Article pubs.acs.org/Organometallics
Gas-Phase Investigations on the Transmetalation Step in Sonogashira Reactions Raphael J. Oeschger, David H. Ringger, and Peter Chen* Laboratorium für Organische Chemie, ETH Zürich, Vladimir-Prelog-Weg 2, 8093 Zürich, Switzerland S Supporting Information *
ABSTRACT: The microscopic reverse of the transmetalation step in the Pd/M (M = Cu, Ag, Au) catalyzed Sonogashiratype reactions has been observed in the gas phase upon collision-induced dissociation (CID) of the heterobimetallic complexes. Measuring the activation energies by quantitative energy-resolved CID experiments provides an upper bound for the internal rearrangement energies. The potential-energy surface is investigated by density functional theory calculations and compared to the experimental values.
■
INTRODUCTION Bimetallic Pd/M (M: Cu, Ag, Au) catalyzed C−C bondforming reactions have been investigated with high interest in recent years.1 A well-known example of such a transformation is the Sonogashira reaction,2 using as cocatalysts primarily Cu(I) or alternatively Ag(I)3 and Au(I)4 salts. The reaction mechanism of these transformations are still not very well understood. Nevertheless, a generally accepted mechanism for the linkage of the system consisting of a Pd- and a M-cycle is via a transmetalation step.5 In comparison to the more extensively studied Pd(0)/Pd(II) catalytic cycles, fewer reports are published on Pd/coinage metal acetylide transmetalation investigations. A better understanding of the mechanisms underlying the transmetalation step could lead to the design of more efficient Sonogashira and related cross-coupling reactions. Mechanistic studies by Osakada and Yamamoto6 revealed that migration of acetylides from Cu(I) to Pd(II) is reversible and passes through bimetallic intermediates. Espinet and coworkers7 reported crystal structures of such previously proposed intermediates and found coinage metal chlorides coordinating to the π-bond of the Pd-acetylide. Lei et al.8 reported the only quantitative kinetic study of the rate-limiting transmetalation step in the Pd/Cu-catalyzed Sonogashira catalytic reaction. Our group recently reported the gas-phase study of the transmetalation of a methyl group from Pt(II) to Cu(I)9 or Au(I),10 indicating that such a transformation occurs through the formation of bimetallic Pt−Cu/Au (d8−d10) bonds. In accordance, Espinet et al.11 and Hashmi et al.12 reported similar findings for the Pd(II)/Au(I)-catalyzed cross-coupling reactions using density functional theory (DFT) calculations; their computed intermediates and transition states suggest that the Pd−Au distances are below the sum of their respective van der Waals radii. In addition, Hashmi et al. showed that transmetalation from Au to Pd needs simultaneous transfer of a halide from © XXXX American Chemical Society
Pd to Au to be thermodynamically feasible. In order to further elucidate the mechanism of the Pd/coinage metal transmetalation steps, we report herein the mass spectrometric study on the microscopic reverse13 of Sonogashira-type transmetalations.
■
RESULTS The cationic heterobimetallic intermediate complexes 4a−c, Scheme 1, were formed in solution by mixing equimolar amounts Scheme 1. Generation of Cations 4a−c
of Pd(II) pincer complex 114 with MBF4 (M = CuI (2a), AgI (2b), AuI (2c)) salts and triphenylphosphine 3 in acetonitrile. Pincer complex 1 was chosen instead of a nonchelating transacetylidearyldiphosphine Pd(II) as model intermediate, as it is assumed to have very similar electronic and steric properties but is not capable of releasing arylacetylene by reductive elimination, making the complex more stable. The resulting charged bimetallic complexes 4a−c were then analyzed in the gas phase by means of ESI-MS/MS spectrometer techniques. The presumably formed bimetallic complexes 4a−c were identified by their m/z ratio, and the measured isotope patterns were found to be in good agreement with calculated patterns, which further confirms their identity (see Figure 3 for computational results). Under collision-induced Received: June 8, 2015
A
DOI: 10.1021/acs.organomet.5b00491 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
(“tight” or “loose”) is a required input. A “tight” transition state model is used when the rate-limiting TS is an intramolecular rearrangement that connects two local minima on the potentialenergy surface prior to dissociation. A “loose” transition state model is appropriate when the rate-limiting TS is a simple dissociation without reverse activation barrier. On the basis of chemical arguments and DFT calculations (Figure 3), it was found that the loose TS model is appropriate for all of the reaction pathways observed for the three bimetallic complexes, 4a−c. DFT modeling of all three bimetallic complexes along the potential-energy surface has revealed that the barriers of the intramolecular rearrangements are much lower in energy than those for the subsequent dissociation. The measured activation barriers of the transmetalation pathways (B) were found to be 47.0 ± 1.0 kcal/mol for copper, 52.8 ± 1.8 kcal/mol for silver, and 37.7 ± 1.0 kcal/mol for gold. The measured values are intriguing, as they suggest similar reactivity for Cu and Ag, but a substantially lower reaction barrier for Au. Such an anomaly within the coinage metals is most likely due to relativistic effects on gold and have previously been observed in the investigation of the bonding energies of coinage metal nitrenes15g and the ring opening of dimethoxycyclopropane with Cu, Ag, and Au.15d The measured activation energies for the loss of the PPh3 ancillary ligand (A) were found to be 43.0 ± 1.1 kcal/mol for Cu and 42.5 ± 1.5 kcal/mol for Ag. For a more detailed mechanistic picture, the observed gasphase reactivities were modeled by DFT calculations at the M06-L/TZP//M06-L/SDD(d,p) level of theory as implemented in ADF16 and the Gaussian 0917 suite. All of the computed energies were zero-point energy and basis set superposition error corrected. For our investigation of the transmetalation (B) and ligand loss (A) processes, we used the M06-L functional, as it has been designed to describe transition-metal thermochemistry, kinetics, and attractive dispersion effects between dissociating fragments. Additionally, previously reported gas-phase and DFT studies have shown that M06-L models the experimentally obtained bond dissociation energies well. The calculations, however, overestimate the transmetalation energy by 2−11 kcal/mol (Figure 2 and Figure 3). The most stable computed bimetallic Pd/Cu and Pd/Ag complexes were found to be the π-coordinated intermediates 4a and 4b. However, in the case of the bimetallic Pd/Au system, complex 5c, with a trigonal carbon, was found to be the lowest lying structure. This phenomenon could be due to the formation of favorable Pd−Au (d8−d10) interactions. Despite these geometric differences, for all three bimetallic complexes the loss of the ancillary PPh3 ligand and the Pd decoordination were found to be the highest TS barriers on the presented potentialenergy surface. Taking into account that the only product ion observed in the gas phase for the transmetalation step (B) is the charged [Pd] complex, in principle, the transmetalation pathways could also proceed via the loss of the ancillary PPh3 ligand prior to transmetalation. Such a stepwise reaction pathway was found to be unlikely, as it is energetically highly unfavorable and does not correlate to our experimentally measured thresholds.
Figure 1. CID spectrum of ions 4a,b,c at 0.2 mTorr and collision offset 90 V. PPh3 loss (A) is represented by black arrows; transmetalation (B), by red arrows. Inset: Experimental (black) and calculated (red) isotope patterns of ions 4a,b,c.
Figure 2. Zero-pressure extrapolated cross sections (circles: pathway A, triangles: pathway B) with L-CID-fitted curves (lines). Inset: Table with experimental and calculated values in kcal/mol.
dissociation (CID) conditions, 4a−c furnished two signals (Scheme 2, Figure 1). The first (A) corresponds to loss of neutral PPh3, and the second (B) to the transmetalation of the phenylacetylide from Pd(II) to M(I) to form the coordinative unsaturated [Pd(PCP)]+ 7 and (PPh3)M−CCPh 8a−c. The signal stemming from PPh3 ligand loss (A) is observed only for Cu and Ag but not for Au, most likely due to the relatively strong Au−P bond. The detection of no more than two competing dissociation proccesses stemming from the same reaction ion renders these bimetallic complexes suitable for twochannel energy-resolved threshold experiments. The activation energies for the PPh3 ligand dissociation (A) and transmetalation (B) processes were determined using a customized Finnigan MAT TSQ-700 ESI-MS/MS spectrometer. The energy-resolved CID experiments were performed by recording the reactant and product ion intensities at different CID gas pressures (20 to 110 μTorrr) as a function of collision offset. The data curves were extrapolated to zero pressure and subsequently fitted with the L-CID15 program. In order to treat the kinetic shift within L-CID, the transition state (TS) type
■
DISCUSSION Except for the M−acetylide dissociation, the overall energy surface is computed to be very flat, even though strong Pd−C bonds are broken and new M−C bonds are formed. We postulate that the low reaction barriers for the transmetalation are due to favorable Pd−coinage metal interactions. B
DOI: 10.1021/acs.organomet.5b00491 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
Figure 3. Calculated zero-point energy corrected potential-energy surface (M06-L/TZP//M06-L/SDD(d,p)) in kcal/mol for reaction pathways A and B. For clarity only structures of Pd/Au are shown and hydrogen atoms are excluded (red: Pd/Au, blue: Pd/Ag, green: Pd/Cu).
we see that the internal σ/π-rearrangementthe actual M−C bond breaking, Pd−C bond forming step from 6a−c to 4a−c is not likely to be rate-limiting in the Sonogashira reaction, although transmetalation was claimed to be rate-limiting in solution. We suggest that there is an explanation consistent with our gas-phase results, as well as the solution-phase kinetics by Lei.8 We suggest that it is rather the preceding alkyne coordination step to form 6a−c that limits the overall rate of the reaction in solution; this ligand exchange is part of the overall transmetalation step in Lei’s mechanism (Figure 4).
Scheme 2. Reaction Pathways of 4a−c upon CID with Ar
Judging from the computed atom distances between Pd and M, the interactions are strongest in the intermediates 5a−c and TSa−c that connect the transmetalation reactant 4a−c with the products 6a−c. The shortest Pd−M distances were found to be Pd−Au 2.87 Å (5c), Pd−Ag 2.91 Å (5b), and Pd−Cu 2.68 Å (5a) and thus only slightly more than the sum of the covalent radii (Pd/Au: 2.75 Å, Pd/Ag: 2.84 Å, Pd/Cu: 2.71 Å).18 DFT modeling is known to have limitations in treating metal−metal interactions accurately, and therefore the computed activation and intermediates energies for the transmetalation processes need to be considered with caution. The experimentally measured energies for pathways B give an upper bound for the transmetalation TSa−c. The computed potential surfaces indicate that we most likely measure only the barriers for the “loose” dissociation pathways. A preceding “tight” TSa−c19 rearrangement20 would have to be at least 15−20 kcal/mol lower in energy than the asymptote for dissociation (with a “loose” transition state) for it to become slower than the subsequent dissociation and thus observable during CID experiments as the rate-determining transition state.15c This provides an upper limit of 27−32 kcal/mol for Pd/Cu, 33−38 kcal/mol for Pd/Ag, and 18−23 kcal/mol for Pd/Au. By microscopic reversibility, the transition state in the elementary transmetalation step in the Sonogashira reaction is the same as the one involved in our gas-phase experiment, but simply in the opposite direction. From the calculated barriers
Figure 4. Catalytic cycle of the Sonogashira reaction with oxidative addition (OA), transmetalation (TM), and reductive elimination (RE). Below TM is demerged into its three main steps (LE: ligand exchange).
While the association of an incoming alkyne to a coordinatively unsaturated Pd center is barrierless in the gas phase, it would proceed by associative displacement of a ligand (solvent, halide, ...) on Pd in solution and, hence, would require a higher activation energy.21 Lei measured an enthalpy of activation of 8.5 kcal/mol for the transmetalation,8 which is in the range of what has been reported for associative ligand substitution on square-planar Pd(II) centers. The measured values vary C
DOI: 10.1021/acs.organomet.5b00491 Organometallics XXXX, XXX, XXX−XXX
Organometallics
■
ACKNOWLEDGMENTS We are grateful to Ilia J. Kobylianskii and Armin Limacher for help with the instruments. Financial support from the ETH Zürich and the Swiss National Science Foundation is also acknowledged.
from 4 to 16 kcal/mol, depending on the nucleophilicity of the incoming ligand and degree of steric hindrance at the Pd(II) center, corresponding to a wide range of ligands and complexes.22 Lei’s value would be rather typical. It is noteworthy that a rate-limiting associative ligand exchange would also be in agreement with Lei’s Hammett correlation plot: Electronaccepting substituents on the aryl cause stronger trans effects and therefore accelerate the reaction.
■
■
EXPERIMENTAL SECTION
■
ASSOCIATED CONTENT
Pd(II) pincer complex 1 was prepared as reported in the literature.14 Spray solutions of Pd/Cu were prepared by mixing (PCP)PdCCPh14 (0.6 mg, 1 μM) with Cu(MeCN)4BF4 (0.3 mg, 1 μM) and PPh3 (0.2 mg, 1 μM) in MeCN (0.1 mL) in a vial in the glovebox. After 1 min 10 μL of the solution was added to 2 mL of MeCN and used immediately. Solutions of Pd/Ag were prepared similarly but with AgBF4 (0.2 mg, 1 μM) instead of Cu(MeCN)4BF4. For Pd/Au solutions, Ph3PAuCl (0.5 mg, 1 μM) was mixed with AgBF4 (0.2 mg, 1 μM) in 0.1 mL of MeCN. After 5 min (PCP)PdCCPh (0.6 mg, 1 μM) was added, and 10 μL of the solution was added to 2 mL of MeCN. Qualitative gas-phase studies were done on a Thermo-Finnigan TSQ Quantum instrument. For quantitative measurements a customized Finnigan MAT TSQ-700 was used as described previously.15h Activation energies were extracted from our L-CID program by fitting the cross-section data.15a Full geometry optimizations were done using Gaussion 0917 with the M06-L density functional and D95(d,p);Pd,Cu,Ag,Au:SDD basis sets. Frequency calculations were performed to obtain zero-point energy values and for confirmation of the nature of each stationary point. Basis set superposition error (BSSE) calculations were done using the “counterpoise” keyword. Single-point energies were computed with the ADF suite16c using the M06-L functional and TZP basis sets (see the SI for more details). S Supporting Information *
Sample preparation, computational details, energies and geometries, experimental distribution of ion kinetic energies, mass spectra, including energy-resolved collision-induced dissociation experimental data. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.organomet.5b00491.
■
REFERENCES
(1) (a) Pérez-Temprano, M. H.; Casares, J. A.; Espinet, P. Chem. Eur. J. 2012, 18, 1864−1884. (b) Hirner, J. J.; Shi, Y.; Blum, S. A. Acc. Chem. Res. 2011, 44, 603−613. (2) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975, 16, 4467−4470. (3) (a) Weibel, J.-M.; Blanc, A.; Pale, P. Chem. Rev. 2008, 108, 3149− 3173. (b) Létinois-Halbes, U.; Pale, P.; Berger, S. J. Org. Chem. 2005, 70, 9185−9190. (c) Halbes, U.; Bertus, P.; Pale, P. Tetrahedron Lett. 2001, 42, 8641−8644. (d) Bertus, P.; Halbes, U.; Pale, P. Eur. J. Org. Chem. 2001, 2001, 4391−4393. (e) Bertus, P.; Pale, P. Tetrahedron Lett. 1996, 37, 2019−2022. (4) (a) Panda, B.; Sarkar, T. K. Synthesis 2013, 45, 817−829. (b) Panda, B.; Sarkar, T. K. Chem. Commun. 2010, 46, 3131−3133. (c) Panda, B.; Sarkar, T. K. Tetrahedron Lett. 2010, 51, 301−305. (d) Lauterbach, T.; Livendahl, M.; Rosellon, A.; Espinet, P.; Echavarren, A. M. Org. Lett. 2010, 12, 3006. (e) Jones, L. A.; Sanz, S.; Laguna, M. Catal. Today 2007, 122, 403−406. (5) (a) Bakherad, M. Appl. Organomet. Chem. 2013, 27, 125−140. (b) Chinchilla, R.; Najera, C. Chem. Soc. Rev. 2011, 40, 5084−5121. (c) Marsden, J. A.; Haley, M. M. Cross-Coupling Reactions to sp Carbon Atoms. In Metal-Catalyzed Cross-Coupling Reactions; WileyVCH Verlag GmbH, 2008; pp 317−394. (d) Doucet, H.; Hierso, J.-C. Angew. Chem., Int. Ed. 2007, 46, 834−871. (e) Chinchilla, R.; Nájera, C. Chem. Rev. 2007, 107, 874−922. (f) Brandsma, L. 16 - Transition Metals-Catalysed Couplings of Acetylenes with sp2-Halides. In Synthesis of Acetylenes, Allenes and Cumulenes; Brandsma, L., Ed.; Academic Press: Oxford, 2004; pp 293−317. (g) Tykwinski, R. R. Angew. Chem., Int. Ed. 2003, 42, 1566−1568. (h) Sonogashira, K. J. Organomet. Chem. 2002, 653, 46−49. (i) Sonogashira, K., 2.4-Coupling Reactions Between sp2 and sp Carbon Centers. In Comprehensive Organic Synthesis; Fleming, B. M. T., Ed.; Pergamon: Oxford, 1991; pp 521−549. (6) (a) Osakada, K.; Yamamoto, T. Coord. Chem. Rev. 2000, 198, 379−399. (b) Osakada, K.; Sakata, R.; Yamamoto, T. Organometallics 1997, 16, 5354−5364. (c) Osakada, K.; Takizawa, T.; Yamamoto, T. Organometallics 1995, 14, 3531−3538. (7) Meana, I.; Espinet, P.; Albéniz, A. C. Organometallics 2014, 33, 1−7. (8) He, C.; Ke, J.; Xu, H.; Lei, A. Angew. Chem., Int. Ed. 2013, 52, 1527−1530. (9) Moret, M.-E.; Serra, D.; Bach, A.; Chen, P. Angew. Chem., Int. Ed. 2010, 49, 2873−2877. (10) Serra, D.; Moret, M.-E.; Chen, P. J. Am. Chem. Soc. 2011, 133, 8914−8926. (11) Pérez-Temprano, M. H.; Casares, J. A.; deLera, Á . R.; Á lvarez, R.; Espinet, P. Angew. Chem., Int. Ed. 2012, 51, 4917−4920. (12) Hansmann, M. M.; Pernpointner, M.; Döpp, R.; Hashmi, A. S. K. Chem. - Eur. J. 2013, 19, 15290−15303. (13) The microscopic reverse is investigated because it is experimentally more convenient. The outcome is not dependent on the direction because in both ways (forward or backward) the same TSs are accessed. In particular, it is easier to work with Pd acetylides than the corresponding Cu acetylides, which are generally explosive. (14) Moulton, C. J.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1976, 1020−1024. (15) (a) Narancic, S.; Bach, A.; Chen, P. J. Phys. Chem. A 2007, 111, 7006−7013. (b) Ringger, D. H.; Kobylianskii, I. J.; Serra, D.; Chen, P. Chem. - Eur. J. 2014, 20, 14270. (c) Couzijn, E. P. A.; Kobylianskii, I. J.; Moret, M.-E.; Chen, P. Organometallics 2014, 33, 2889−2897. (d) Batiste, L.; Chen, P. J. Am. Chem. Soc. 2014, 136, 9296−9307. (e) Kobylianskii, I. J.; Widner, F. J.; Kräutler, B.; Chen, P. J. Am. Chem.
CONCLUSIONS In summary, the microscopic reverse of the transmetalation step in Sonogashira reactions has been investigated experimentally in the gas phase and theroretically by DFT calculations. The measured activation barriers of ligand dissociation and transmetalation/Pd dissociation were somewhat overestimated by DFT calculations. The optimized intermediates showed strong heterobimetallic bonding interactions, which appear to be responsible for the overall low rearrangement barriers. This feature might be applicable to other bimetallic transition-metal-catalyzed reactions involving a transmetalation step and potentially gives access to more efficient catalysts. This study serves as continuation of the exploration of such favorable metal−metal dispersion interactions, which are currently investigated within our research group, and this work is useful for benchmarking future endeavors in this field.
■
Article
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest. D
DOI: 10.1021/acs.organomet.5b00491 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics Soc. 2013, 135, 13648−13651. (f) Fedorov, A.; Batiste, L.; Bach, A.; Birney, D. M.; Chen, P. J. Am. Chem. Soc. 2011, 133, 12162−12171. (g) Fedorov, A.; Couzijn, E. P. A.; Nagornova, N. S.; Boyarkin, O. V.; Rizzo, T. R.; Chen, P. J. Am. Chem. Soc. 2010, 132, 13789−13798. (h) Couzijn, E. P. A.; Zocher, E.; Bach, A.; Chen, P. Chem. - Eur. J. 2010, 16, 5408−5415. (16) (a) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra, C.; van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931−967. (b) Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor. Chem. Acc. 1998, 99, 391− 403. (c) ADF2012, S; Theoretical Chemistry, Vrije Universiteit, Amsterdam: The Netherlands, http://www.scm.com. (17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J.; Brothers, E. N.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Gaussian, Inc.: Wallingford, CT, USA, 2009. (18) Bondi, A. J. Phys. Chem. 1964, 68, 441−451. (19) Calculated TSa−c are lower in energy than intermediates 6a−c. Geometry optimizations were done at a lower level of accuracy, M06L/SDD(d,p), followed by more precise single points at M06-L/TZP. Such an approach leads to minor inconsistencies due to the distinct flatness of the potential-energy surface. (20) TSa−c are tight in the sense that there are not two degrees of freedom that become rotations as there would be in a loose TS. Even if there are two degrees of freedom that have exceedingly low vibrational frequencies, it does not change our conclusions because it would be even more unlikely that the TS for the actual elementary transmetalation step is rate-limiting. (21) To investigate the activation energy of ligand displacement on Pd(II) by acetylides, one could think of measuring gas-phase energies for decoordination of solvent molecules as a comparison. However, this would not be conclusive because (a) ligand exchange in d8 squareplanar complexes is usually associative and (b) dissociation limits are much higher in the gas phase than in solution due to solvation effects, electrostatic screening, and dispersion effects. (22) (a) Hallinan, N.; Besancon, V.; Forster, M.; Elbaze, G.; Ducommun, Y.; Merbach, A. E. Inorg. Chem. 1991, 30, 1112−1114. (b) Hellquist, B.; Elding, L. I.; Ducommun, Y. Inorg. Chem. 1988, 27, 3620−3623. (c) Roulet, R.; Gray, H. B. Inorg. Chem. 1972, 11, 2101− 2104.
E
DOI: 10.1021/acs.organomet.5b00491 Organometallics XXXX, XXX, XXX−XXX
