 4 in CD2Cl2 at room temperature, related to Scheme 4A.

178.74 178.56

C4

ppm

C1

220 200

to Scheme 4A. 152.02 151.92

214.67

C1 COOCH3

C4

180

C

160 140

C3

C5

ppm

5

120 ppm

ppm

100 80

C6 COOCH3

ppm

C9

C6 C10C

60 40

ppm

20

23.61

23.96

51.81 51.75

119.16 119.13

119.54

126.56 126.49

109.33 109.27

C2 C7

ppm ppm

8

C7

0 ppm

Figure S29. 13C NMR spectrum (150.9 MHz) of complex 4 in CD2Cl2 at room temperature, related

0.00

178.74 178.56 172.62 170.50 168.79 152.02 151.92 143.41 141.93 134.77 133.94 133.70 132.75 132.51 130.72 129.53 129.35 129.16 126.95 126.56 126.49 119.89 119.74 119.54 119.16 119.13 116.78 116.21 115.01 109.33 109.27 106.20 55.21 53.06 52.88 52.70 52.52 52.34 51.81 51.75 43.34 42.30 23.96 23.61

214.67

to Scheme 4A. 5 4 3 2 2.01

6

CH

8

2.08

7 2.03

8

C10H

2.00

9

C3 H C7 H

6.08 2.08

10 6.05

C1 H

2.08 2.09

2.00

COOCH3

H2O C7 H C10H

1 0 ppm

Figure S30. 1H NMR spectrum (400.1 MHz) of complex 5 in CD2Cl2 at room temperature, related

2.46 2.42 2.09 2.05 1.71 1.53 1.49 1.45 1.04 1.02 1.00 0.97

4.33 4.30 4.27 4.21 3.88 3.82 3.53 3.47

5.26

7.84 7.67 7.66 7.64 7.61 7.54 7.42 7.34 7.05 7.02 7.02 7.00 6.98 6.96 6.95 6.93 6.36

9.20 9.16

7.67 7.60

15.19

26.05 25.02

C7PPh3

RhPPh3

100

90

80

70

60

50

40

30

20

C2PPh3

10

0

-10

-20

-30

-40

-50

-60

ppm

Figure S31. 31P NMR spectrum (161.9 MHz) of complex 5 in CD2Cl2 at room temperature, related to Scheme 4A.

210

C1 C4

ppm 177

C1

200 190

to Scheme 4A. 151.28 151.18

176.50

176.81

198.69 198.57 198.38 198.26

C3

COOCH3

ppm

C4

180

C

170 160

5

150

C

140 130 120

C2

ppm 122

C

110 100

120

90

118

80

116

C5

70

54.03 53.89 53.76 53.49 53.22 53.17 52.95

114.04 113.95

119.54 118.66 117.89 117.04

123.12 122.44

129.76 129.55

COOCH3

C6

ppm

6

60

54.0

ppm

C 10

2

50 40

53.5

30 20

ppm

C9 C8 C7

10 ppm

Figure S32. 13C NMR spectrum (150.9 MHz) of complex 5 in CD2Cl2 at room temperature, related

198.69 198.57 198.38 198.26 176.81 176.50 172.61 171.28 151.28 151.18 135.91 135.69 134.88 134.10 133.70 133.60 133.45 133.36 132.05 131.22 130.66 130.30 130.18 130.07 129.76 129.55 128.26 127.83 123.12 122.44 119.54 118.66 117.89 117.04 114.04 113.95 56.59 54.03 53.96 53.89 53.76 53.49 53.22 53.17 52.95 44.43 43.28 26.30 25.79

1.56

2.86 2.74

3.74

5.32

8.28 7.84 7.82 7.81 7.57 7.44 7.42 7.40 7.38 7.27 7.25 7.23 6.87 6.85 6.84 6.82 6.21

9.75 9.71

CH2Cl2

COOCH3

C10H CH 8

C7 H

8

7

6

5

4

3

2

1

0

ppm

2.02 2.03

9 0.99

10 1.00

11

H2O

C3 H

6.01

CH

1.05

1

Figure S33. 1H NMR spectrum (400.1 MHz) of complex 6 in CD2Cl2 at room temperature, related to Scheme 4A.

8.29 8.21

26.48 25.82

Rh(PPh3)2

CPPh3

120

Figure S34.

100 31P

to Scheme 4A.

80

60

40

20

0

-20

-40

-60

-80

ppm

NMR spectrum (161.9 MHz) of complex 6 in CD2Cl2 at room temperature, related

174.02 173.95

144.38 144.16

C4 C3

ppm

220 174

C

C1

200

to Scheme 4A. ppm

C

180

4

160

COOCH3

C5 C3

140 120 100 ppm

C9

7 ppm

C

80 60 40 20

54.0

53.30

53.84 53.63 53.57

54.11

54.38

122.79

123.51

216.07 215.72

C1 C2 COOCH3

ppm

8

C10

C2

0 ppm

Figure S35. 13C NMR spectrum (100.6 MHz) of complex 6 in CD2Cl2 at room temperature, related

193.06 174.02 173.95 171.16 157.65 144.38 144.16 135.17 135.15 134.93 134.89 134.84 134.29 134.14 134.01 133.91 131.53 131.00 130.76 130.62 130.49 130.28 128.57 128.53 128.40 128.35 128.30 123.51 122.79 119.58 118.70 57.99 54.38 54.24 54.11 53.84 53.63 53.57 53.44 53.30 41.73 39.35

216.07 215.72

to Scheme 4A. 8 7 6

C8 H

5 4 2.04

9 6.07

10

C7 H C3 H

1.00

CH

1.00

1

1.02

1.00

COOCH3

CH2Cl2

C10H H2O Et2O

Et2O

3 2 1 0 ppm

Figure S36. 1H NMR spectrum (500.2 MHz) of complex 7 in CD2Cl2 at room temperature, related

0.00

1.48 1.08 1.07 1.06

2.29

3.57 3.51 3.36 3.34

7.61 7.60 7.60 7.58 7.41 7.40 7.39 7.39 7.38 7.37 7.36 7.35 7.16 7.15 7.04 6.92 6.91 6.90 6.88 6.54 5.77 5.25 5.23 4.68

10.21 10.21 10.16 10.15

3.89

32.56 31.98

Rh(PPh3)2

CPPh3

140

120

100

80

60

40

20

0

-20

-40

-60

-80

ppm

Figure S37. 31P NMR spectrum (202.5 MHz) of complex 7 in CD2Cl2 at room temperature, related to Scheme 4A.

Table S1. Crystal data and structure refinement of 2a, 3 and 4, related to Schemes 2 and 4. 2a·1.5C2H4Cl2

3·C4H8O

4·1.5CH2Cl2

Empirical formula

C71H64BCl4F4O4P3Rh

C72H67ClO6RhP3

C78.5H68BCl4F4NO5P3Rh

Mol. weight

1405.65

1259.52

1529.77

Temperature [K]

153.00(14)

168(20)

100.01(10)

Crystal system

Monoclinic

Monoclinic

Triclinic

Space group

P21/n

P2(1)/c

P-1

a [Å]

12.27280(13)

18.5774(4)

10.4683(4)

b [Å]

20.6749(3)

13.0396(3)

13.4616(4)

c [Å]

25.2565(3)

27.1373(5)

25.3697(9)

 []

90

90

94.693(3)

β []

98.6364(11)

99.351(2)

95.913(3)

γ []

90

90

97.674(3)

V [Å3]

6335.89(13)

6486.4(2)

3507.6(2)

Z

4

4

2

ρcalcd [g cm-3]

1.474

1.290

1.448

 [mm-1]

4.961

0.429

4.546

F(000)

2884.0

2616.0

1570.0

Crystal size [mm3]

0.2 × 0.2 × 0.1

0.200.200.20

0.1 × 0.1 × 0.05

Radiation

CuKα (λ = 1.54184)

MoKα (λ = 0.71073)

CuKα (λ = 1.54184)

2θ range [°]

7.608 to 124.592

3.474 to 49.998

7.04 to 134.996

Coll. refl.

35912

31595

24419

Indep. refl.

9985

11408

12635

data/restraints/params

9985/1/822

11408/150/825

12635/12/1090

GOF on F2

1.046

1.099

1.027

R1/wR2 [I  2σ(I)]

0.0433/0.1120

0.0572/0.1556

0.0539/0.1438

R1/wR2 (all data)

0.0482/0.1172

0.0694/0.1631

0.0576/0.1484

Largest peak/hole [e Å-3]

1.45/-1.04

1.90/-0.76

1.93/-1.44

Table S2. Crystal data and structure refinement of 5, 6 and 7, related to Scheme 4. 5·6CH2Cl2

6·1.5CH2Cl2

7·1.25CH2Cl2

Empirical formula

C142H130B3Cl15F12O8P6Rh2

C69.5H61BCl4F4O5P3Rh

C69.25H59.5Cl3.5O4P3Rh

Mol. weight

3148.27

1400.61

1275.56

Temperature [K]

173(2)

179(9)

178(5)

Crystal system

Triclinic

Monoclinic

Monoclinic

Space group

P-1

P21/n

P21/c

a [Å]

14.4220(4)

12.3594(2)

12.79757(18)

b [Å]

15.2205(4)

20.3965(4)

21.9384(3)

c [Å]

33.0221(7)

25.3431(5)

22.2894(3)

 []

94.6105(18)

90

90

β []

91.6756(19)

97.6116(19)

101.1559(15)

γ []

92.942(2)

90

90

V [Å3]

7211.6(3)

6332.4(2)

6139.70(15)

Z

2

4

4

ρcalcd [g cm-3]

1.450

1.469

1.380

 [mm-1]

0.644

4.974

4.773

F(000)

3208.0

2868.0

2626.0

Crystal size [mm]

0.4 × 0.2 × 0.2

0.2 × 0.15 × 0.1

0.2 × 0.2 × 0.2

Radiation

MoKα (λ = 0.71073)

CuKα (λ = 1.54184)

CuKα (λ = 1.54184)

2θ range [°]

3.05 to 50

7.038 to 124.488

7.04 to 124.286

Coll. refl.

48907

21698

20543

Indep. refl.

25391

9907

9589

data/restraints/params

25391/73/1787

9907/18/831

9589/48/790

GOF on F2

1.063

1.065

1.046

R1/wR2 [I  2σ(I)]

0.0675 / 0.1682

0.0496/0.1225

0.0399/0.1176

R1/wR2 (all data)

0.0894 / 0.1801

0.0566/0.1283

0.0451/0.1228

Largest peak/hole [e Å-3]

1.71/-0.91

1.07/-1.17

1.38/-0.69

Table S3. Wiberg bond indices of model metallapentalenes Os-1, Os-2, Ru-1, Ru-2 and 2′, related to Figure 3. M–C1/ Compound

M–C7

M–C4

M–Cl/

C1–C2/

C2–C3/

C3–C4/

M–CO

C6–C7

C5–C6

C4–C5

Os-1

1.02

0.85

0.62

1.48

1.39

1.36

Os-2

0.96

0.72

1.30

1.51

1.36

1.41

Ru-1

0.91

0.77

0.73

1.50

1.38

1.38

Ru-2

0.86

0.69

1.11

1.54

1.35

1.41

2′

0.76

0.74

0.60

1.56

1.33

1.41

Transparent Methods General Information. All syntheses were performed under an N2 atmosphere using standard Schlenk techniques unless otherwise stated. Hexane, tetrahydrofuran and diethyl ether solvents were distilled from sodium/benzophenone. Dichloromethane was distilled over calcium hydride under N2 prior to use. Triynes 1a and 1b (purchased from J&K® Chemical) and other reagents were used as received from commercial sources without further purification. Column chromatography was performed on silica gel (200-300 mesh) in air. Nuclear magnetic resonance (NMR) spectroscopy was performed using a Bruker Advance II 400 spectrometer, a Bruker Advance III 500 spectrometer or a Bruker Ascend III 600 spectrometer at room temperature. The 1H

and

13C

NMR chemical shifts (δ) are relative to tetramethylsilane, and the

31P

NMR chemical

shifts are relative to 85% H 3PO4. The absolute values of the coupling constants are given in hertz (Hz). Elemental analyses were performed on a Vario EL III elemental analyzer. Thermal gravimetric analyses (TGA) were carried out on an SDT Q600 at a heating rate of 10 °C min–1 from room temperature to 800 °C under an air atmosphere. Single-Crystal X-Ray Diffraction Experiments. Single-crystal X-ray diffraction data were collected on an Oxford Gemini S Ultra CCD area detector with graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) for 3 and 5 and mirror-monochromated Cu Kα radiation (λ = 1.54184 Å) for 2a, 6 and 7. An Agilent SuperNova Dual system with mirror-monochromated Cu Kα radiation (λ = 1.54184 Å) was used for 4. Absorption corrections were applied by using the program CrysAlis (Version 1.171.36.24 for 2a, 6 and 7, Version 1.171.39.46e for 3 and 5, and Version 1.171.37.35 for 4; multi-scan mode). Using Olex2 (Dolomanov et al., 2009), the structures were solved using the ShelXT (Sheldrick, 2015b) structure solution program using the intrinsic phasing method (2a, 3, 4, 5, 6 and 7), and all of the structures were refined with the ShelXL (Sheldrick, 2015a) refinement package using least-squares minimization. Non-H atoms were refined anisotropically unless otherwise stated. Hydrogen atoms were introduced at their geometric positions and refined as riding atoms unless otherwise stated. Single crystals suitable for X-ray diffraction were grown from a solution of ClCH2CH2Cl (2a), THF (3), or CH2Cl2 (4, 5, 6 and 7) layered with hexane. The THF solvent molecules in 3, some of the CH2Cl2 solvent molecules in 5, 6 and 7, and two of the BF4 counter anions in 5 were disordered and refined with suitable constraints. CCDC-1506335 (2a), CCDC-1506333 (3), CCDC-1848362 (4), CCDC-1848363 (5), CCDC-1848364 (6) and CCDC-1848365 (7) contain the supplementary crystallographic data for this paper. Further details on the crystal data, data collection, and refinements are provided in Tables S1 and S2. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Computational details. All structures were optimized at the B3LYP level of DFT (Becke, 1993; Lee et al., 1988; Miehlich et al., 1989). Frequency calculations were performed to identify all the stationary points as minima (zero imaginary frequency). In the B3LYP calculations, the effective core potentials (ECPs) of Hay and Wadt with a double-ζ valence basis set (LanL2DZ) (Hay and Wadt, 1985) were used to describe the Rh, Os, Ru, Cl, and P atoms, whereas the standard 6311++G** basis set was used for the C, O and H atoms. Polarization functions were added for Rh (ζ(f) = 1.350), Os (ζ(f) = 0.886), Ru (ζ(f) = 1.235), Cl (ζ(d) = 0.514) and P (ζ(d) = 0.340) (Huzinaga, 1984) in all calculations. All optimizations were performed with the Gaussian 09 software package (Frisch et al., 2013), whereas the Wiberg bond index (Wiberg, 1968) calculations were carried out with the NBO 6.0 program (Glendening, 2013) interfaced with the Gaussian 09 program. Nucleus-independent chemical shift (NICS) (Chen et al. 2005; FallahBagher-Shaidaei et al., 2006; Schleyer et al., 1996) values were calculated at the same level. The anisotropy of the induced current density (ACID) calculations was carried out with the ACID program (Geuenich et al. 2005). The energies (in kcal/mol) are given and include the zero-point energy corrections.

Experimental Procedures

Synthesis and characterization of complex 2a: Method A: A dichloromethane solution (5 mL) of 1a (1.03 g, 3.93 mmol) and a solution of HBF 4 (48 wt% solution in H2O, 1.20 mL, 9.25 mmol) were sequentially added to a red solution of RhCl(PPh 3)3 (3.04 g, 3.29 mmol) in dichloromethane (120 mL) slowly. The reaction mixture was stirred at room temperature for 10 min to yield a red solution. The solvent volume was reduced to approximately 15 mL under vacuum, and the mixture was purified by column chromatography on silica gel (eluent: dichloromethane/acetone = 20:1) to afford complex 2a as a red solid. Yield: 3.63 g, 88%. Method B: HBF4 (48 wt% solution in H2O, 0.24 mL, 1.9 mmol) was added to a suspension of complex 3 (800 mg, 0.67 mmol) in dichloromethane (30 mL). The reaction mixture was stirred at room temperature for 10 min to yield a red solution. The solvent volume was reduced to approximately 5 mL under vacuum, and

the

mixture

was

purified

by

column

chromatography

on

silica

gel

(eluent:

dichloromethane/acetone = 20:1) to afford complex 2a as a red solid. Yield: 771 mg, 91%. Method C: HBF4 (48 wt% solution in H2O, 83 μL, 0.64 mmol) was added to a suspension of complex 7 (250 mg, 0.214 mmol) in dichloromethane (8 mL). The reaction mixture was stirred at room temperature for 5 min to yield a red solution. The solvent volume was reduced to approximately 2 mL under vacuum, and the mixture was purified by column chromatography on silica gel (eluent: dichloromethane/acetone = 20:1) to afford complex 2a as a red solid. Yield: 250 mg, 93%. 1H NMR (600.1 MHz, CD2Cl2): δ 12.83 (s, 1H, C7H), 11.76 (d, J(HP) = 21.49 Hz, 1H, C1H), 7.91 (s, 1H, C3H), 6.89–7.82 (45H, Ph), 3.64 (s, 6H, COOCH3), 3.04 (s, 2H, C10H), 2.33 ppm (s, 2H, C8H). 31P NMR (242.9 MHz, CD2Cl2): δ 26.82 (d, J(PRh) = 105.77 Hz, RhPPh3), 9.72 ppm (s, CPPh3).

13C

NMR (150.9 MHz, CD2Cl2, plus

13C

DEPT-135, 1H-13C HSQC and 1H-13C

HMBC): δ 265.29 (br, C7), 239.66 (br, C1), 188.77 (s, C5), 188.66 (ddt, J(CRh) = 27.43 Hz, J(CP) = 25.42 Hz, J(CP) = 4.93 Hz, C4), 172.54 (s, C6), 171.10 (s, COOCH3), 157.38 (d, J(CP) = 25.91 Hz, C3), 134.14–135.39 (Ph), 131.86 (ddt, J(CP) = 55.99 Hz, J(CRh) = 4.31 Hz, J(CP) = 4.31 Hz, C2), 128.40–131.03 (Ph), 119.19 (d, J(CP) = 87.55 Hz, Ph), 63.73 (s, C9), 53.68(s, COOCH3), 39.83 (s, C8), 38.60 ppm (s, C10). Anal. Calcd (%) for C68H58BClF4O4P3Rh: C, 64.96; H, 4.65. Found: C, 65.04; H, 4.90.

Synthesis and characterization of complex 2b: 1b (180 mg, 1.23 mmol) and HBF4 (48 wt% solution in H2O, 0.36 mL, 2.8 mmol) were sequentially slowly added to a red solution of RhCl(PPh3)3 (1.04 g, 1.12 mmol) in dichloromethane (35 mL). The reaction mixture was stirred at room temperature for 10 min to yield a red solution. The solvent volume was reduced to approximately 5 mL under vacuum, and the mixture was purified by column chromatography on silica gel (eluent: dichloromethane/acetone = 20:1) to afford complex 2b as a red solid. Yield: 1.04 g, 81%. 1H NMR (600.1 MHz, CD2Cl2): δ 13.08 (s, 1H, C7H), 11.50 (d, J(HP) = 21.48 Hz, 1H, C1H), 8.07 (s, 1H, C3H), 6.88–7.80 (45H, Ph), 2.20 (m, 2H, C10H), 1.47 (tt, apparent quint, J(HH) = 7.31 Hz, J(HH) = 7.31 Hz, 2H, C9H), 1.36 ppm (t, J(HH) = 7.31 Hz, 2H, C8H). 31P NMR (242.9 MHz, CD2Cl2): δ 28.70 (dd, J(PRh) = 108.51 Hz, J(PP) = 5.87 Hz, RhPPh3), 9.14 ppm (t, J(PP) = 5.87 Hz, CPPh3).

13C

NMR (150.5 MHz, CD2Cl2, plus

13C-dept

135 , 1H-13C HSQC and 1H-13C

HMBC): δ 261.20 (br, C7), 239.86 (dt, J(CRh) = 35.44 Hz, J(CP) = 10.37 Hz, C1), 195.42 (s, C5), 187.62 (ddt, J(CRh) = 29.21 Hz, J(CP) = 25.20 Hz, J(CP) = 4.83 Hz, C4), 178.70 (s, C6), 157.32

(d, J(PC) = 26.58 Hz, C3), 130.20–134.95 (Ph), 129.41 (ddt, J(CP) = 58.75 Hz, J(CRh) = 8.59 Hz, J(CP) = 4.48 Hz, C2), 127.80–128.45 (Ph), 119.00 (d, J(CP) = 87.27 Hz, Ph), 32.50 (s, C8), 31.46 (s, C9), 28.42 ppm (s, C10). Anal. Calcd (%) for C64H54BClF4P3Rh: C, 67.36; H, 4.77. Found: C, 67.72; H, 4.89.

Synthesis and characterization of complex 3: A dichloromethane solution (5 mL) of 1a (786 mg, 3.00 mmol) was slowly added to a red solution of RhCl(PPh 3)3 (2.73 g, 2.95 mmol) in dichloromethane (90 mL). The reaction mixture was stirred at room temperature for 15 min to yield a green solution. The solvent volume was evaporated under vacuum to approximately 5 mL, and the mixture was washed with hexane (3 × 100 mL) to afford complex 3 as a yellow solid. Yield: 3.26 g, 93%. 1H NMR (600.1 MHz, CD2Cl2): δ 10.27 (d, J(HP) = 29.75 Hz, 1H, C1H), 6.87– 8.20 (45H, Ph), 6.53 (s, 1H, C7H), 3.57 (s, 3H, COOCH3), 3.54 (s, 3H, COOCH3), 3.45 (d, J(HH) = 8.95 Hz, 1H, C3H), 2.35 (d, J(HH) = 16.47 Hz, 1H, C10H), 2.27 (d, J(HH) = 16.47 Hz, 1H, C10H), 2.01 (d, J(HH) = 17.02 Hz, 1H, C 8H), 1.78 (d, J(HH) = 17.02 Hz, 1H, C8H), 0.17 ppm (d, J(HH) = 8.95 Hz, 1H, OH).

31P

NMR (242.9 MHz, CD2Cl2): δ 33.33 (ddd, J(PP) = 431.55 Hz, J(PRh) =

123.37 Hz, J(PP) = 5.47 Hz, RhPPh3), 31.05 (ddd, J(PRh) = 431.55 Hz, J(PP) = 123.37 Hz, J(PP) = 5.47 Hz, RhPPh3), 8.00 ppm (dt, J(PRh) = 11.00 Hz, J(PP) = 5.47 Hz, CPPh3). MHz, CD2Cl2, plus

13C

13C

NMR (150.9

DEPT-135, 1H-13C HSQC and 1H-13C HMBC): δ 221.27 (m, C1), 174.08 (s,

COOCH3), 173.27 (s, COOCH3), 167.43 (ddt, J(CRh) = 30.24 Hz, J(CP) = 22.91 Hz, J(CP) = 7.00 Hz, C4), 154.85 (s, C6), 153.49 (m, C7), 145.57 (s, C5), 127.32–136.42 (Ph), 122.91 (d, J(PC) = 46.59 Hz, C2), 122.89 (d, J(PC) = 84.89 Hz, Ph), 80.32 (d, J(PC) = 25.69 Hz, C3), 64.39 (s, C9), 52.90(s, COOCH3), 52.70 (s, COOCH3), 39.06 (s, C10), 36.01 ppm (s, C8). Anal. Calcd (%) for C68H59ClO5P3Rh: C, 68.78; H, 5.01. Found: C, 68.67; H, 5.23.

Synthesis and characterization of complex 4: A mixture of 2a (300 mg, 0.239 mmol) and 8hydroxyquinoline (104 mg, 0.716 mmol) was stirred at RT in dichloromethane (10 mL) for 30 min to yield a yellow solution. The solvent volume was reduced to approximately 5 mL under vacuum and the mixture was washed with Et2O (3 × 30 mL) to afford complex 4 as a yellow solid. Yield: 288 mg, 86%. 1H NMR (600.1 MHz, CD2Cl2):  9.96 (d, J(HP) = 18.51 Hz, 1H, C1H), 6.61–8.07 ppm (51H, Ph and C9H6NO), 4.21 (s, 1H, C3H), 3.71 (dd, J (HH) = 15.20 Hz, J (HP) = 15.20 Hz, 1H, C10H), 3.67 (s, 3H, COOCH3), 3.41(s, 3H, COOCH3), 2.09 (dd, J (HH) = 13.79 Hz, J (HP) = 13.79 Hz, 1H, C7H), 1.63 (s, 2H, C8H), 1.26 (dd, J (HH) = 14.32 Hz, J (HP) = 14.32 Hz, 1H, C7H), 0.55 (dd, J(HH) = 17.17 Hz, J(HP) = 9.55 Hz, 1H, C10H). (d, J(PRh) = 168.18 Hz, RhPPh3), 14.31 (s, 13C

C7PPh3),

NMR (150.9 MHz, CD2Cl2, plus 1H-13C HSQC and

31P

NMR (161.9 MHz, CD2Cl2):  24.65

6.75 ppm (d, J(PRh) = 11.24 Hz, C2PPh3). 13C-dept

135):  214.66 (m, C1), 178.74 (m,

C4), 172.62 (s, COOCH3), 170.50 (s, COOCH3), 168.79 (s, C9H6NO), 151.97 (d, J(CRh) = 15.41 Hz, C5), 126.91–143.4 (Ph and C9H6NO), 126.53 (d, J(CP) = 10.39 Hz, C3), 119.89 (s, C9H6NO), 119.45 (d, J(CP) = 87.55 Hz, Ph), 119.33 (d, J(CP) = 62.07 Hz, C2), 116.49 (d, J(CP) = 85.02 Hz, Ph), 115.00 (s, C9H6NO), 109.30 (d, J(CP) = 9.01 Hz, C6), 106.20 (s, C9H6NO), 55.21 (s, C9), 51.81 (s, COOCH3), 51.75 (s, COOCH3), 44.34 (s, C10), 42.30 (s, C8), 23.79 ppm (d, J(CP) = 53.03 Hz, C7). Anal. Calcd (%) for C77H65BClF4NO5P3Rh: C, 65.94; H, 4.67; N, 1.00. Found: C, 66.01; H, 4.57; N, 0.65.

Synthesis and characterization of complex 5: A mixture of 2a (450 mg, 0.358 mmol) and PPh3 (18.6 mg, 0.071 mmol) was stirred at RT in dichloromethane (15 mL) for 15 min to yield a yellow

solution. The solvent volume was reduced to approximately 5 mL under vacuum, and the mixture was washed with Et2O (3 × 30 mL) to afford complex 5 as a yellow solid. Yield: 401 mg, 85%. 1H NMR (400.1 MHz, CD2Cl2):  9.18 (d, J(HP) = 15.99 Hz, 2H, C1H), 6.36–7.84 (90H, Ph) , 4.30 (dd, J (HH) = 15.20 Hz, J (HP) = 15.20 Hz, 2H, C7H), 4.21 (s, 2H, C3H), 3.83(s, 6H, COOCH3), 3.53(s, 6H, COOCH3), 3.47 (br, 2H, C10H), 2.44 (d, J (HH) = 17.14 Hz, 2H, C8H), 2.07 (d, J (HH) = 17.14 Hz, 2H, C8H) , 1.49 (dd, J (HH) = 15.20 Hz, J (HP) = 15.20 Hz, 2H, C7H), 1.01 (dd, J (HH) = 16.77 Hz, J (HP) = 9.72 Hz, 2H, C10H).

31P

NMR (161.9 MHz, CD2Cl2):  25.54 (d, J(PRh) =

167.25 Hz, RhPPh3), 15.19 (s, C7PPh3), 7.64 ppm (d, J(PRh) = 12.0 Hz, C2PPh3). (100.6 MHz, CD2Cl2, plus

1H-13C

HSQC and

13C-dept

13C

NMR

135):  198.53 (dd, J(CP) = 31.44 Hz, J(CP)

= 13.77 Hz, C1), 176.80 (m, C4), 172.61 (s, COOCH3), 171.28 (s, COOCH3), 151.23 (d, J(CRh) = 9.70 Hz, C5), 130.07–135.91 (Ph), 129.66 (d, J(CP) = 21.83 Hz, C3), 127.83–128.26 (Ph), 122.78 (d, J(CP) = 67.52 Hz, C2), 119.10 (d, J(CP) = 88.40 Hz, Ph), 117.46 (d, J(CP) = 84.92 Hz, Ph), 113.99 (d, J(CP) = 9.61 Hz, C6), 56.59 (s, C9), 53.17 (s, COOCH3), 52.95 (s, COOCH3), 44.43 (s, C10), 43.28 ppm (s, C8), 26.04 ppm (d, J(CP) = 51.36 Hz, C7). Anal. Calcd (%) for C136H118B3Cl3F12O8P6Rh2: C, 61.90; H, 4.51. Found: C, 62.05; H, 4.48.

Synthesis and characterization of complex 6: A mixture of 2a (325 mg, 0.258 mmol) and Pyridine-N-Oxide (73.7 mg, 0.775 mmol) was stirred at RT in dichloromethane (10 mL) for 15 min to yield a blue solution. The solvent volume was reduced to approximately 3 mL under vacuum, and

the

mixture

was

purified

by

column

chromatography

on

silica

gel

(eluent:

dichloromethane/acetone = 20:1) to afford complex 6 as a blue solid. Yield: 269 mg, 82%. 1H NMR (400.1 MHz, CD2Cl2): δ 9. 73 (d, J(HP) = 16.11 Hz, 1H, C1H), 8.28 (s, 1H, C7H), 6.82–7.84 (45H, Ph), 6.21 (s, 1H, C3H), 3.74 (s, 6H, COOCH3), 2.86 (s, 2H, C8H), 2.74 ppm (s, 2H, C10H). 31P

NMR (161.9 MHz, CD2Cl2): δ 26.15 (d, J(PRh) = 108.92 Hz, OsPPh3), 8.25 ppm (d, J(PRh) =

14.05 Hz, CPPh3).

13C

NMR (100.6 MHz, CD2Cl2, plus

13C-dept

135, 1H-13C HSQC and 1H-13C

HMBC): δ 215.90 (m, C1), 193.06 (s, C7), 173.99 (m, C4), 171.16 (s, COOCH3), 157.65 (s, C5), 144.27 (d, J(CP) = 22.03 Hz, C3), 133.91–135.17 (Ph), 131.53 (s, C6), 128.30–131.00 (Ph), 123.15 (d, J(CP) = 72.70 Hz, C2), 119.14 (d, J(CP) = 88.78 Hz, C2), 56.20 (s, C9), 53.63 (s, COOCH3), 41.73 (s, C8), 39.35 ppm (s, C10). Anal. Calcd (%) for C68H58BClF4O5P3Rh: C, 64.14; H, 4.59. Found: C, 63.82; H, 4.40.

Synthesis and characterization of complex 7: Complex 2a (300 mg, 0.239 mmol) was dissolved in dichloromethane (5 mL) and put on a neutral alumina column. The red starting material turned a yellow color on the alumina surface. The yellow fraction was eluted with acetone, and the solvent was evaporated in vacuo to afford complex 7 as a yellow solid. Yield: 251 mg, 90%. 1H NMR (500.2 MHz, CD2Cl2):  10.18 (dd, J(HP) = 27.54 Hz, J(HP) = 3.30 Hz, 1H, C1H), 6.88–7.61 (45H, Ph), 6.53 (s, 1H, C7H), 5.77 (s, 1H, C3H), 4.68 (s, 1H, C8H), 3.51 (s, 6H, COOCH3), 2.29 (s, 2H, C10H).

31P

NMR (202.5 MHz, CD2Cl2):  32.27 (d, J(PRh) = 117.85 Hz,

RhPPh3), 3.89 ppm (s, CPPh3). Unfortunately,

13C

NMR characterization was failed because of

the poor solubility of 7. Anal. Calcd (%) C68H57ClO4P3Rh: C, 69.84; H, 4.91. Found: C, 69.88; H, 5.19.

Supplemental References Becke, A.D. (1993). Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 98, 5648–5652. Chen, Z., Wannere, C.S., Corminboeuf, C., Puchta, R., and Schleyer, P.v.R. (2005). NucleusIndependent Chemical Shifts (NICS) as an Aromaticity Criterion. Chem. Rev. 105, 3842–3888. Dolomanov, O.V., Bourhis, L.J., Gildea, R.J., Howard, J.A.K., and Puschmann, H. (2009). OLEX2: A Complete Structure Solution, Refinement and Analysis Program. J. Appl. Cryst. 42, 339–341. Fallah-Bagher-Shaidaei, H., Wannere, C.S., Corminboeuf, C., Puchta, R., and Schleyer, P.v.R. (2006). Which NICS Aromaticity Index for Planar π Rings Is Best? Org. Lett. 8, 863–866. Frisch, M. J., et al. (2013). Gaussian 09, Revision D.01 (Gaussian: Wallingford CT). Geuenich, D., Hess, K., Köhler, F., and Herges, R. (2005). Anisotropy of the Induced Current Density (ACID), a General Method to Quantify and Visualize Electronic Delocalization. Chem. Rev. 105, 3758–3772. Glendening, E.D., Badenhoop, J.K., Reed, A.E., Carpenter, J.E., Bohmann, J.A., Morales, C.M., Landis, C.R., and Weinhold, F. (2013). NBO 6.0 (Theoretical Chemistry Institute, University of Wisconsin, Madison, WI). Hay, P.J., and Wadt, W.R. (1985). Ab Initio Effective Core Potentials for Molecular Calculations. Potentials for K to Au Including the Outermost Core Orbitals. J. Chem. Phys. 82, 299–310. Huzinaga, S. (1984). Gaussian Basis Sets for Molecular Calculations (Elsevier, Amsterdam). Lee, C., Yang, W., and Parr, R.G. (1988). Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B 37, 785–789. Miehlich, B., Savin, A., Stoll, H., and Preuss, H. (1989). Results Obtained with the Correlation Energy Density Functionals of Becke and Lee, Yang and Parr. Chem. Phys. Lett. 157, 200–206. Schleyer, P.v.R., Maerker, C., Dransfeld, A., Jiao, H., and Hommes, N.J.R.v.E. (1996). NucleusIndependent Chemical Shifts: A Simple and Efficient Aromaticity Probe. J. Am. Chem. Soc. 118, 6317–6318. Sheldrick, G.M. (2015). Crystal Structure Refinement with SHELXL. Acta Cryst. C71, 3–8. Sheldrick, G.M. (2015). SHELXT-Integrated Space-Group and Crystal-Structure Determination. Acta Cryst. A71, 3–8. Wiberg,

K.B.

(1968).

Application

of

the

Pople-Santry-Segal

CNDO

Method

to

Cyclopropylcarbinyl and Cyclobutyl Cation and to Bicyclobutane. Tetrahedron 24, 1083–1096.

the