Article pubs.acs.org/IC
Synthesis and Structural Studies of Nickel(0) Tetracarbene Complexes with the Introduction of a New Four-Coordinate Geometric Index, τδ Mark H. Reineke, Matthew D. Sampson, Arnold L. Rheingold, and Clifford P. Kubiak* Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, Mail Code 0358, La Jolla, California 92093-0358, United States S Supporting Information *
ABSTRACT: The synthesis and characterization of two homoleptic chelating nickel(0) tetracarbene complexes are reported. These are the first group 10 M(0) (M = Ni, Pd, Pt) tetracarbene complexes. These species have geometries intermediate between C2v sawhorse and tetrahedral and show high UV−vis absorption in the 350−600 nm range, with extinction coefficients (ϵ) between 5600 and 9400 M−1 cm−1. Density functional theory analysis indicates that this high absorptivity is due to metal-to-ligand charge transfer. In order to better describe the unusual geometries encountered in these complexes, an adjustment to the popular τ4 index for fourcoordinate geometries is introduced in order to better delineate between sawhorse and distorted tetrahedral geometries.
■
INTRODUCTION Coordinatively unsaturated group 10 complexes have been successfully used in many catalytic applications, including palladium- and platinum-catalyzed cross-coupling reactions1,2 and C−H activation reactions.3,4 Typically, palladium and platinum have been the primary focus of development in this field. Recently, however, nickel complexes have found growing popularity as earth-abundant analogues to their second- and third-row counterparts.5,6 This surge in popularity has occurred in tandem with the proliferation of N-heterocyclic carbene (NHC) ligands in the coordination chemistry of first-row transition metals, resulting in new and varied reactivity in organonickel chemistry.6−9 Such NHC-stabilized nickel complexes have been reported as catalysts for a variety of transformations,7,10 including aryl halide reduction11 and cross-coupling reactions.12,13 Typically, NHC ligands are incorporated into these systems as σ-donating analogues to two-electron donor ligands such as phosphines or amines.14 It is therefore surprising that, despite the rich reductive electrochemistry of nickel complexes containing phosphorus and nitrogen donors for applications such as carbon dioxide (CO2) reduction15 and proton reduction,16 there are only a handful of reports in the literature discussing the effects of NHC ligands on the electrochemistry or electrocatalytic properties of nickel complexes.17−23 Similarly, isolated nickel(0) complexes containing carbene ligands (the products of reduction of these systems) are unusual, primarily restricted to homoleptic bis-24 and tris-NHC19 complexes, heteroleptic nickel olefin complexes,25−27 and NHC adducts of Ni(CO)4.28−31 © 2015 American Chemical Society
Herein we report the in situ electrochemical generation, direct synthesis, and structural characterization of bis(1,1′methylene-3,3′-diimidazole-1,1′-2,2′-diylidene)nickel(0) (3a) and bis(1,1′-methylene-3,3′-dibenzimidzole-2,2′-diylidene)nickel(0) (3b) from bis(1,1′-methylene-3,3′-diimidazole-1,1′2,2′-diylidene)nickel(II) bromide (1a) and bis(1,1′-methylene3,3′-dibenzimidzole-2,2′-diylidene)nickel(II) bromide (1b), respectively. These are homoleptic NHC analogues to existing nickel diphosphine complexes. 3a and 3b are, to our knowledge, the first isolated and structurally characterized nickel(0) tetracarbene complexes. Furthermore, when these species are compared to nickel(0) diphosphine complexes, it was found that the most popular index for the classification of four-coordinate geometries, τ4,32 is insufficient to distinguish between sawhorse geometries seen in this study and the pinched tetrahedra common in analogous complexes. As such, this index has been modified with a new term to take into account the marked differences between these two coordination geometries. These modifications have been incorporated into a new index for the rapid and convenient classification of fourcoordinate complexes, τδ.
■
EXPERIMENTAL SECTION
Synthesis. All reagents were obtained from commercial suppliers and used without purification unless otherwise noted. All reactions were carried out under a N2 atmosphere using standard Schlenk and glovebox techniques. Acetonitrile (MeCN) and tetrahydrofuran Received: November 24, 2014 Published: March 10, 2015 3211
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry (THF) were dried over activated 3 Å molecular sieves and alumina and degassed prior to use. Dimethyl sulfoxide (DMSO) was dried by distillation of volatiles under vacuum, stored over activated molecular sieves, and degassed prior to use. NiBr2·6H2O was dehydrated dried under vacuum at 60 °C overnight. Complete dehydration was verified gravimetrically. Tetrabutylammonium hexafluorophosphate (TBAPF6) was twice recrystallized from methanol (MeOH) and dried under vacuum. Complexes 1a and 1b were synthesized as previously reported.12,33 KC8 was synthesized as previously reported.34 Note that KC8 is extremely pyrophoric and should be handled only under an inert atmosphere and stored in a freezer at −30 °C. Synthesis of the Hexafluorophosphate Salt 2a. To a solution of the tetracarbene dibromide 1a (499 mg, 0.88 mmol) dissolved in a minimum of boiling MeOH was added a boiling methanolic solution of ammonium hexafluorophosphate (NH4PF6; 502 mg, 2.99 mmol). A pale-yellow precipitate began to form immediately. The reaction was allowed to stand and cool to room temperature. The product was collected by vacuum filtration and dried overnight under vacuum at 60 °C, resulting in a pale-yellow solid (504 mg, 0.72 mmol, 82% yield). 1 H NMR (400 MHz, DMSO-d6): δ 7.70 (s, 4H, CH), 7.31 (s, 4H, CH), 6.68 (dd, 4H, CH2, J = 202.3 and 12.9 Hz), 3.18 (s, 12H, CH3). 13 C NMR (500 MHz, DMSO-d6): δ 172.96 (Ni−C), 123.93, 122.84 (CH), 62.51 (CH2), 36.56 (CH3). ESI-HRMS ([M]2+). Calcd for [C18H24N8Ni]2+: m/z 205.0732. Found: m/z 205.0733. Elem anal. Calcd for C18H24N8NiP2F6: C, 30.84; H, 3.45; N, 15.98. Found: C, 30.85; H, 30.5; N, 16.00. UV−vis (DMSO): λmax = 326 nm. Synthesis of the Hexafluorophosphate Salt 2b. To a solution of the tetracarbene dibromide 1b (100 mg, 0.13 mmol) dissolved in a minimum of boiling MeOH was added a boiling methanolic solution of NH4PF6 (60 mg, 0.36 mmol). A pale-yellow precipitate began to form immediately. The reaction was allowed to stand and cool to room temperature. The product was collected by vacuum filtration and dried overnight under vacuum at 60 °C, resulting in a bright-yellow solid (98 mg, 0.11 mmol, 84% yield). 1H NMR (400 MHz, DMSOd6): δ 8.34 (d, 4H, CH, J = 8.2 Hz), 7.68 (d, 4H, CH, J = 8.1 Hz), 7.54 (m, 8H, CH, CH2), 7.46, (t, 4H, CH, J = 7.7 Hz), 3.49 (s, 12H, CH3). 13 C NMR (500 MHz, DMSO-d6): δ 183.38 (Ni−C), 134.77, 133.66, 124.95, 124.62, 112.52, 111.68 (CH), 122.84 (CH), 57.39 (CH2), 34.57 (CH3). ESI-HRMS ([M]+). Calcd: m/z 305.1046. Found: m/z 305.1046. Elem anal. Calcd for C34H32N8NiP2F6: C, 45.31; H, 3.58; N, 12.43. Found: C, 44.96; H, 3.42; N, 12.45. UV−vis (DMSO): λmax = 270, 340 nm. Synthesis of the Nickel(0) Tetracarbene Complex 3a. To a suspension of the tetracarbene dibromide 1a (150 mg, 0.26 mmol) in THF (15 mL) was added 2.1 equiv of KC8 (75 mg, 0.55 mmol). This suspension was allowed to react at room temperature overnight. The dark suspension was then filtered, reserving the supernatant. Solvent was removed under vacuum to give a dark-red solid product (54 mg, 0.13 mmol, 51% yield). 1H NMR (400 MHz, THF-d8): δ 6.73 (s, 4H, CH), 8.65 (s, 4H, CH), 5.26 (s, 4H, CH2), 3.05 (s, 12H, CH3). 13C NMR (500 MHz, DMSO-d6): δ 121.16, 119.38 (CH), 58.04 (CH2), 31.52 (CH3). Note: no carbenic 13C resonance was observed, even after extended collection times. ESI-HRMS ([M − H]+). Calcd for [C34H31N8Ni]+: m/z 609.2022. Found: m/z 609.2020. UV−vis (THF): λmax = 392 nm. Synthesis of the Nickel(0) Tetracarbene Complex 3b. To a suspension of the tetracarbene dibromide 1b (158 mg, 0.21 mmol) in THF (15 mL) was added 2.1 equiv of KC8 (81 mg, 0.60 mmol). This suspension was allowed to react at room temperature overnight. The dark suspension was then filtered, reserving the supernatant. The solvent was removed under vacuum to give a dark-violet solid product (72 mg, 0.12 mmol, 57% yield). 1H NMR (400 MHz, THF-d8): δ 7.43 (d, 4H, CH), 6.93 (m, 12H, CH), 6.00 (s, 4H, CH2), 3.27 (s, 12H, CH3). 13C NMR (500 MHz, DMSO-d6): δ 209.38 (Ni−C), 137.09, 135.56, 119.68, 119.35, 106.66, 105.30 (CH), 53.91 (CH2), 25.38 (CH3). ESI-HRMS ([M − H]2+). Calcd for [C18H23N8Ni]+: m/z 409.1394. Found: m/z 409.1394. UV−vis (THF): λmax = 390, 544 nm. Electrochemistry. Electrochemical experiments were performed using a BASi Epsilon potentiostat. Cyclic voltammograms (CVs) were performed under nitrogen in a one-compartment cell using a glassy
carbon BASi working electrode (3 mm diameter), a platinum wire counter electrode, and a nonaqueous silver reference electrode from BASi containing a silver wire in a DMSO solution of 0.1 M TBAPF6 and 0.1 mM AgNO3. All experiments were performed in dry DMSO using 0.1 M TBAPF6 as the supporting electrolyte. After the initial CVs were taken, ferrocene (Fc) was added as an internal reference. X-ray Crystallographic Studies. Single-crystal X-ray diffraction studies were carried out on a Bruker Kappa APEX-II CCD diffractometer equipped with Mo Kα radiation (λ = 0.71073 Å). The crystals were mounted on a Cryoloop with Paratone oil, and data were collected under a nitrogen gas stream at 100(2) K using ω and φ scans. Data were integrated and scaled using the Bruker SAINT software program. Solution by direct methods (SHELXS) produced a complete phasing model consistent with the proposed structure. All non-hydrogen atoms were refined anisotropically by full-matrix least squares (SHELXL-97). All hydrogen atoms were placed using a riding model. Their positions were constrained relative to their parent atom using the appropriate HFIX command in SHELXL-97. Crystallographic data and structure refinement parameters are summarized in the Supporting Information (SI). Spectroscopy. 1H and 13C NMR spectra were recorded on a Varian VX 500 MHz spectrometer at 298 K, and data were processed using MestReNova software. Chemical shifts are reported relative to tetramethylsilane (δ = 0) and referenced against solvent residual peaks. Microanalyses were performed by NuMega Resonance Laboratories, Inc. High-resolution mass spectrometry (HRMS) spectra were recorded on an Agilent 6230 Accurate-Mass time-of-flight mass spectrometer. UV−vis spectra were recorded on a Shimadzu UV 3600 spectrometer. UV−vis absorbance spectra of 3a and 3b were taken at 700 μM in THF in an airtight cell with a 1 mm path length and CaF2 windows. Spectra of 2a and 2b were taken at 70 μM in DMSO in a quartz cuvette with a 1 cm path length. Density Functional Theory (DFT) Calculations. DFT calculations were performed with the Amsterdam Density Functional (ADF) program suite (version 2012.01).35−37 The triple-ζ Slater-type orbital TZ2P basis set was utilized without frozen cores for all atoms. Relativistic effects were included via the zeroth-order regular approximation.38,39 The BP86 functional and the local density approximation of Vosko, Wilk, and Nusair40 were coupled with the generalized gradient approximation corrections described by Becke41 and Perdew42,43 for electron exchange and correlation, respectively. Frequency calculations were performed to verify that the optimized geometries were minima. Geometry-optimized positional (xyz) coordinates and a sample input file are included in the SI.
■
RESULTS AND DISCUSSION Synthesis of Nickel(II) Precursors. Complexes 1a and 1b (Scheme 1) were synthesized using previously reported procedures.12,33 Anion exchange using NH4PF6 in hot MeOH gave the corresponding hexafluorophosphate (PF6−) salts of these complexes, 2a and 2b, respectively. Crystals suitable for X-ray diffraction were grown by vapor diffusion of diethyl ether into a concentrated MeCN solution of each complex. As noted Scheme 1. Structures of Homoleptic Chelating Bis-NHC Nickel(II) Complexes
3212
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry by others in previous studies of 1a and 1b,12,33 these cations display near-ideal square planarity around the nickel(II) center, with ligand bite angles of 86.39(7)° and 86.64(10)°, respectively (see the SI). This is enforced by the rigidity of the ligands with respect to boat−chair interconversion of the nickelocycle, as verified by diastereotopic 1H NMR shifts for the linking methylene. This rigidity appears to be enforced by the steric clash of neighboring ligand methyl groups. Electrochemical Studies. Complexes 2a and 2b were studied electrochemically in order to explore their redox behavior. CVs were taken in anhydrous DMSO with 0.1 M TBAPF6 as the supporting electrolyte. The CV of 2a shows two one-electron reductions (Figure 1). The first feature is
indicated by a linear dependence of the peak current against the square root of the scan rate (Figures S6 and S7 in the SI). Synthesis and Crystallography of Nickel(0) Species. In order to better understand the redox behavior of these species, their doubly reduced states were synthesized by chemical reduction. These reductions were performed by suspending dibromide salt 1a or 1b in THF followed by the addition of ∼2 equiv of potassium-intercalated graphite (KC8). Filtration of the resulting dark suspensions followed by removal of the solvent under vacuum resulted in isolation of the reduced species 3a and 3b as brick-red and violet powders, respectively. Care was taken during solvent removal because both complexes sublime at room temperature under reduced pressure. Scheme 2 shows the reductive chemistry of these systems. Scheme 2. Reductive Chemistry of Homoleptic Chelating bis-NHC Nickel(II) Complexes
Both complexes are highly sensitive to oxygen and moisture. In the 1H NMR spectrum, the characteristic diastereotopic proton signals of the sterically locked nickel(II) complexes 2a and 2b become magnetically equivalent. Despite complete asymmetry of the solid-state structures (vide infra), the solution-phase 1H NMR spectra show equivalent shifts for all methyl groups, all aryl systems, and all methylene linkers, suggesting structural flexibility in solution. Crystals of 3a and 3b suitable for X-ray diffraction were grown by vapor diffusion of diethyl ether into a concentrated THF solution of each complex. The solid-state structures of 3a and 3b are shown in Figure 2. Upon reduction, the nickel(II) complexes undergo rearrangement to a pseudotetrahedral sawhorse geometry. Because of the steric clash between the methyl groups in the nickel(II) state (enforced by ligand rigidity, as indicated by diasterotopic linking methylene proton shifts), it is expected that there is a large barrier to this rearrangement, requiring significant lengthening or breaking of a Ni−C bond. This is likely the main contributor to both the irreversibility and extremely negative reduction potentials for the corresponding nickel(II) precursors. In the solid-state structures of both 3a and 3b, the bite angles of the ligands remain constrained by the methylene spacer at 92.33(14)° and 91.87(14)° in 3a and 93.44(11)° and 93.87(11)° in 3b. Analogous nickel(0) diphosphine complexes typically adopt a pinched tetrahedral geometry in such a ligand environment. However, these complexes distort toward a sawhorse geometry, particularly in the case of 3a (vide infra). Electronic Spectroscopy. UV−vis spectroscopic studies were carried out in order to experimentally probe the electronic structures of these complexes. The spectra are presented in Figure 3. In 3a, the broad absorbance with λmax at 392 nm is assigned to a metal-to-ligand charge transfer (MLCT), with an
Figure 1. CVs of 2a and 2b under identical conditions. Experiments were performed at a complex concentration of 1 mM in DMSO with 0.1 M TBAPF6 as the supporting electrolyte under a N2 atmosphere with a 3-mm-diameter glassy carbon working electrode, platinum wire counter electrode, and nonaqueous Ag+/Ag reference electrode. Potentials were referenced to an internal Fc standard.
irreversible with a peak anodic current at −2.30 V versus Fc+/Fc. The second reduction is reversible, E1/2 = −2.33 V versus Fc+/Fc with a peak-to-peak separation of 58 mV. These reductive features merge with increasing scan rate (Figure S3 in the SI). These CVs are consistent with an ECE mechanism, wherein an initial reduction from nickel(II) to nickel(I) is followed by a chemical step before subsequent reduction to nickel(0). The second reoxidation is shifted cathodically by ∼460 mV, indicating that this chemical step stabilizes the nickel(I) state relative to the nickel(II) starting state. The merging peaks at increased scan rates are therefore a kinetic phenomenon, reflective of a slow chemical step that occurs on the electrochemical time scale. The CV of 2b, in contrast, shows a fully irreversible twoelectron reduction with a peak cathodic current at −2.00 V versus Fc+/Fc, ∼330 mV more positive than the corresponding reductions of 2a. This is consistent with an ECE mechanism, wherein the second reduction occurs at a potential more positive than the first. At high scan rates, it is possible to rereduce the complex before reversal of the chemical step, allowing isolation of two one-electron features in the absence of the chemical step (Figure S5 in the SI). Variable scan rate studies indicate that both species were freely diffusing, as 3213
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry
Figure 2. Solid-state structures of 3a (left) and 3b (right). Hydrogen atoms and solvent molecules are omitted for clarity, and ellipsoids are set at the 50% probability level. The structure of 3a is positioned to emphasize the open coordination site at the nickel center. Selected bond lengths (Å) and angles (deg) for 3a: Ni1−C1, 1.913(3); Ni1−C2, 1.903(3); Ni1−C3, 1.902(3); Ni1−C4, 1.886(3); C4−Ni1−C3, 91.87(14); C2−Ni1−C1, 92.33(14); C3−Ni1−C2, 102.93(14); C4−Ni1−C1, 112.64(14); C3−Ni1−C1, 119.36(14); C4−Ni1−C2, 139.86(14). Selected bond lengths (Å) and angles (deg) for 3b: Ni1−C1, 1.886(3); Ni1−C2, 1.882(3); Ni1−C3, 1.886(3); Ni1−C4, 1.883(3); C2−Ni1−C1, 93.44(11); C4−Ni1−C3, 93.87(11); C3−Ni1−C1, 108.84(11); C2−Ni1−C4, 112.57(11); C4−Ni1−C1, 121.18(11); C2−Ni1−C3, 129.64(11).
Figure 4. DFT-calculated HOMO (left) and LUMO (right) of 3a and 3b.
Figure 3. UV−vis absorbance spectra of 3a and 3b in THF. Spectra of 2a and 2b in DMSO are included for reference.
extinction coefficient (ϵ) of ∼5600 M−1 cm−1. The broadness of the feature relative to that of 2a may be due to the decreased rigidity of the molecule in solution in combination with the availability of near-degenerate, low-lying ligand-based π* molecular orbitals. Complex 3b has similarly broad features with λmax at 390 and 544 nm and ϵ of ∼7700 and ∼9400 M−1 cm−1, respectively. We attribute the lower-energy absorbance to a MLCT similar to that described for 2a, while the higherenergy feature corresponds to transitions from the highest occupied molecular orbital (HOMO) to a second manifold of π orbitals centered in the benzimidazole structures of the ligands (see Figure S8 in the SI). DFT Analysis. Solid-state structural parameters of 3a and 3b were used to perform gas-phase DFT analysis on these complexes. The agreement between the calculated and solidstate geometries was generally good, with the calculation capturing the sawhorse distortion for both molecules. Relative to the crystal structure, this distortion was slightly underexaggerated in 3a and slightly overexaggerated in 3b (Table S4 in the SI). The HOMO and lowest unoccupied molecular orbital (LUMO) of each complex are shown in Figure 4. In both cases, the HOMO is nickel-based, with the bulk of its
density in the open coordination site, suggesting that this complex may be amenable to reductive chemistry at the nickel center. This HOMO is comprised mainly of Ni pz character, with contributions from parallel carbon-based p orbitals. The symmetry between the metal-based HOMO and the ligandbased LUMO is well matched. As such, strong MLCT bands are expected in the electronic structure (vide supra), explaining the extremely high coloration of these species. The LUMO in 3b is much more delocalized than that in 3a, consistent with the ∼150 nm shift in absorption. Furthermore, both molecules are predicted to have low-lying π* orbitals near in energy to their LUMO, perhaps explaining the broadness of the observed spectral features. Finally, these calculations, in combination with the above crystallographic data, allow confident assignment of a true nickel(0) ground state, with no apparent redox noninnocence in the carbene ligands as proposed in other examples.20 Structural Analysis and Comparisons, τδ. In order to compare the geometries of 3a and 3b to similar complexes, τ4 values were calculated from the largest (α) and second-largest (β) L−Ni−L angles by the method of Yang, Powell, and Houser (eq 1).32 3214
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry
τ4 =
360 − (α + β) 141
qualitatively has a geometry closer to sawhorse than that of any reported nickel(0) diphosphine complex.46−51 This illustrates incompleteness in the popular τ4 index as a measure of the variation in the four-coordinate geometry. Because this index only takes into account the sum of the largest and secondlargest angles, it cannot properly differentiate between “true” sawhorse complexes and those that are tetrahedral but pinched because of ligand bite angle or steric constraints. In terms of shape maps described by Alvarez et al.,52 τ4 is limited to describing the procession along the “spreading” mode, the least distorted continuum of geometries between square and tetrahedral,53 and cannot adequately take into account the effects of “scissor” or “plier” distortions enforced by bidentate ligands.52 In order to address this shortcoming, a modification to τ4 was devised to account for differences between α and β as well as their sum as a measure of the deviation from the ideal tetrahedral geometry (eqs 2 and 3).
(1)
With τ4 values of 0.715 and 0.774, respectively, 3a is best described as a sawhorse complex with tetrahedral distortion and 3b is best described as a distorted tetrahedral complex. The slightly lower τ4 value for 3a can be attributed to increased ligand donation, resulting in distorted sawhorse geometry akin to certain heavy p-block coordination compounds.44,45 This is an especially appropriate structural description of 3a given the Ni p orbital nature of the HOMO of this complex. τ4 values for these complexes are presented alongside those of analogous nickel(0) diphosphine complexes in Table 1. Table 1. Selected Structural Parameters and τ4 Values for 3a, 3b, and Reported Analogues complex
largest angle, α (deg)
second-largest angle, β (deg)
τ4
ref
Ni(dcpm)2a Ni(dppm)2c Ni(diprpe)2d Ni(dppe)2e Ni(dmpp)2f Ni(dppp)2g 3a 3b
136.4b 130.17(3) 127.3(8) 129.3(1) 115.41(2) 120.49(3) 139.86(14) 129.64(11)
134.1b 128.54(3) 127.3(8) 120.2(1) 113.41(4) 115.43(3) 119.36(14) 121.18(11)
0.63 0.72 0.75 0.78 0.93 0.87 0.71 0.77
46 47 48 49 50 51 this work this work
τδ =
360 − (α + β) δ 141
(2)
δ=
β α
(3)
Note that τδ is simply the product of τ4 and a new term, δ, which is the ratio of the second-largest angle β to the largest angle α. This new index preserves many of the features of τ4 that have made this index popular among coordination chemists: it is simple to calculate, it has a limited scale from 0 to 1, and it can be utilized without visual inspection of the structure in question. However, τδ better serves those working with bidentate ligands without sacrificing broader usefulness. τδ values of several idealized geometries and selected examples from the literature are presented in Table 2. The top portion of the table shows values for idealized geometries. Complexes with τδ values near 1.0 remain best described as tetrahedral, while sawhorse geometries now cluster
a dcpm = bis(dicyclohexylphosphino)methane. bNo error values were given in ref 46. cdppm = bis(diphenylphosphino)methane. ddiprpe = 1,2-bis(diisoprophylphosphino)ethane. e dppe = 1,2-bis(diphenylphosphino)ethane. fdmpp = 1,3-bis(dimethylphosphino)propane. gdppp =1 ,3-bis(diphenylphosphino)propane.
Of note in this table is the assignment of Ni(dcpm)20 as the most sawhorse-like of these complexes, despite the α and β angles of this complex clearly indicating a pinched tetrahedral geometry with bite angles of
Synthesis and Structural Studies of Nickel(0) Tetracarbene Complexes with the Introduction of a New Four-Coordinate Geometric Index, τδ Mark H. Reineke, Matthew D. Sampson, Arnold L. Rheingold, and Clifford P. Kubiak* Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, Mail Code 0358, La Jolla, California 92093-0358, United States S Supporting Information *
ABSTRACT: The synthesis and characterization of two homoleptic chelating nickel(0) tetracarbene complexes are reported. These are the first group 10 M(0) (M = Ni, Pd, Pt) tetracarbene complexes. These species have geometries intermediate between C2v sawhorse and tetrahedral and show high UV−vis absorption in the 350−600 nm range, with extinction coefficients (ϵ) between 5600 and 9400 M−1 cm−1. Density functional theory analysis indicates that this high absorptivity is due to metal-to-ligand charge transfer. In order to better describe the unusual geometries encountered in these complexes, an adjustment to the popular τ4 index for fourcoordinate geometries is introduced in order to better delineate between sawhorse and distorted tetrahedral geometries.
■
INTRODUCTION Coordinatively unsaturated group 10 complexes have been successfully used in many catalytic applications, including palladium- and platinum-catalyzed cross-coupling reactions1,2 and C−H activation reactions.3,4 Typically, palladium and platinum have been the primary focus of development in this field. Recently, however, nickel complexes have found growing popularity as earth-abundant analogues to their second- and third-row counterparts.5,6 This surge in popularity has occurred in tandem with the proliferation of N-heterocyclic carbene (NHC) ligands in the coordination chemistry of first-row transition metals, resulting in new and varied reactivity in organonickel chemistry.6−9 Such NHC-stabilized nickel complexes have been reported as catalysts for a variety of transformations,7,10 including aryl halide reduction11 and cross-coupling reactions.12,13 Typically, NHC ligands are incorporated into these systems as σ-donating analogues to two-electron donor ligands such as phosphines or amines.14 It is therefore surprising that, despite the rich reductive electrochemistry of nickel complexes containing phosphorus and nitrogen donors for applications such as carbon dioxide (CO2) reduction15 and proton reduction,16 there are only a handful of reports in the literature discussing the effects of NHC ligands on the electrochemistry or electrocatalytic properties of nickel complexes.17−23 Similarly, isolated nickel(0) complexes containing carbene ligands (the products of reduction of these systems) are unusual, primarily restricted to homoleptic bis-24 and tris-NHC19 complexes, heteroleptic nickel olefin complexes,25−27 and NHC adducts of Ni(CO)4.28−31 © 2015 American Chemical Society
Herein we report the in situ electrochemical generation, direct synthesis, and structural characterization of bis(1,1′methylene-3,3′-diimidazole-1,1′-2,2′-diylidene)nickel(0) (3a) and bis(1,1′-methylene-3,3′-dibenzimidzole-2,2′-diylidene)nickel(0) (3b) from bis(1,1′-methylene-3,3′-diimidazole-1,1′2,2′-diylidene)nickel(II) bromide (1a) and bis(1,1′-methylene3,3′-dibenzimidzole-2,2′-diylidene)nickel(II) bromide (1b), respectively. These are homoleptic NHC analogues to existing nickel diphosphine complexes. 3a and 3b are, to our knowledge, the first isolated and structurally characterized nickel(0) tetracarbene complexes. Furthermore, when these species are compared to nickel(0) diphosphine complexes, it was found that the most popular index for the classification of four-coordinate geometries, τ4,32 is insufficient to distinguish between sawhorse geometries seen in this study and the pinched tetrahedra common in analogous complexes. As such, this index has been modified with a new term to take into account the marked differences between these two coordination geometries. These modifications have been incorporated into a new index for the rapid and convenient classification of fourcoordinate complexes, τδ.
■
EXPERIMENTAL SECTION
Synthesis. All reagents were obtained from commercial suppliers and used without purification unless otherwise noted. All reactions were carried out under a N2 atmosphere using standard Schlenk and glovebox techniques. Acetonitrile (MeCN) and tetrahydrofuran Received: November 24, 2014 Published: March 10, 2015 3211
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry (THF) were dried over activated 3 Å molecular sieves and alumina and degassed prior to use. Dimethyl sulfoxide (DMSO) was dried by distillation of volatiles under vacuum, stored over activated molecular sieves, and degassed prior to use. NiBr2·6H2O was dehydrated dried under vacuum at 60 °C overnight. Complete dehydration was verified gravimetrically. Tetrabutylammonium hexafluorophosphate (TBAPF6) was twice recrystallized from methanol (MeOH) and dried under vacuum. Complexes 1a and 1b were synthesized as previously reported.12,33 KC8 was synthesized as previously reported.34 Note that KC8 is extremely pyrophoric and should be handled only under an inert atmosphere and stored in a freezer at −30 °C. Synthesis of the Hexafluorophosphate Salt 2a. To a solution of the tetracarbene dibromide 1a (499 mg, 0.88 mmol) dissolved in a minimum of boiling MeOH was added a boiling methanolic solution of ammonium hexafluorophosphate (NH4PF6; 502 mg, 2.99 mmol). A pale-yellow precipitate began to form immediately. The reaction was allowed to stand and cool to room temperature. The product was collected by vacuum filtration and dried overnight under vacuum at 60 °C, resulting in a pale-yellow solid (504 mg, 0.72 mmol, 82% yield). 1 H NMR (400 MHz, DMSO-d6): δ 7.70 (s, 4H, CH), 7.31 (s, 4H, CH), 6.68 (dd, 4H, CH2, J = 202.3 and 12.9 Hz), 3.18 (s, 12H, CH3). 13 C NMR (500 MHz, DMSO-d6): δ 172.96 (Ni−C), 123.93, 122.84 (CH), 62.51 (CH2), 36.56 (CH3). ESI-HRMS ([M]2+). Calcd for [C18H24N8Ni]2+: m/z 205.0732. Found: m/z 205.0733. Elem anal. Calcd for C18H24N8NiP2F6: C, 30.84; H, 3.45; N, 15.98. Found: C, 30.85; H, 30.5; N, 16.00. UV−vis (DMSO): λmax = 326 nm. Synthesis of the Hexafluorophosphate Salt 2b. To a solution of the tetracarbene dibromide 1b (100 mg, 0.13 mmol) dissolved in a minimum of boiling MeOH was added a boiling methanolic solution of NH4PF6 (60 mg, 0.36 mmol). A pale-yellow precipitate began to form immediately. The reaction was allowed to stand and cool to room temperature. The product was collected by vacuum filtration and dried overnight under vacuum at 60 °C, resulting in a bright-yellow solid (98 mg, 0.11 mmol, 84% yield). 1H NMR (400 MHz, DMSOd6): δ 8.34 (d, 4H, CH, J = 8.2 Hz), 7.68 (d, 4H, CH, J = 8.1 Hz), 7.54 (m, 8H, CH, CH2), 7.46, (t, 4H, CH, J = 7.7 Hz), 3.49 (s, 12H, CH3). 13 C NMR (500 MHz, DMSO-d6): δ 183.38 (Ni−C), 134.77, 133.66, 124.95, 124.62, 112.52, 111.68 (CH), 122.84 (CH), 57.39 (CH2), 34.57 (CH3). ESI-HRMS ([M]+). Calcd: m/z 305.1046. Found: m/z 305.1046. Elem anal. Calcd for C34H32N8NiP2F6: C, 45.31; H, 3.58; N, 12.43. Found: C, 44.96; H, 3.42; N, 12.45. UV−vis (DMSO): λmax = 270, 340 nm. Synthesis of the Nickel(0) Tetracarbene Complex 3a. To a suspension of the tetracarbene dibromide 1a (150 mg, 0.26 mmol) in THF (15 mL) was added 2.1 equiv of KC8 (75 mg, 0.55 mmol). This suspension was allowed to react at room temperature overnight. The dark suspension was then filtered, reserving the supernatant. Solvent was removed under vacuum to give a dark-red solid product (54 mg, 0.13 mmol, 51% yield). 1H NMR (400 MHz, THF-d8): δ 6.73 (s, 4H, CH), 8.65 (s, 4H, CH), 5.26 (s, 4H, CH2), 3.05 (s, 12H, CH3). 13C NMR (500 MHz, DMSO-d6): δ 121.16, 119.38 (CH), 58.04 (CH2), 31.52 (CH3). Note: no carbenic 13C resonance was observed, even after extended collection times. ESI-HRMS ([M − H]+). Calcd for [C34H31N8Ni]+: m/z 609.2022. Found: m/z 609.2020. UV−vis (THF): λmax = 392 nm. Synthesis of the Nickel(0) Tetracarbene Complex 3b. To a suspension of the tetracarbene dibromide 1b (158 mg, 0.21 mmol) in THF (15 mL) was added 2.1 equiv of KC8 (81 mg, 0.60 mmol). This suspension was allowed to react at room temperature overnight. The dark suspension was then filtered, reserving the supernatant. The solvent was removed under vacuum to give a dark-violet solid product (72 mg, 0.12 mmol, 57% yield). 1H NMR (400 MHz, THF-d8): δ 7.43 (d, 4H, CH), 6.93 (m, 12H, CH), 6.00 (s, 4H, CH2), 3.27 (s, 12H, CH3). 13C NMR (500 MHz, DMSO-d6): δ 209.38 (Ni−C), 137.09, 135.56, 119.68, 119.35, 106.66, 105.30 (CH), 53.91 (CH2), 25.38 (CH3). ESI-HRMS ([M − H]2+). Calcd for [C18H23N8Ni]+: m/z 409.1394. Found: m/z 409.1394. UV−vis (THF): λmax = 390, 544 nm. Electrochemistry. Electrochemical experiments were performed using a BASi Epsilon potentiostat. Cyclic voltammograms (CVs) were performed under nitrogen in a one-compartment cell using a glassy
carbon BASi working electrode (3 mm diameter), a platinum wire counter electrode, and a nonaqueous silver reference electrode from BASi containing a silver wire in a DMSO solution of 0.1 M TBAPF6 and 0.1 mM AgNO3. All experiments were performed in dry DMSO using 0.1 M TBAPF6 as the supporting electrolyte. After the initial CVs were taken, ferrocene (Fc) was added as an internal reference. X-ray Crystallographic Studies. Single-crystal X-ray diffraction studies were carried out on a Bruker Kappa APEX-II CCD diffractometer equipped with Mo Kα radiation (λ = 0.71073 Å). The crystals were mounted on a Cryoloop with Paratone oil, and data were collected under a nitrogen gas stream at 100(2) K using ω and φ scans. Data were integrated and scaled using the Bruker SAINT software program. Solution by direct methods (SHELXS) produced a complete phasing model consistent with the proposed structure. All non-hydrogen atoms were refined anisotropically by full-matrix least squares (SHELXL-97). All hydrogen atoms were placed using a riding model. Their positions were constrained relative to their parent atom using the appropriate HFIX command in SHELXL-97. Crystallographic data and structure refinement parameters are summarized in the Supporting Information (SI). Spectroscopy. 1H and 13C NMR spectra were recorded on a Varian VX 500 MHz spectrometer at 298 K, and data were processed using MestReNova software. Chemical shifts are reported relative to tetramethylsilane (δ = 0) and referenced against solvent residual peaks. Microanalyses were performed by NuMega Resonance Laboratories, Inc. High-resolution mass spectrometry (HRMS) spectra were recorded on an Agilent 6230 Accurate-Mass time-of-flight mass spectrometer. UV−vis spectra were recorded on a Shimadzu UV 3600 spectrometer. UV−vis absorbance spectra of 3a and 3b were taken at 700 μM in THF in an airtight cell with a 1 mm path length and CaF2 windows. Spectra of 2a and 2b were taken at 70 μM in DMSO in a quartz cuvette with a 1 cm path length. Density Functional Theory (DFT) Calculations. DFT calculations were performed with the Amsterdam Density Functional (ADF) program suite (version 2012.01).35−37 The triple-ζ Slater-type orbital TZ2P basis set was utilized without frozen cores for all atoms. Relativistic effects were included via the zeroth-order regular approximation.38,39 The BP86 functional and the local density approximation of Vosko, Wilk, and Nusair40 were coupled with the generalized gradient approximation corrections described by Becke41 and Perdew42,43 for electron exchange and correlation, respectively. Frequency calculations were performed to verify that the optimized geometries were minima. Geometry-optimized positional (xyz) coordinates and a sample input file are included in the SI.
■
RESULTS AND DISCUSSION Synthesis of Nickel(II) Precursors. Complexes 1a and 1b (Scheme 1) were synthesized using previously reported procedures.12,33 Anion exchange using NH4PF6 in hot MeOH gave the corresponding hexafluorophosphate (PF6−) salts of these complexes, 2a and 2b, respectively. Crystals suitable for X-ray diffraction were grown by vapor diffusion of diethyl ether into a concentrated MeCN solution of each complex. As noted Scheme 1. Structures of Homoleptic Chelating Bis-NHC Nickel(II) Complexes
3212
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry by others in previous studies of 1a and 1b,12,33 these cations display near-ideal square planarity around the nickel(II) center, with ligand bite angles of 86.39(7)° and 86.64(10)°, respectively (see the SI). This is enforced by the rigidity of the ligands with respect to boat−chair interconversion of the nickelocycle, as verified by diastereotopic 1H NMR shifts for the linking methylene. This rigidity appears to be enforced by the steric clash of neighboring ligand methyl groups. Electrochemical Studies. Complexes 2a and 2b were studied electrochemically in order to explore their redox behavior. CVs were taken in anhydrous DMSO with 0.1 M TBAPF6 as the supporting electrolyte. The CV of 2a shows two one-electron reductions (Figure 1). The first feature is
indicated by a linear dependence of the peak current against the square root of the scan rate (Figures S6 and S7 in the SI). Synthesis and Crystallography of Nickel(0) Species. In order to better understand the redox behavior of these species, their doubly reduced states were synthesized by chemical reduction. These reductions were performed by suspending dibromide salt 1a or 1b in THF followed by the addition of ∼2 equiv of potassium-intercalated graphite (KC8). Filtration of the resulting dark suspensions followed by removal of the solvent under vacuum resulted in isolation of the reduced species 3a and 3b as brick-red and violet powders, respectively. Care was taken during solvent removal because both complexes sublime at room temperature under reduced pressure. Scheme 2 shows the reductive chemistry of these systems. Scheme 2. Reductive Chemistry of Homoleptic Chelating bis-NHC Nickel(II) Complexes
Both complexes are highly sensitive to oxygen and moisture. In the 1H NMR spectrum, the characteristic diastereotopic proton signals of the sterically locked nickel(II) complexes 2a and 2b become magnetically equivalent. Despite complete asymmetry of the solid-state structures (vide infra), the solution-phase 1H NMR spectra show equivalent shifts for all methyl groups, all aryl systems, and all methylene linkers, suggesting structural flexibility in solution. Crystals of 3a and 3b suitable for X-ray diffraction were grown by vapor diffusion of diethyl ether into a concentrated THF solution of each complex. The solid-state structures of 3a and 3b are shown in Figure 2. Upon reduction, the nickel(II) complexes undergo rearrangement to a pseudotetrahedral sawhorse geometry. Because of the steric clash between the methyl groups in the nickel(II) state (enforced by ligand rigidity, as indicated by diasterotopic linking methylene proton shifts), it is expected that there is a large barrier to this rearrangement, requiring significant lengthening or breaking of a Ni−C bond. This is likely the main contributor to both the irreversibility and extremely negative reduction potentials for the corresponding nickel(II) precursors. In the solid-state structures of both 3a and 3b, the bite angles of the ligands remain constrained by the methylene spacer at 92.33(14)° and 91.87(14)° in 3a and 93.44(11)° and 93.87(11)° in 3b. Analogous nickel(0) diphosphine complexes typically adopt a pinched tetrahedral geometry in such a ligand environment. However, these complexes distort toward a sawhorse geometry, particularly in the case of 3a (vide infra). Electronic Spectroscopy. UV−vis spectroscopic studies were carried out in order to experimentally probe the electronic structures of these complexes. The spectra are presented in Figure 3. In 3a, the broad absorbance with λmax at 392 nm is assigned to a metal-to-ligand charge transfer (MLCT), with an
Figure 1. CVs of 2a and 2b under identical conditions. Experiments were performed at a complex concentration of 1 mM in DMSO with 0.1 M TBAPF6 as the supporting electrolyte under a N2 atmosphere with a 3-mm-diameter glassy carbon working electrode, platinum wire counter electrode, and nonaqueous Ag+/Ag reference electrode. Potentials were referenced to an internal Fc standard.
irreversible with a peak anodic current at −2.30 V versus Fc+/Fc. The second reduction is reversible, E1/2 = −2.33 V versus Fc+/Fc with a peak-to-peak separation of 58 mV. These reductive features merge with increasing scan rate (Figure S3 in the SI). These CVs are consistent with an ECE mechanism, wherein an initial reduction from nickel(II) to nickel(I) is followed by a chemical step before subsequent reduction to nickel(0). The second reoxidation is shifted cathodically by ∼460 mV, indicating that this chemical step stabilizes the nickel(I) state relative to the nickel(II) starting state. The merging peaks at increased scan rates are therefore a kinetic phenomenon, reflective of a slow chemical step that occurs on the electrochemical time scale. The CV of 2b, in contrast, shows a fully irreversible twoelectron reduction with a peak cathodic current at −2.00 V versus Fc+/Fc, ∼330 mV more positive than the corresponding reductions of 2a. This is consistent with an ECE mechanism, wherein the second reduction occurs at a potential more positive than the first. At high scan rates, it is possible to rereduce the complex before reversal of the chemical step, allowing isolation of two one-electron features in the absence of the chemical step (Figure S5 in the SI). Variable scan rate studies indicate that both species were freely diffusing, as 3213
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry
Figure 2. Solid-state structures of 3a (left) and 3b (right). Hydrogen atoms and solvent molecules are omitted for clarity, and ellipsoids are set at the 50% probability level. The structure of 3a is positioned to emphasize the open coordination site at the nickel center. Selected bond lengths (Å) and angles (deg) for 3a: Ni1−C1, 1.913(3); Ni1−C2, 1.903(3); Ni1−C3, 1.902(3); Ni1−C4, 1.886(3); C4−Ni1−C3, 91.87(14); C2−Ni1−C1, 92.33(14); C3−Ni1−C2, 102.93(14); C4−Ni1−C1, 112.64(14); C3−Ni1−C1, 119.36(14); C4−Ni1−C2, 139.86(14). Selected bond lengths (Å) and angles (deg) for 3b: Ni1−C1, 1.886(3); Ni1−C2, 1.882(3); Ni1−C3, 1.886(3); Ni1−C4, 1.883(3); C2−Ni1−C1, 93.44(11); C4−Ni1−C3, 93.87(11); C3−Ni1−C1, 108.84(11); C2−Ni1−C4, 112.57(11); C4−Ni1−C1, 121.18(11); C2−Ni1−C3, 129.64(11).
Figure 4. DFT-calculated HOMO (left) and LUMO (right) of 3a and 3b.
Figure 3. UV−vis absorbance spectra of 3a and 3b in THF. Spectra of 2a and 2b in DMSO are included for reference.
extinction coefficient (ϵ) of ∼5600 M−1 cm−1. The broadness of the feature relative to that of 2a may be due to the decreased rigidity of the molecule in solution in combination with the availability of near-degenerate, low-lying ligand-based π* molecular orbitals. Complex 3b has similarly broad features with λmax at 390 and 544 nm and ϵ of ∼7700 and ∼9400 M−1 cm−1, respectively. We attribute the lower-energy absorbance to a MLCT similar to that described for 2a, while the higherenergy feature corresponds to transitions from the highest occupied molecular orbital (HOMO) to a second manifold of π orbitals centered in the benzimidazole structures of the ligands (see Figure S8 in the SI). DFT Analysis. Solid-state structural parameters of 3a and 3b were used to perform gas-phase DFT analysis on these complexes. The agreement between the calculated and solidstate geometries was generally good, with the calculation capturing the sawhorse distortion for both molecules. Relative to the crystal structure, this distortion was slightly underexaggerated in 3a and slightly overexaggerated in 3b (Table S4 in the SI). The HOMO and lowest unoccupied molecular orbital (LUMO) of each complex are shown in Figure 4. In both cases, the HOMO is nickel-based, with the bulk of its
density in the open coordination site, suggesting that this complex may be amenable to reductive chemistry at the nickel center. This HOMO is comprised mainly of Ni pz character, with contributions from parallel carbon-based p orbitals. The symmetry between the metal-based HOMO and the ligandbased LUMO is well matched. As such, strong MLCT bands are expected in the electronic structure (vide supra), explaining the extremely high coloration of these species. The LUMO in 3b is much more delocalized than that in 3a, consistent with the ∼150 nm shift in absorption. Furthermore, both molecules are predicted to have low-lying π* orbitals near in energy to their LUMO, perhaps explaining the broadness of the observed spectral features. Finally, these calculations, in combination with the above crystallographic data, allow confident assignment of a true nickel(0) ground state, with no apparent redox noninnocence in the carbene ligands as proposed in other examples.20 Structural Analysis and Comparisons, τδ. In order to compare the geometries of 3a and 3b to similar complexes, τ4 values were calculated from the largest (α) and second-largest (β) L−Ni−L angles by the method of Yang, Powell, and Houser (eq 1).32 3214
DOI: 10.1021/ic502792q Inorg. Chem. 2015, 54, 3211−3217
Article
Inorganic Chemistry
τ4 =
360 − (α + β) 141
qualitatively has a geometry closer to sawhorse than that of any reported nickel(0) diphosphine complex.46−51 This illustrates incompleteness in the popular τ4 index as a measure of the variation in the four-coordinate geometry. Because this index only takes into account the sum of the largest and secondlargest angles, it cannot properly differentiate between “true” sawhorse complexes and those that are tetrahedral but pinched because of ligand bite angle or steric constraints. In terms of shape maps described by Alvarez et al.,52 τ4 is limited to describing the procession along the “spreading” mode, the least distorted continuum of geometries between square and tetrahedral,53 and cannot adequately take into account the effects of “scissor” or “plier” distortions enforced by bidentate ligands.52 In order to address this shortcoming, a modification to τ4 was devised to account for differences between α and β as well as their sum as a measure of the deviation from the ideal tetrahedral geometry (eqs 2 and 3).
(1)
With τ4 values of 0.715 and 0.774, respectively, 3a is best described as a sawhorse complex with tetrahedral distortion and 3b is best described as a distorted tetrahedral complex. The slightly lower τ4 value for 3a can be attributed to increased ligand donation, resulting in distorted sawhorse geometry akin to certain heavy p-block coordination compounds.44,45 This is an especially appropriate structural description of 3a given the Ni p orbital nature of the HOMO of this complex. τ4 values for these complexes are presented alongside those of analogous nickel(0) diphosphine complexes in Table 1. Table 1. Selected Structural Parameters and τ4 Values for 3a, 3b, and Reported Analogues complex
largest angle, α (deg)
second-largest angle, β (deg)
τ4
ref
Ni(dcpm)2a Ni(dppm)2c Ni(diprpe)2d Ni(dppe)2e Ni(dmpp)2f Ni(dppp)2g 3a 3b
136.4b 130.17(3) 127.3(8) 129.3(1) 115.41(2) 120.49(3) 139.86(14) 129.64(11)
134.1b 128.54(3) 127.3(8) 120.2(1) 113.41(4) 115.43(3) 119.36(14) 121.18(11)
0.63 0.72 0.75 0.78 0.93 0.87 0.71 0.77
46 47 48 49 50 51 this work this work
τδ =
360 − (α + β) δ 141
(2)
δ=
β α
(3)
Note that τδ is simply the product of τ4 and a new term, δ, which is the ratio of the second-largest angle β to the largest angle α. This new index preserves many of the features of τ4 that have made this index popular among coordination chemists: it is simple to calculate, it has a limited scale from 0 to 1, and it can be utilized without visual inspection of the structure in question. However, τδ better serves those working with bidentate ligands without sacrificing broader usefulness. τδ values of several idealized geometries and selected examples from the literature are presented in Table 2. The top portion of the table shows values for idealized geometries. Complexes with τδ values near 1.0 remain best described as tetrahedral, while sawhorse geometries now cluster
a dcpm = bis(dicyclohexylphosphino)methane. bNo error values were given in ref 46. cdppm = bis(diphenylphosphino)methane. ddiprpe = 1,2-bis(diisoprophylphosphino)ethane. e dppe = 1,2-bis(diphenylphosphino)ethane. fdmpp = 1,3-bis(dimethylphosphino)propane. gdppp =1 ,3-bis(diphenylphosphino)propane.
Of note in this table is the assignment of Ni(dcpm)20 as the most sawhorse-like of these complexes, despite the α and β angles of this complex clearly indicating a pinched tetrahedral geometry with bite angles of
