Article pubs.acs.org/Organometallics
Density Functional Study of Oxygen Insertion into Niobium− Phosphorus Bonds: Novel Mechanism for Liberating P3− Synthons Glenn R. Morello*,† and Thomas R. Cundari‡ †
Centre for Theoretical and Computational Chemistry (CTCC) and Department of Chemistry, University of Tromsø, N-9037 Tromsø, Norway ‡ Department of Chemistry, Center for Advanced Scientific Computing and Modeling (CASCaM), University of North Texas, P.O. Box 305070, Denton, Texas 76203-5070, United States S Supporting Information *
ABSTRACT: We explore the mechanism of oxygen insertion into niobium−phosphorus bonds to liberate synthetically relevant, phosphorus-containing molecules. Oxygen insertion mechanisms generally proceed through either direct oxygen insertion from an oxo ligand, MO (oxy-insertion), or an insertion of an oxygen atom from an external oxidant, OY (Baeyer−Villiger, BV). Computational methods were employed to elucidate the preferred mechanism for the liberation of the phosphorus moiety from [(η2-P3)Nb(ODipp)3] (Dipp = 2,6-iPr2C6H3, P3 = P3-SnPh3) when treated with pyridine-N-oxide as an external oxidant. Careful analysis of conformational isomers and energies clearly suggests that the BV mechanism is the preferred pathway toward phosphorus liberation. Once free, the P3 moiety can react with 1,3-cyclohexadiene to form the Diels−Alder product, which is also modeled in the computational study.
1. INTRODUCTION The functionalization of molecules by transition-metal-assisted oxygen atom transfer is one of the most important reactions for the production of organic molecules. Recent applications utilize transition-metal-oxo (MO) complexes to catalytically functionalize organic compounds.1−6 Understanding the energetic requirements of metal−oxo bond formation is fundamental in the development of molecules suited for active oxygen atom transfer. Work from Holm and Donahue has outlined a reactivity scale for oxygen atom transfer on the basis of thermodynamics for a large number of molecules including organic small molecules, inorganic compounds, and transition metal complexes.7 Computational methodologies have provided valuable insight into the mechanistic details and overall energetics of oxygen atom transfer with transition metal complexes.8−10 Lee and Holm have compiled a large database of reaction enthalpies for oxygen atom transfer, using density functional theory (DFT), with various basis sets, G3 methodologies for non-transition-metal reactions, and DFT for transition metal inclusion, that compare well with experimental results.7,8,11 Calculations by Gisdakis and Rösch provided insight into the effect L (L = Cp, CH3, Cl, O) has on reaction energies, bond dissociation energies, and bond lengths of the transition states for [2+3] cycloaddition of ethylene to metaloxos, LMO3 (M = Mo, W, Mn, Tc, Re, and Os).9 Structural characterization of transition-metal-oxo complexes and analysis of their reactivity and transition states have been well represented in the literature.9,12−18 For example, previous work by Goddard et al. explored the involvement of transitionmetal-oxo complexes in olefin metathesis and alkene © XXXX American Chemical Society
epoxidation, noting that spectator oxo ligands are influential in the thermodynamics of insertion reactions.19,20 In addition, Ziegler has computationally explored C−H and O−H bond activation by transition-metal (TM)-oxo complexes of groups 5−8 and rationalized that the difference in energetic trends of TM-d0 oxo complexes, in abstraction and addition reactions, is related to the radial extension of the metal d orbital.10 Much research in the realm of oxidation of organic molecules has centered on the cytochrome P450 family of enzymes and their inorganic mimics.21−23 Newcomb has focused on the heme iron center of P450 and has deduced a variety of mechanistic pathways available for hydroxylation of molecules.24 The efficiency with which the heme iron performs oxygen atom transfer (OAT) has allowed many investigators to design metalloporphyrins as catalysts for oxidation reactions. Specifically, Chiavarino et al. have explored oxo-Fe(IV) porphyrin complexes for the oxidation of amines and have gained valuable insight into the intermediates along the reaction path.23 Work by Shaik and collaborators have provided computational insight into heme and nonheme iron complexes for catalytic epoxidation and hydroxylation of small organic molecules.22,25,26 More recently, groups have developed metalloporphyrin-oxo complexes, inspired by the P450 enzyme reactive center, for functionalization of organic molecules that utilize different transition metal centers such as Mn, Mo, Co, and Cr.27−30 Received: August 26, 2016
A
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
group (R).39 While the transfer group was found to be the biggest contributor to overall kinetics of the reaction, varying the oxidant was also determined to impact the kinetic barriers, and oxidants with the best leaving group provided more accessible oxygen insertion pathways.39 Moreover, most investigations into oxygen atom insertion have used mid to late transition metal complexes,4,5,20,35,37,40 while reactions with early transition metals remain rare. Recently, Mei and coworkers have identified different pathways for a tungsten complex, Cp*W(O)2(CH2SiMe3), which show dependence on the presences of acid or base in the reaction conditions.41 Beyond the functionalization of organic molecules, Baeyer− Villiger oxygen insertion pathways may be exploited for liberation of a ligand from a transition metal complex to serve as a synthon for chemical transformations. Direct insertion of the oxygen atom into a metal−ligand bond can generate a metastable M−O−L species that proceeds in forming the MO complex by loss of the ligand fragment, L. The reorganization of the M−O−L to generate the MO complex is dependent on the nature of the metal and ligands, i.e., high- or low-valent metal, stable or labile ligand. In particular, forming a strong MO bond with early, high-valent transition metals can serve as the driving force for the formation of the metal-oxide (MO) and liberation of the synthon.13,32 As an example, consider the liberation of phosphorus-containing ligands, specifically the P3− moiety, from such stable oxo complexes as a direct route for phosphorus transfer reactions to production of phosphoruscontaining molecules.42 Previous work from the Cummins group explored the reactivity of white phosphorus, P4, with a niobium complex as a method for generating such phosphorus cluster ligands suited for transfer reactions.43−48
Oxidations of organic moieties by transition metal oxides have been widely discussed in the literature.4,5,10,19,20,31 Complexes containing oxo ligands (typically depicted as either LnMO or LnMO) are employed for oxygen atom insertion where the oxo ligand functionalizes the coligand (typically an alkyl or aryl) and inserts into the M−R bond, Scheme 1.4,5,32,33 Scheme 1. Reaction Mechanisms for Transition-MetalMediated Oxygen Insertion into an M−R Bond: (Top) O Atom Insertion by Metal-Oxo Complex (oxy); (Bottom) Metal-Mediated O Atom Insertion by External Oxidant (Baeyer-Villiger, BV)a
a
OY represents the external oxidant.
Brown and Mayer studied the utility of [TpRe(O)(Ph)]+ (Tp = hydrido-tris(pyrazolyl)borate complexes) for the oxidation of organic molecules and proposed an intramolecular description of the migration of phenyl to the oxo ligand without the presence of radicals.4,5 The proposed mechanism uses an external oxidant to oxidize the metal center, forming the [TpRe(O)2(Ph)]+ intermediate, followed by phenyl group transfer to the new oxo ligand, resulting in a [TpRe(O)(OPh)]+ complex.34 Further studies by Mayer explored transition-metal-oxo complexes (Re, Mn, Cu) for oxidizing C−H bonds, which led to the surprising conclusion that unpaired spin density on the metal is not required for hydrogen atom abstraction.33 In contrast to the number of reports for oxidation by metaloxo complexes, reports of metal-mediated oxygen insertion with an external oxidant (YO) are rare, Scheme 1.2,35−37 Work from Periana’s group has shown that rhenium trioxides, XReO3 (X = CH3, Ph), in the presence of H2O2, PyO, ON(CH3)3, DMSO, IO4−, and PhIO proceed through a Baeyer−Villiger (BV)-type mechanism to produce methanol in good yield.2,36−38 Highlighted in the mechanism is the direct insertion of oxygen from an external oxidant into the Re−C bond while the initial Re−O bonds act as spectators.2,37 DFT calculations show that oxygen insertion from Re−O bonds has activation barriers almost 3 times as high as the external oxidant insertion.38 Similarly, Goddard and Periana have also examined rhenium carbonyl complexes, which in the presence of catalytic amounts of oxidant, SeO2, produce methanol through a concerted oxygen atom insertion mechanism.1 Although organometallic BV can serve as an atom-economical method for organic molecule functionalization, mechanistic details are still vague and further work is needed to resolve the key factors in the oxygen atom insertion process with TMs and external oxidants.6 Work in the Cundari group has explored oxygen atom insertion into platinum−carbon bonds for the production of methanol from methane.39 Thermodynamic studies indicate a nonredox pathway that is reminiscent of a BV-type mechanism (Scheme 1) for the inserting oxygen atom and show dependence on the nature of the oxidant and transfer
Scheme 2. (Top) Synthetically Characterized Reaction That Converts [(η3-P3)Nb(ODipp)3], 1-Dipp, into [(η2P3SnPh3)Nb(ODipp)3], 2-Dipp;47 (Bottom) Computationally Modeled Complexes for This Study Employ the Smaller OMe Ligands (1, 2) in Place of the Larger ODipp Ligandsa
a
Dipp = 2,6-iPr2C6H3.
A reagent for phosphorus group transfer reactions is [(η3P3)Nb(ODipp)3]−Na+, 1-Dipp, Scheme 2 (Dipp = 2,6i Pr2C6H3). Treatment of 1-Dipp with Ph3SnCl yields complex 2-Dipp, [(η2-P3SnPh3)Nb(ODipp)3], which upon the addition of pyridine-N-oxide, releases a Ph3SnP3 fragment, Scheme 3, that participates in a Diels−Alder reaction with 1,3-cyclohexadiene.47 This work employs computational chemistry methods to gain insight into the mechanism of release of the Ph3SnP3 fragment, B
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
Scheme 3. Proposed Mechanism for Oxidation of [(η2-P3SnPh3)Nb(OMe)3], 2, with Pyridine-N-Oxide (Oxidant) Liberating the Reactive Ph3SnP3 Fragment, 8a
Compound 8 can be trapped by 1,3-cyclohexadiene in a Diels−Alder reaction, orange box, and subsequently used as a transfer reagent for the P3− fragment. a
coordinates. Triple-zeta basis set corrections employed the CEP-121G basis set55 on Nb and Sn atoms and the 6-311G(d,p) basis set58,59 on all other atoms. The empirical dispersion correction of Grimme et al., GD3,60 was employed to account for dispersion effects, while the IEFPCM solvation model,61 using THF as the solvent, was used to better represent reaction conditions. Thermochemistry was determined at 298.15 K and 1 atm utilizing B3LYP/CEP-31G(d) unscaled vibrational frequencies.
Scheme 3, and its subsequent reaction with a diene. Reaction coordinates are compared to experimental results, and calculated thermodynamic and kinetic profiles are discussed. Specifically, detailed analysis of oxidant activation and insertion mechanisms is presented to gain insight into the proposed mechanistic pathways (Scheme 3) toward generation of the experimentally observed niobium(V)-oxo, 7, and, when using 1,3-cyclohexadiene as a chemical trap, the Diels−Alder product, Ph3SnP3-C6H8, 9. The expelled Ph3SnP3-C6H8 molecule is both nucleophilic at the phosphorus and capable of losing the neutral 1,3-cyclohexadiene when reacted with an appropriate P3− acceptor, such as electron-deficient transition metal complexes, or main group atoms.45 Transfer reactions of P3− from 1-Dipp have led to the synthesis of exotic SbP3 species49 and modifications of Wilkinson’s catalyst in a much cleaner manner than previous attempts.47 Perhaps of most intrigue is the synthesis of AsP3 from 1-Dipp, which by itself is a possible material for semiconductors46,50 or can further react with [GaC(SiMe3)3}4] to generate AsGaP3-containing clusters toward the development of novel materials.51
3. RESULTS AND DISCUSSION In order to reduce the computational demand of complex 2, the bulky ODipp ligands were initially replaced with OMe ligands, Scheme 2. Although the steric environment about the niobium center will be altered, the electronic structure is not expected to change drastically, thus allowing 2 to serve as a good model for phosphorus transfer chemistry. Isolated key transition states were identified and optimized more efficiently with the less computationally demanding OMe ligands. Altering the OMe complexes to incorporate the full ODipp ligands was later performed to investigate the steric impact of the larger ligands (see section 3.6). 3.1. Geometries of Niobium Complexes 1 and 2 and Isomers of 3. Complexes 1-Dipp and 2-Dipp have been previously reported by the Cummins group.62 Replacing the ODipp ligands in 1-Dipp with OMe yields 1, Figure 1. Complex 1 is in a slightly distorted tetrahedral geometry about the Nb, due to the larger size of the η3-P3 ligand in comparison to the methoxy coligands, with all O−Nb−Pc bond angles equal to 112° (Pc = centroid of P3 ring). Bonding within the triphosphorus ring exhibits three nearly equal P−P bonds (2.23 ± 0.01 Å), consistent with sp3-hybridized phosphorus atoms. The calculated P−P bond lengths for 1 agree well with experimental P−P bond lengths from Cummins et al. on the 1Dipp complex (2.20 ± 0.01 Å).62 The addition of Ph3SnCl transforms 1 into [(η2-P3SnPh3)Nb(OMe)3], 2, with the loss of NaCl. The binding of niobium to the triphosphorus ring is best considered η2 in complex 2, as there are two similar Nb−P
2. COMPUTATIONAL METHODS The Gaussian 09 package was used for all calculations described herein.52 The B3LYP hybrid density functional was employed.53,54 Modeling of relativistic effects and description of the bonding and energetics were accomplished via the relativistic effective core potentials (RECPs) and the double-ζ valence basis sets of Stephens, Basch, and Krauss55 (CEP-31G) augmented with a d-polarization function for main group atoms. Optimized geometries and transition states (TS) were confirmed by the presence of zero and one imaginary frequency, respectively, in the calculated energy Hessian. Additional confirmation of transition states was achieved by intrinsic reaction coordinate (IRC) calculations.56,57 Forward and reverse reaction IRC pathways were followed with 25 points sampled in each direction and a step size of 0.01 amu1/2-Bohr (see Supporting Information). Single-point corrections to account for a basis set size, dispersion, and solvation were performed on optimized structures of all reaction C
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
The addition of pyridine-N-oxide (PyO) to complex 2 can generate four coordination isomers: (i) eq/ax, the oxidant occupies an equatorial position placing the Ph3SnP3 ligand axial; (ii) ax/eq, the oxidant binds in the axial position placing Ph3SnP3 equatorial; (iii) ax/ax, the oxidant and Ph3SnP3 are both in axial positions; or (iv) eq/eq, both the oxidant and Ph3SnP3 are in equatorial positions, Figure 2. Although the relative free energy for the ax/ax, 3ax/ax, is found to be the lowest, no direct insertion transition state of the oxygen atom into a niobium−phosphorus bond from the 3ax/ax coordination isomer can be found without prior rearrangement of the oxidant to the equatorial position (the Pc−Nb−Otrans bond angle is 166.4°, Pc = centroid of the P1−P2 bond). Optimization of the 3eq/eq isomer shows miniscule differences in geometry (overlay of structures yielded no RMS greater than 0.2 Å) in comparison to 3ax/eq, suggesting that the two conformations are isomers only in that there is a reflection in the xz plane, with a change in the orientation of the pyridine ring accounting for the small change in energy (ΔG = 0.3 kcal/ mol). Due to complex 3eq/eq being equal to 3ax/eq, and 3ax/ax
Figure 1. B3LYP/CEP-31G geometry optimized structures of [(η3P3)Nb(OMe)3]−, 1, left, and [(η2-P3SnPh3)Nb(OMe)3], 2, right, highlighting the different bonding modes of the P3 ring. Bond lengths are reported in Å.
bond lengths (2.58 and 2.59 Å) and an elongated third Nb−P bond of 2.83 Å, Figure 1.
Figure 2. Comparison of coordination isomers for complex 3, [(η2-P3SnPh3)Nb(OPy)(OMe)3]. Pyridine-N-oxide/Ph3SnP3 ligands can coordinate in the equatorial/axial positions, 3eq/ax, top left; axial/equatorial position, 3ax/eq, top right; axial/axial positions, 3ax/ax, bottom left; or equatorial/ equatorial, 3eq/eq, bottom right. Relative free energies (kcal/mol) are listed in parentheses for each conformation. Bond lengths are reported in Å. D
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics unable to directly insert the oxygen atom, only 3eq/ax and 3ax/eq are discussed as plausible conformations for the direct oxy insertion pathway. If it is thermodynamically favored to coordinate the oxidant in the 3ax/ax configuration, the system may be expected to pass through a fairly low energy Berry pseudorotation transition state to orient its ligands in the most favorable conformation for oxy insertion (Baeyer−Villiger, nonredox pathway).63 Such rotations have been studied in the literature for five-coordinate d0 TM complexes describing the orbital interactions and energetics.64,65 Calculated energy values for pseudorotations have been in the range of 10−20 kcal/mol for five-coordinate transition metal complexes.66−68 It is expected, however, that the reduced size of the OMe ligands in 2 will offer slightly lower energy barriers to rotation than the more sterically demanding ODipp ligands in 2-Dipp. 3.2. Transition State (TS3−4) and Metallocycles 4 and 5. Activation of the pyridine-N-oxide nitrogen−oxygen bond, coincident with oxygen atom insertion into the Nb−P bond, TS3−4, is hypothesized to be the rate-determining step (RDS) for the reaction pathway in Scheme 3.47 Thus, accurately determining the geometry of the transition state, TS3−4, is crucial in understanding the full reaction scheme and testing the hypothesis put forth by Cummins and Cossairt. Both the eq/ax and ax/eq conformations of TS3−4 exhibit an elongation of the NPyr−O bond and shortening of the O−P bond. Furthermore, in the imaginary mode, the Nb−P bond that is being broken elongates while the remaining Nb−P bond does not markedly change, indicating the Ph3SnP3 fragment remains bound to the Nb center, as expected for a nonredox oxygen insertion mechanism. Intermediates 4 and 5, which follow TS3−4, can best be considered [2,1]-bicyclo-Nb(OP3SnPh3) complexes with the Nb−P2 bond slightly elongated, 2.63 and 2.65 Å, respectively, compared to the η2 complex 3 (Nb−P1 = 2.58 Å, Nb−P2 = 2.59 Å), Figure 3. In order to sample additional pathways toward oxygen insertion, formation of the oxo complex prior to insertion into the Nb−P bond was also modeled. Transition-state optimizations of an equatorial bound oxo, [(η2-P3SnPh3)Nb(O)(Py)(OMe)3], resulted in formation of the local minimum metallocycle, complex 4. However, TS calculations with axial PyO coordination yielded a NbO complex where the Ph3SnP3 fragment is in a slightly altered geometry, TSoxo, where the two shortest Nb−P bonds, Nb−P1 and Nb−P2, are elongated by ∼0.2 Å each in comparison with its starting structure 3ax/ax (Figure 2). Additionally, the P1−P2 bond is shortened from 2.20 Å in 3ax/ax to 2.10 Å in TSoxo. These two results indicate that during the formation of the NbO bond the P3 fragment starts to resemble the separate Ph3SnP3 fragment (P1−P2 bond is 2.03 Å). 3.3. Generation of Niobium-Oxo (7) and P3− Synthon (8). Metal oxo complexes, LnMO, for high-valent early transition metal species have been well characterized in the literature and are known to be stable.32,69,70 It is hypothesized that upon insertion of the oxygen atom from pyridine-N-oxide into the Nb−P bond, formation of the Nb(O)(OMe)3 complex and the Ph3SnP3 fragment will readily proceed. While the formation of the niobium oxo complex is experimentally verified for the full ODipp complex by Cummins et al.,47 the mechanism of oxygen atom transfer could not be determined through experimental techniques, necessitating the use of computational methods to elucidate the energetically favored pathway.
Figure 3. Geometry-optimized metallocycle, with equatorially coordinated pyridine, 4, top, and pyridine-free complex metallocycle, 5, bottom. Bond lengths are reported in Å and angles in deg.
Optimizations found a weakly bound adduct as a local minimum, 6, with two niobium−phosphorus bonds of lengths 3.59 and 3.56 Å. The niobium in 6 is pseudotetrahedral with a short NbO bond, 1.73 Å. The formation of the NbO bond, 7, presumably provides the driving force for the release of the Ph3SnP3 fragment in which the P1−P2 bond length reduces from 2.29 Å, 5, to 2.03 Å, 6. Optimized NbO bond lengths are calculated to be 1.73 Å for [Nb(O)(OMe)3], 7, and show excellent agreement (NbO = 1.71 Å,10 1.72 Å,71 1.75 Å,72 1.70 Å73) with previously reported NbO bond lengths and against the experimentally isolated [Nb(O)(ODipp)3]2 complex, 7-DippX‑ray, Figure 4. Analysis of the calculated geometry of the Ph3SnP3 fragment, 8, shows excellent agreement with the characterized material.47 All bonds exhibit less than 2% difference compared to experimental data. The addition of the SnPh3 to the P3− fragment leads to two single bonds in the direction of the SnPh3 substituent (P1−P2 and P1−P3) and one shorter multiple bond (P2−P3). The P1−P2 and P1−P3 single bonds have calculated bond lengths of 2.29 and 2.28 Å, respectively, while the P2−P3 bond is shortened to a length of 2.03 Å, suggestive of multiple-bond character. In addition, the P−P bonds in 8 agree well with experimentally reported bond lengths of substituted triphosphorus rings (P−P = 2.23 Å,75 2.12 Å76) and doublebonded organic phosphines (PP = 2.03 Å,77 2.02 Å78) found in the literature.75−78 3.4. Thermodynamics of Oxidant Activation and Baeyer−Villiger Pathway. Mild conditions for the salt metathesis reaction (1 h, 20 °C) are used to synthesize the larger ODipp analogue, 2-Dipp, in good yield.47 Substitution of OMe groups for the larger ODipp reduces computational demand for calculations on active complex 2 toward the generation of the P3 synthon. E
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
Figure 5. Calculated oxygen insertion pathway, BV, for formation of niobium-oxo complex 7 and release of Ph3SnP3 fragment 8. Green energies represent the PyO/P3 ligands in the equatorial/axial conformation, while black energy values have the ligands in the axial/equatorial conformation. Oxygen insertion transition state TS3−4 is labeled as TSax/eq and TSeq/ax for mechanistic clarity. Geometries displayed in the diagram represent the lower energy conformations at each reaction coordinate. All energies are calculated relative to complex 2 and reported in kcal/mol.
Figure 4. Geometry-optimized structure of niobium-oxo complexes [Nb(O)(OMe)3], 7, top, and X-ray structure of [Nb(O)(ODipp)3], 7DippX‑ray, bottom.47,74 7-DippX‑ray was isolated from hydrogenolysis reaction, but also reported for the mechanism discussed in this work, Scheme 3.47 Isopropyl substituents on ODipp ligands are removed for clarity. Bond lengths are reported in Å and angles in deg.
the equatorial/axial conformation of PyO/P3. The Nb(O)(OMe)3 fragment of the adduct, 6, rearranges to take on the four-coordinate tetrahedral geometry. The calculated free energy for adduct formation is +3.6 kcal/mol, followed by release of Ph3SnP3, which has a calculated ΔG of −6.8 kcal/ mol. The complete separation of the fragments is favored by the formation of a stable Nb(V)-oxo complex, 7, and the Ph3SnP3 synthon, 8. The reaction pathway depicted in Figure 5 indicates that once the barrier to oxygen insertion is surmounted, a highly exergonic pathway is followed to generate the Nb(O)(OMe)3 complex, 7. The free energy liberated from the [(η2P3SnPh3)(O--Py)Nb(OMe)3]⧧ transition state, TS3−4, to the formation of Nb(O)(OMe)3, 7, is highly exergonic, with a calculated value of −91.9 kcal/mol from the TS ax/eq conformation and −87.0 kcal/mol from the TSeq/ax isomer. Liberation of the Ph3SnP3 fragment from the niobium complex provides an atom-economical way to introduce P3− moieties into organic molecules as intermediate transfer agents for the modification of organometallic complexes.47 Calculations on the Diels−Alder reaction of 8 with 1,3-cyclohexadiene were performed, and the full reaction energy profile is reported in Section 3.8. Geometric data compare well to Xray experimental data, and a low activation barrier (ΔG) of +12.6 kcal/mol is found for the formation of the Diels−Alder product, Ph3SnP3C6H8, making this product a viable synthon for the production of phosphorus-containing molecules. 3.5. Thermodynamics of Oxidant Activation and “Oxy” Pathway. Formation of the niobium-oxo complex Nb(O)(OMe)3, 7, prior to the release of the P3 fragment can occur only in the axial position, as discussed in Section 3.2. Coordination of the oxidant in the axial position results in the lowest energy isomer, 3ax/ax. Transition-state complex TSoxo exhibits the oxygen atom moving close to the Nb center, forming the NbO bond with an elongation of the O−N
The addition of pyridine-N-oxide to 2 generates complex 3eq/ax, [(η2-P3SnPh3)Nb(PyO)(OMe)3], with a calculated ΔG of +16.2 kcal/mol, Figure 5. In contrast, coordination of the oxidant to form 3ax/eq is calculated to be exergonic with a ΔG of +6.2 kcal/mol. The remaining unique conformation from Figure 2, 3ax/ax, has a calculated free energy of +5.1 kcal/mol for oxidant binding. Complex 3ax/ax has no direct pathway for oxygen insertion into the niobium−phosphorus bond without first rearranging to 3eq/ax or 3ax/eq at a cost of +11.1 and +1.1 kcal/mol, respectively. Calculations reveal free energy transition-state barrier heights of +24.7 and +29.6 kcal/mol for the eq/ax and ax/eq geometries, respectively. The calculated ΔΔG⧧ for TSax/eq ↔ TSeq/ax is 4.9 kcal/mol in favor of oxygen atom insertion from the eq/ax conformation, 3eq/ax, of the PyO/P3 ligands. Complexes 3ax/eq and 3ax/ax can therefore be considered precursor complexes to the transition state where pseudorotations will first generate 3eq/ax, which may then surmount the lowest energy barrier for oxygen insertion (+24.7 kcal/mol). Once the barrier to N−O bond activation is overcome, formation of the oxygen-inserted complex, 4, is highly exergonic, with a calculated ΔG of −75.7 kcal/mol from the lowest energy transition state, TSeq/ax, down to −51.0 kcal/mol in reference to starting complex 2. While the molecular motifs are different between Nb-OP3 and R3PO, the calculated free energy for forming complex 4 agrees well with reported oxidation values of triphenylphosphine (−73.8 kcal/mol) and trimethylphosphine (−79.7 kcal/mol).7 Loss of the pyridine ligand from 4 is entropically favored with a calculated ΔG of −8.1 kcal/mol to form complex 5 from F
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics bond. DFT calculations, ΔG⧧, place TSoxo +35.7 kcal/mol above the starting complex 2, Figure 6. Thus, the “oxy” pathway
Figure 7. Calculated oxygen insertion pathway for formation of niobium-oxo complex 7-Dipp and release of Ph3SnP3 fragment 8. Green energies represent the PyO/P3 ligands in the equatorial/axial conformation, while black energy values have the ligands in the axial/ equatorial coordination. Geometries displayed in the diagram represent the lower energy conformations at each reaction coordinate. All energies are calculated relative to complex 2-Dipp and reported in kcal/mol.
Figure 6. Calculated potential energy surface for formation of the oxo complex (TSoxo) prior to the release of the Ph3SnP3 fragment (8). All energies are calculated relative to complex 2 and reported in kcal/mol.
analogue, Figure 5. The size of the ODipp ligands seemingly results in a less selective coordination of the oxidant, as the three isomers of 3-Dipp exhibit closer energies to each other than the OMe isomers of 3. The larger ODipp ligands raise the energies for ax/ax and ax/eq coordination and lowered eq/ax coordination in the 3-Dipp isomers compared to isomers of 3. Transition-state activation barriers for oxygen atom insertion into the Nb−P bond are calculated to be lowest for TSDippeq/ax, +20.9 kcal/mol (ΔG⧧), with TS3‑Dippax/eq calculated to be +29.6 kcal/mol (ΔG⧧), 8.7 kcal/mol higher than TSDippeq/ax (ΔΔG⧧). The energetic trend of the eq/ax TS configuration lying lower in energy agrees with the previously discussed results for the OMe system, Figure 5. However, oxygen atom insertion into the Nb−P bond in the eq/ax position, compared to the ax/eq position, is more favored for the larger ODipp complex. Specifically, the ΔΔG⧧ for the two OMe TS complexes (TSax/eq − TSeq/ax) is calculated to be +4.9 kcal/mol compared to +8.7 kcal/mol for the ODipp TSs, Table 1. Furthermore, the
exhibits the highest energy pathway of all systems identified in this research, with a calculated ΔΔG⧧ of +11.0 (3ax/ax → TSoxo) compared to the lowest energy “BV” path (3eq/ax → TSeq/ax). Similar to the BV pathway, loss of the pyridine ligand following the transition state is thermodynamically favored. Complex 6oxo exhibits elongated Nb−P bonds, 5.12 and 5.71 Å, and a shortened P−P bond, 2.03 Å, indicating the release of the P3 fragment upon forming the Nb(V) complex. Complete separation of 7 and 8 is entropically favored and calculated to be −9.8 kcal/mol downhill from 6oxo. 3.6. Oxygen Insertion with [(η2-P3SnPh3)Nb(ODipp)3]. In order to compare calculated results directly to experimental findings, the reaction mechanisms investigated above were explored for the complete niobium complex [(η2-P3SnPh3)Nb(ODipp)3], 2-Dipp. Analysis of the OMe model catalyst geometric and thermochemical results provides insight into the mechanism for niobium oxidation, which supplies an ad hoc “template” for the study of the larger ODipp complex. Inclusion of the large ODipp ligands will undoubtedly have an effect on oxidant coordination energetics due to steric considerations and will provide a description of the system utilized in the experimental procedure,47 which may or may not alter the mechanistic and energetic trends calculated with the OMe complex 2. Similar to the methoxy complex, three unique isomers were optimized with the oxidant, PyO, coordinated to niobium and exhibit the same energetic pattern; 3-Dippeq/ax > 3-Dippax/eq > 3-Dippax/ax (the fourth isomer, 3-Dippeq/eq, optimizes to 3Dippax/eq upon minimization). The relative order of isomers of 3-Dipp exhibits the same pattern as that of the OMe complexes; however, the calculated energy barriers for oxidant coordination are higher compared to the OMe system for 3Dippax/eq (ΔΔG⧧ = +6.9 kcal/mol) and 3-Dippax/ax (ΔΔG⧧ = +6.2 kcal/mol), most certainly due to the steric bulk of the ODipp ligands. Interestingly, while 3-Dippeq/ax remains the highest energy isomer for oxidant coordination, the calculated ΔΔG⧧ is −1.7 kcal/mol, Figure 7, compared to the OMe
Table 1. Calculated Activation Energy Barriers (kcal/mol) for the Transition States of OMe and ODipp Complexes Relative to the Starting Complexes, 2 or 2-Dipp, Respectively TSoxo TSax/eq TSeq/ax a
ΔG⧧OMea
ΔG⧧ODippa
ΔG⧧OMeb
ΔG⧧ODippb
+35.7 +29.6 +24.7
+35.9 +29.6 +20.9
42.5 33.5 30.9
40.5 32.7 23.7
B3LYP/CEP-31G/gas. THF.
b
B3LYP-GD3/CEP-121G|6-311G(d,p)/
difference in activation barriers between the TSeq/ax (OMe) and TSDippeq/ax (ODipp) is 3.8 kcal/mol in favor of the ODipp complex. It is hypothesized that the steric bulk of the ODipp ligands, compared to the OMe ligands, results in a reduced barrier for oxygen atom insertion into the niobium− phosphorus bond due, in part, to the enhanced selectivity of oxidant binding in the eq/ax position (ΔG⧧ reduced by 1.7 G
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics kcal/mol, 3-Dippeq/ax), but more importantly to the decrease in selectivity for the ax/eq and ax/ax coordination isomers (ΔG⧧ increased by 6.9 kcal/mol, 3-Dippax/eq, and 6.2 kcal/mol, 3Dippax/ax). Following oxygen insertion into the Nb−P bond, the reaction proceeds favorably to generate the metallocycle, calculated to be downhill by −66.4 kcal/mol for the eq/ax configuration and −78.1 kcal/mol for the ax/eq isomer. The small difference in energies (ΔΔG⧧ = 3.0 kcal/mol) in 4-Dipp, Figure 7, is alleviated with the removal of the coordinated pyridine ligand 5-Dipp, which is expected based on experimental isolation of the dimer of 7-Dipp with no coordinated pyridine.47,74 Formation of the strong NbO bond drives the reaction through the adduct geometry 6-Dipp, which is calculated to be +2.4 kcal/mol uphill from 5-Dipp. Complete dissociation of the Ph3SnP3 fragment 8 from 7-Dipp is entropically favored and is calculated to be −8.6 kcal/mol downhill from 6-Dipp. Calculations were also performed to determine the reaction path for the oxo (redox) pathway in which the oxo is generated from the 3-Dippax/ax isomer. Animation of the imaginary frequency in the calculated TSDippoxo exhibits the simultaneous NbO bond formation and Nb−P bond breaking, indicating the Ph3SnP3 fragment is lost when the NbO bond forms. Additionally, the P2−P1 bond length decreases to 2.11 Å as the NbO bond forms, indicating the formation of a PP bond (see discussion above for PP bond lengths), Figure 8. Adduct complex 6-Dippoxo has the Ph3SnP3 fragment displaced from the Nb(O)(ODipp)3 fragment by >5.0 Å, Figure 8, and also exhibits a PP bond, with a calculated distance of 2.03 Å. Calculations reveal the free energy of activation for the Nb− O bond to form the NbO bond is uphill by +23.9 kcal/mol (3-Dippax/ax → TSDippoxo), Figure 9. The effective barrier, ΔG⧧, for this TS is calculated to be 35.9 kcal/mol. Compared to TSDippax/eq and TSDippeq/ax this TS sits higher in energy by 6.3 and 15.0 kcal/mol, respectively, Table 1. In agreement with the previously presented results for the OMe model catalyst, the oxo (redox) pathway is calculated to be the least favorable pathway of all systems identif ied in this research. 3.7. Basis Set, Dispersion, and Solvent-Corrected Values. To improve accuracy and ensure our results represent the synthetic environment, additional single-point calculations were performed to correct for a larger basis set, dispersion, and solvent (see Computational Methods). Table 1 shows the corrected free energy values for rate-limiting steps TS3−4 and TSoxo compared to the originally calculated values for 2 and 2Dipp (see Supporting Information for energies of reaction coordinates using the larger basis set method). The corrected values of the reaction path for complex 2 exhibit the same mechanistic selectivity previously determined with the smaller basis set. Specifically, 3ax/ax provides the lowest energy conformation of bound oxidant but proceeds through the highest transition state, TSoxo, with a calculated energy barrier of +40.5 kcal/mol. Conformation 3eq/ax is the highest energy intermediate, lying +8.7 kcal/mol higher than 3ax/ax and +9.2 kcal/mol higher than conformer 3ax/eq. However, TS3eq/ax is calculated to have the lowest activation energy at +30.9 kcal/ mol, Table 1, agreeing with previous results. Similarly for the 2-Dipp pathway, the 3-Dippeq/ax conformation is again the highest, but has the lowest activation energy, +23.7 kcal/mol, toward generation of the P3-containing moiety, 8. As discussed above, the differences in energies between the three conformations of 3-Dipp are reduced,
Figure 8. Transition-state structure TSDippoxo, top, showing the formation of the NbO bond and elongated Nb−P bonds. Adduct complex 6-Dippoxo, bottom, indicated the dissociation of the Ph3SnP3 fragment upon oxo formation and the creation of a PP bond (2.03 Å). iPr groups are removed from TSDippoxo for clarity. Bond lengths are reported in Å.
Figure 9. Calculated potential energy surface for the formation of the oxo complex (TSDippoxo) prior to the release of the Ph3SnP3 fragment (8). All energies calculated relative to complex 2 and reported in kcal/ mol.
H
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
4. CONCLUSIONS The use of early transition metal complexes to generate phosphorus-containing synthons from elemental P4 is a valuable process toward the production of phosphoruscontaining molecules. DFT methods were employed to understand the mechanism by which [(η2-P3SnPh3)Nb(ODipp)3], 2, is oxidized by PyO to release the Ph3SnP3 fragment. Reports of oxygen insertion into a Nb−P bond are rare in the literature, and results presented here offer an investigation into generating phosphorus-containing fragments for synthetic purposes via a nonredox, oxy-insertion pathway. Calculations identify oxygen insertion as the rate-determining step for the process, with formation of the metallocycle then eventual metal-oxo complex being highly exothermic. Thus, the driving force of the process is the formation of the MO bond, which is strong for high-valent, early transition metals, forming the stable d0, Nb(V)-oxo complex.32 It is determined from calculations that binding the oxidant, pyridine-N-oxide, with respect to the Ph3SnP3 ligand (PyO/P3) is more favored for the axial/equatorial coordination isomer, 3ax/eq, versus the equatorial/axial coordination isomer, 3eq/ax. However, formation of the transition state is favored (ΔΔG = −2.6 kcal/mol, corrected) when the oxidant occupies the equatorial/axial position, TSeq/ax. This suggests an internal pseudorotation to achieve the low-energy, thermodynamically favored configuration, TSeq/ax. Once the barrier to oxygen insertion is overcome, liberation of the Ph3SnP3 fragment and formation of the oxo are calculated to be thermodynamically favorable. Conversely, calculations reveal the “oxo” mechanism dissociates the Ph3SnP3 fragment upon NbO bond formation in the axial coordination site. Coordination of pyridine-N-oxide to 2 in the axial site, 3ax/ax, exhibits the lowest calculated free energy. The release of the Ph3SnP3 fragment happens spontaneously at the transition state, TSoxo, as the influential trans niobium−oxo bond forms, characterizing the mechanism as a redox pathway in which the niobium is oxidized from Nb(III) to Nb(V). Calculations on the full [(η2-P3SnPh3)Nb(ODipp)3] complex, 2-Dipp, and its potential energy surfaces for both the organometallic BV (nonredox) and oxy (redox) pathways reveal a similar trend in energetics. Specifically, the oxidantbound coordination isomers show the pattern 3-Dippeq/ax > 3Dippax/ax > 3-Dippax/eq. Moreover, the transition-state activation energies exhibit the same pattern as the methoxy complex, which calculates the eq/ax, TSDippeq/ax, to have the lowest activation energy. The BV pathway proceeds in the same manner, through the metallocycle, 4-Dipp, adduct, 6-Dipp, and formation of the oxo complex 7-Dipp upon the release of the Ph3SnP3 fragment. It is determined from DFT calculations that the generation of the Ph3SnP3 fragment, 8, is energetically favored when PyO/P3 ligands are in the equatorial/axial arrangement, TS eq/ax (TSDippeq/ax). In comparison, the axial/equatorial isomers, TSax/eq (TS Dippax/eq ), and the trans oxo isomers, TSoxo (TSDippoxo), have corrected free energy barriers 2.6 kcal/mol (9.0 kcal/mol) and 11.6 kcal/mol (7.73 kcal/mol) higher than TSeq/ax (TSDippeq/ax), respectively. This identifies the organometallic Baeyer−Villiger, nonredox, pathway to be the favored mechanism for oxidation of the model [(η2-P3SnPh3)Nb(OMe)3] complex and [(η2-P3SnPh3)Nb(ODipp)3] to generate the synthetically useful P3− synthon.
suggesting both an enhanced selectivity for 3-Dippeq/ax and a decreased selectivity for 3-Dippax/eq, due to the increased steric requirement of the ODipp ligands. Including dispersion effects did affect the energies of complexes 5 and 5-Dipp, where the loss of the weakly coordinated pyridine ligand becomes slightly uphill, +3.8 and +5.5 kcal/mol, respectively. Similarly, the formation of 6 and 6Dipp is raised compared to results reported above. Furthermore, the calculated free energy moving from coordinate 6 → 7 in all reactions is reduced compared to original values. This can be expected with such adduct-type complexes where dispersion is expected to play an important role. In accord with experimental evidence and previously calculated energy barriers, the corrected calculations predict the rate-determining barrier to be TS3−4 in all mechanisms, Table 1. 3.8. Diels−Alder Reaction of Ph3SnP3 (8) with 1,3Cyclohexadiene. Liberation of Ph3SnP3, 8, from starting material 2 or 2-Dipp is an overall thermodynamically favored process with only one major barrier to overcome: bonding and insertion of the oxidant, Figure 10. While complexes 6 and 6-
Figure 10. Calculated reaction pathway for the Diels−Alder adduct 9. Values in parentheses from basis set, dispersion, and solvationcorrected energy calculations. Free energies are reported in kcal/mol.
Dipp are calculated to be just slightly endergonic relative to 5 and 5-Dipp, respectively, it is expected that the release of 8 will proceed with little resistance in order to form the stable niobium-oxo complex 7 or 7-Dipp, which has been isolated as a dimer experimentally47 for the ODipp system. The Diels−Alder addition of 8 and 1,3-cyclohexadiene proceeds with 8 acting as the dienophile. Formation of the transition-state complex, TS8−9, is calculated to possess an activation barrier of +12.6 kcal/mol at the lower level. The additional corrections to the computational method reduced the energy barrier to +4.8 kcal/mol for TS8−9, implying a fast reaction, which correlates with experimental studies of 9. Transition-state TS8−9 has a small imaginary frequency describing the formation of two P−C bonds and the slight elongation of the PP bond. Generation of the Diels−Alder adduct, 9, from TS8−9 is exergonic with a corrected ΔG of −21.8 kcal/mol. Formation of 9 provides a synthetically feasible route to making phosphorus-containing molecules. Specifically, it has been shown that 9 can be employed as an efficient delivery source of phosphorus atoms to different transition metal complexes47 and arsenic (AsP3) for use in the development of novel materials.49,50 I
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics
■
(22) Visser, S. P. d.; Ogliaro, F.; Harris, N.; Shaik, S. J. Am. Chem. Soc. 2001, 123, 3037−3047. (23) Chiavarino, B.; Cipollini, R.; Crestoni, M. E.; Fornarini, S.; Lanucara, F.; Lapi, A. J. Am. Chem. Soc. 2008, 130, 3208−3217. (24) Newcomb, M.; Hollenberg, P.; Coon, M. Arch. Biochem. Biophys. 2003, 409, 72−79. (25) Hirao, H.; Kumar, D.; Que, L., Jr.; Shaik, S. J. Am. Chem. Soc. 2006, 128, 8590−8606. (26) Kumar, D.; Derat, E.; Khenkin, A. M.; Neumann, R.; Shaik, S. J. Am. Chem. Soc. 2005, 127, 17712−17718. (27) Zhang, R.; Newcomb, M. Acc. Chem. Res. 2007, 41, 468−477. (28) Vasbinder, M. J.; Bakac, A. Inorg. Chem. 2007, 46, 2921−2928. (29) Goldberg, D. P. Acc. Chem. Res. 2007, 40, 626−634. (30) Yin, G. Coord. Chem. Rev. 2010, 254, 1826−1842. (31) Almeida, K. J. d.; Silva, T. C.; Neto, J. L.; Rocha, M. V. J.; Ramalho, T. C.; Miranda, M. N. d.; Duarte, H. A. J. Organomet. Chem. 2016, 802, 49−59. (32) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; John Wiley & Sons, Inc., 1988. (33) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441−450. (34) Brown, S. N.; Mayer, J. M. J. Am. Chem. Soc. 1996, 118, 12119− 12133. (35) Yu, H.; Jia, G.; Lin, Z. Organometallics 2008, 27, 3825−3833. (36) Conley, B. L.; Ganesh, S. K.; Gonzales, J. M.; Ess, D. H.; Nielsen, R. J.; Ziatdinov, V. R.; Oxgaard, J.; Goddard, W. A., III; Periana, R. A. Angew. Chem., Int. Ed. 2008, 47, 7849−7852. (37) Gonzales, J. M.; Distasio, R., Jr.; Periana, R. A.; Goddard, W. A., III; Oxgaard, J. J. Am. Chem. Soc. 2007, 129, 15794−15804. (38) Bischof, S. M.; Cheng, M.-J.; Nielsen, R. J.; Gunnoe, T. B.; William, A.; Goddard, I.; Periana, R. A. Organometallics 2013, 30, 2079−2082. (39) Figg, T. M.; Cundari, T. R.; Gunnoe, T. B. Organometallics 2011, 30, 3779−3785. (40) Pardue, D. B.; Mei, J.; Cundari, T. R.; Gunnoe, T. B. Inorg. Chem. 2014, 53, 2968−2975. (41) Mei, J.; Carsch, K. M.; Freitag, C. R.; Gunnoe, T. B.; Cundari, T. R. J. Am. Chem. Soc. 2013, 135, 424−435. (42) Piro, N. A.; Cummins, C. C. Angew. Chem., Int. Ed. 2009, 48, 934−938. (43) Cossairt, B. M.; Cummins, C. C. Angew. Chem., Int. Ed. 2008, 47, 8863−8866. (44) Cossairt, B. M.; Cummins, C. C. Inorg. Chem. 2008, 47, 9363− 9371. (45) Cossairt, B. M.; Piro, N. A.; Cummins, C. C. Chem. Rev. 2010, 110, 4164−4177. (46) Cossairt, B. M.; Cummins, C. C. J. Am. Chem. Soc. 2009, 131, 15501−15511. (47) Cossairt, B. M.; Cummins, C. C. Angew. Chem., Int. Ed. 2010, 49, 1595−1598. (48) Velian, A.; Cummins, C. C. Chem. Sci. 2012, 3, 1003−1006. (49) Cossairt, B. M.; Diawara, M.-C.; Cummins, C. C. Science 2009, 323, 602. (50) Manasreh, O. Semiconductor Heterojunctions and Nanostructures; McGraw-Hill: New York, 2005. (51) Cossairt, B. M.; Cummins, C. C. Chem. - Eur. J. 2010, 16, 12603−12608. (52) 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.; J. Bloino, G. Z.; 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.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; 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,
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.organomet.6b00679. Coordinates of all optimized structures in xyz format (XYZ) Further discussion on calculated geometries and energies of structures, IRC analysis of calculated transition states, enthalpies and free energies of all optimized structures (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Support of our work by CASCAM and the U.S. Department of Energy, through grant DE-FG02-03ER15387, the Research Council of Norway (Grant No.23170/F20), and from the Norwegian supercomputing program NOTUR (Grant No. NN4654K) is gratefully acknowledged. Special thanks to Dr. Kit Cummins (MIT) and Dr. Alexandria Velian for their insight and discussions about this work.
■
REFERENCES
(1) Tenn, W. J., III; Conley, B. L.; Hovelmann, C. H.; Ahlquist, M.; Nielsen, R. J.; Ess, D. H.; Oxgaard, J.; Bischof, S. M.; Goddard, W. A., III; Periana, R. A. J. Am. Chem. Soc. 2009, 131, 2466−2468. (2) Conley, B. L.; Ganesh, S. K.; Gonzales, J. M.; Tenn, W. J., III; Young, K. J. H.; Oxgaard, J.; Goddard, W. A., III; Periana, R. A. J. Am. Chem. Soc. 2006, 128, 9018−9019. (3) Tenn, W. J., III; Young, K. J. H.; Bhalla, G.; Oxgaard, J.; Goddard, W. A., III; Periana, R. A. J. Am. Chem. Soc. 2005, 127, 14172−14173. (4) Brown, S. N.; Mayer, J. M. Organometallics 1995, 14, 2951−2960. (5) Brown, S. N.; Mayer, J. M. J. Am. Chem. Soc. 1994, 116, 2219− 2220. (6) Strukul, G. Angew. Chem., Int. Ed. 1998, 37, 1198−1209. (7) Holm, R. H.; Donahue, J. P. Polyhedron 1993, 12, 571−589. (8) Lee, S. C.; Holm, R. H. Inorg. Chim. Acta 2008, 361, 1166−1176. (9) Gisdakis, P.; Rosch, N. J. Am. Chem. Soc. 2001, 123, 697−701. (10) Deng, L.; Ziegler, T. Organometallics 1996, 15, 3011−3021. (11) Acree, W. E., Jr.; Tucker, S. A.; Ribeiro da Silva, M. D. M. C.; Matos, M. A. R.; Goncalves, J. M.; Ribeiro da Silva, M. A. V.; Pilcher, G. J. Chem. Thermodyn. 1995, 27, 391−398. (12) Carter, E. A.; Goddard, W. A., III J. Phys. Chem. 1988, 92, 2109−2115. (13) Luo, L.; Lanza, G.; Fragala, I. L.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1998, 120, 3111−3112. (14) Benson, M. T.; Cundari, T. R.; Lim, S. J.; Nguyen, H. D.; PierceBeaver, K. J. Am. Chem. Soc. 1994, 116, 3955−3966. (15) Gutsev, G. L.; Rao, B. K.; Jena, P. J. Phys. Chem. A 2000, 104, 11961−11971. (16) Che, C.-M. Pure Appl. Chem. 1995, 67, 225−232. (17) Yin, G.; Danby, A. M.; Kitko, D.; Carter, J. D.; Scheper, W. M.; Busch, D. H. J. Am. Chem. Soc. 2008, 130, 16245−16253. (18) Czernuszewicz, R. S.; Mody, V.; Czader, A.; Gałezowski, M.; Gryko, D. T. J. Am. Chem. Soc. 2009, 131, 14214−14215. (19) Rappe, A. K.; Goddard, W. A., III. Nature 1980, 285, 311−312. (20) Rappe, A. K.; Goddard, W. A., III J. Am. Chem. Soc. 1982, 104, 3287−3294. (21) Tahsini, L.; Bagherzadeh, M.; Nam, W.; Visser, S. P. d. Inorg. Chem. 2009, 48, 6661−6669. J
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
Article
Organometallics P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.02; Gaussian Inc.: Wallingford, 2013. (53) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. (54) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38, 3098− 3100. (55) Stevens, W. J.; Basch, H.; Krauss, M. J. Chem. Phys. 1984, 81, 6026−6033. (56) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154− 2161. (57) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523− 5527. (58) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72, 5639. (59) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650. (60) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010, 132, 154104. (61) Tomasi, J.; Mennucci, B.; Cancès, E. J. Mol. Struct.: THEOCHEM 1999, 464, 211. (62) Cossairt, B. M.; Diawara, M. C.; Cummins, C. C. Science 2009, 323, 602. (63) Berry, R. S. J. Chem. Phys. 1960, 32, 933−938. (64) Westmoreland, T. D. Inorg. Chim. Acta 2007, 361, 1187−1191. (65) Kosma, K.; Trushin, S. A.; Fuß, W.; Schmid, W. E.; Schneider, B. M. R. Phys. Chem. Chem. Phys. 2010, 12, 13197−13214. (66) Nakazawa, H.; Kawamura, K.; Ogawa, T.; Miyoshi, K. J. Organomet. Chem. 2002, 646, 204−211. (67) Cundari, T. R.; Taylor, C. D. Organometallics 2003, 22, 4047− 4059. (68) Couzijn, E. P. A.; Slootweg, J. C.; Ehlers, A. W.; Lammertsma, K. J. Am. Chem. Soc. 2010, 132, 18127−18140. (69) Kurihara, M.; Katoh, N.; Kojima, T.; Ishii, Y.; Matsuda, Y. Inorg. Chem. 1995, 34, 4888−4895. (70) Perjessy, A.; Ertl, P.; Pronayova, N.; Gautheron, B.; Broussier, R. J. Organomet. Chem. 1994, 466, 133−137. (71) Lecomte, C.; Protas, J. J. Chem. Soc., Chem. Commun. 1976, 434−435. (72) Lecomte, C.; Protas, J.; Richard, P.; Barbe, J. M.; Guilard, R. J. Chem. Soc., Dalton Trans. 1982, 247−250. (73) Hubert-Pfalzgraf, L. G.; Tsunoda, M.; Borgne, G. L. J. Chem. Soc., Dalton Trans. 1988, 533−535. (74) Visciglio, V. M.; Fanwick, P. E.; Rothwell, I. P. Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 1994, C50, 900−902. (75) Bianchini, C.; Vaira, M. D.; Meli, A.; Sacconi, L. Inorg. Chem. 1981, 20, 1169−1173. (76) Vaira, M. D.; Sacconi, L.; Stoppioni, P. J. Organomet. Chem. 1983, 250, 183−195. (77) Yoshifuji, M.; Shima, I.; Inamoto, N. J. Am. Chem. Soc. 1981, 103, 4587−4589. (78) Smith, R. C.; Urnezius, E.; Lam, K.-C.; Rheingold, A. L.; Protasiewicz, J. D. Inorg. Chem. 2002, 41, 5296−5299.
K
DOI: 10.1021/acs.organomet.6b00679 Organometallics XXXX, XXX, XXX−XXX
