Synthesis and Characterization of Tantalum-Based Early-Late Heterobimetallic Complexes Supported by 2-(diphenylphosphino)pyrrolide Ligands

Abstract

The synthesis of the metalloligand Ta(κ²-NP)₃Cl₂ (NP = 2-diphenylphosphinopyrrolide) and its coordination chemistry with group 9 and 10 metals is reported. Treatment of Ta(κ²-NP)₃Cl₂ with group 9 and 10 metals resulted in clean formation of the heterobimetallic complexes Cl₂Ta(μ₂-NP)₃M (M = Ni (2), Pd (3)) or Cl₂Ta(μ₂-NP)₃MCl (M = Rh (4), Ir (5)). Each pair of complexes is isostructural and contains three phosphinopyrrolide ligands that bridge the metal centers. The d¹⁰ or d⁸ complexes are all diamagnetic and X-ray crystallographic analysis reveals similarly short metal-metal distances, ranging from 2.2979(5) Å to 2.4366(2) Å. Despite the similar bonding metrics in 2-5, treatment with an L type donor (2,6-dimethylphenylisocyanide (CNXylyl)) reveals 3 different coordination geometries in TaNi(CNXylyl) (6), TaPd(CNXylyl) (7), and TaIr(CNXylyl) (8). While complexes 6, 7, and 8 all bind the isocyanide at the late metal, ligand rearrangements are observed in the first row complex 6. Complex 7 binds the isocyanide in the axial position while equatorial binding is observed in 8. All isocyanide adducts maintain close metal-metal contacts in the solid state.

Graphical Abstract

The synthesis of the metalloligand Ta(κ²-NP)₃Cl₂ (NP = 2-diphenylphosphinopyrrolide) and its coordination chemistry with group 9 and 10 metals is reported. Treatment of Ta(κ²-NP)₃Cl₂ with group 9 and 10 metals resulted in clean formation of the heterobimetallic complexes Cl₂Ta(μ₂-NP)₃M (M = Ni (2), Pd (3)) or Cl₂Ta(μ₂-NP)₃MCl (M = Rh (4), Ir (5)).

Introduction

Pioneered by Cotton’s characterization of a quadruple bond in [Re₂Cl₈]²⁻, the study of metal-metal bonds has received significant attention over the past 50 years[1–4]. While initial studies focused heavily on homobimetallic complexes, advances in ligand design have enabled the study of metal-metal bonding in early-late heterobimetallic complexes[5–8]. Development of hard/soft bifunctional ligands (often bearing hard oxygen/nitrogen and soft phosphorous donors) by the Wolczanski[9], Nagashima[10–13], Lu[14–18], Thomas[19–29] and others has resulted in a plethora of early-late bimetallic complexes containing strong metal-metal bonds. Several trends have emerged from these classes of complexes: metal-metal bonding is often strongest when the combined d-electron count of the two metals is ten, and bond strength decreases as the metal centers become more disparate[30]. For example, Lu reported a series of Cr-M (M = Mn, Fe, Co, and Ni) complexes supported by a double-decker N((o-C₆H₄)NCH₂PⁱPr₂)₃ ligand in a trigonal framework[31], and as the apical metal is varied the bond order decreases from a quintuple bond in Cr-Mn to a single bond in Cr-Ni.

Despite significant synthetic efforts, many bimetallic combinations of the transition metal block remain underexplored. There are many examples of early-late bimetallic complexes featuring group 5/group 9 or 10 metals, however the presence of bridging ligands often forces close contact between metal centers and convolutes the bonding picture, making it difficult to extract meaningful trends (Figure 1). Most of these complexes contain single atom bridging ligands (e.g. imidos, hydrides, alkylidenes, or carbonyls) and/or long metal-metal distances with formal shortness ratios (FSR) > 1.1, indicative of weak or no direct metal-metal interaction[32–34]. For example, Bergman (Figure 1A) has used methylene bridges derived from Cp₂Ta(CH₂)(CH₃) to synthesize Ta-Pd, Ta-Co, and Ta-Ir complexes that feature two methylene bridges, resulting in M-M distances that range from 2.7 Å to 2.8 Å depending on the metal pairing [35–39]. Nikonov reported a similar Ta-Rh complex (Figure 1B) supported by bridging phosphides which catalyzes the hydrosilylation of ketones[40]. Mashima has reported Ta-Ni, Ta-Rh, and Ta-Ir complexes derived from tantalocyclopentadienes (Figure 1C), where the metallacycle π system donates into the late metal center which in turn datively donates to Ta[41, 42]. The Ta-Ni distance is 2.5841(7) Å, likely indicative of weak metal-metal bonding.

Figure 1:

Select examples of bimetallic complexes group 5 metals paired with group 9 and 10 metals.

In fact, there are relatively few examples of strong metal-metal bonds in early-late bimetallic complexes featuring group 5 metals. Suzuki—and more recently Camp—published a number of Ta-Ir complexes (Figure 1D) that feature close, unsupported metal-metal double bonds with distances of ~2.45 Å [43]-[44]. Lu has synthesized a series V-M (M = Fe, Co, and Ni) complexes (Figure 1E) that span from metal-metal triple bonds (V-Fe) to no bond (V-Ni)[45], while Thomas has also made V-M (M = Fe, Co, Ni, and Cu) bimetallics.[16] Thomas’ Nb-Fe complex (Figure 1F) was the first report of a metal-metal triple bond in an early-late bimetallic complex containing Nb.[17] Building on this, a suite of bimetallic complexes containing Nb oxos were reported.[22]

We have previously described the ability of phosphinopyrrolide ligands to facilitate metal-metal bonding in group 4-group 10 bimetallic complexes and sought to expand our efforts to the group 5 triad[46–49]. Herein, we report the synthesis of a Ta tris (phosphinopyrrolide) metalloligand and its successful coordination to group 9 and 10 metals (Figure 1G), resulting in complexes with Ta-M bond distances significantly shorter than those previously reported.

Materials and Methods

General Considerations and Instrumentation.

All air and moisture sensitive compounds were manipulated in a glovebox under a nitrogen atmosphere. Solvents for air and moisture sensitive reactions were vacuum transferred from sodium benzophenone ketyl (THF, Et₂O, pentane, d_6-benzene and d₈-toluene) or predried by passing through activated alumina columns of a Pure Process Technology solvent purification system. Li(NP)[46], Ni(COD)₂[50], 2,6-xylyl isocyanide[51] and TaCl₅(THF)[52] were prepared according to literature procedure. Pd(P^tBu₃)₂, [Rh(COD)Cl]₂, and [Ir(COD)Cl]₂ were purchased from Strem Chemicals and used without further purification. ¹H and ³¹P spectra were recorded on Varian INOVA 500 MHz, Bruker Avance III 400 MHz, or Bruker AVANCE III 500 MHz spectrometers. Complexes 2-8 are quite insoluble, which prevented proper ¹³C NMR characterization. Chemical shifts are reported with respect to residual protio solvent impurity for ¹H (s, 7.16 ppm for C₆D₅H), a PPh₃ standard for ³¹P (s, −6 ppm in C₆D₆), and solvent carbons for ¹³C (p, 67.21 ppm for THF-d₈). Attempts at elemental analysis of these air-sensitive complexes were unsuccessful.

X-ray Crystal Data: General Procedure.

Crystals were removed quickly from a scintillation vial to a microscope slide coated with oil. Samples were selected and mounted on the tip of a 0.1 mm diameter glass capillary. Data collection was carried out on a Bruker APEX II CCD diffractometer with a 0.71073 Å Mo Kα source. The data intensity was corrected for absorption and decay (SADABS).[53] Final cell constants were obtained from least-squares fits from all reflections. Crystal structure solution was done through intrinsic phasing (SHELXT-2014/5),[54] which provided most nonhydrogen atoms. Full matrix least-squares/difference Fourier cycles were performed (using SHELXL-2016/6 and GUI ShelXle)[55, 56] to locate the remaining non-hydrogen atoms. All non-hydrogen atoms were refined with anisotropic displacement parameters. Hydrogen atoms were placed in ideal positions and refined as riding atoms with relative isotropic displacement parameters. A disordered THF molecule was removed from the unit cell of 4 and 6 using Platon SQUEEZE.[57] Details regarding refined data and cell parameters are available in Table 1 and 2.

Table 1.

Crystal and refinement data for complexes 1–4.

	1	2	3	4
CCDC Number	1842024	1842023	1842026	1842029
Empirical Formula	C48H39N3P3Cl2Ta	C48H39N3P3Cl2TaNi • 3(C6H6)	C48H39N3P3Cl2TaPd	C48H39N3P3Cl3TaRh
Formula weight	1002.58	1295.60	1108.98	1140.95
T (K)	100(2)	100(2)	100(2)	100(2)
a, Å	15.0010(13)	12.5097(3)	23.2731(19)	11.82400
b, Å	16.3582(12)	23.6118(6)	11.6716(7)	11.96200
c, Å	16.7487(13)	18.9624(5)	19.2249(14)	19.99500
α, deg	90	90	90	74.6100
β, deg	92.923(3)	94.487(1)	125.576(2)	85.0200
γ, deg	90	90	90	61.3200
Volume, Å3	4104.6(6)	5620.0(2)	4247.4(5)	2389.383
Z	4	4	4	2
Crystal System	Monoclinic	Monoclinic	Monoclinic	Triclinic
Space Group	P2(1)/n	P2(1)/n	Cc	P-1
dcalc, g/cm3	1.622	1.531	1.734	1.586
θ Range, deg	2.248 to 30.534	2.367 to 28.949	2.7835 to 28.269	2.4765 to 30.357
μ, mm−1	2.965	2.506	3.276	2.937
Abs. Correction	Multi-scan	Multi-scan	Multi-scan	Multi-scan
GOF	1.179	1.072	1.122	1.080
R1,a	R1 = 0.0408	R1 = 0.0497	R1 = 0.0346	R1 = 0.0488
wR2b [I>2σ (I)]	wR2 = 0.1237	wR2 = 0.1058	wR2 = 0.1116	wR2 = 0.1351

^aR₁ = Σ||F_o| - |F_c||/Σ|F_o|.

^bwR₂ = [Σ[w(F_o²-F_c²)²]/Σ[w(F_o²)²]^1/2.

Table 2.

Crystal and refinement data for complexes 5–8.

	5	6	7	8
CCDC Number	1842025	1842028	1842030	1842022
Empirical Formula	C48H39N3P3Cl3TaIr • 2(C4H8O)	C57H48Cl2N4P3TaPd	C57H48Cl2N4P3TaNi • 2(C4H8O)	C57H48Cl3N4P3TaIr • 2(C6H6) • (C4H8O)
Formula weight	1374.46	1240.16	1336.66	1589.75
T (K)	100(2)	100(2)	123(2)	100(2)
a, Å	11.9034(7)	35.502(3)	11.37200	12.2652(8)
b, Å	15.5335(10)	37.018(2)	16.44400	15.9047(10)
c, Å	27.7150(19)	30.510(2)	16.66900	18.3995(12)
α, deg	90	90	92.2100	83.035(2)
β, deg	98.146(2)	90	90.8100	80.491(2)
γ, deg	90	90	92.2100	86.035(2)
Volume, Å3	5072.8(6)	40097(5)	3062.406	3237.8(4)
Z	4	32	2	2
Crystal System	Monoclinic	Orthorhombic	Triclinic	Triclinic
Space Group	P2(1)/c	Fdd2	P-1	P-1
dcalc, g/Cm3	1.800	1.643	1.450	1.631
θ Range, deg	2.374 to 30.482	2.543 to 28.271	2.355 to 36.350	2.440 to 28.233
μ, mm−1	5.075	2.786	2.305	3.998
Abs. Correction	Multi-scan	Multi-scan	Multi-scan	Multi-scan
GOF	1.084	1.037	1.100	1.031
R1,a	R1 = 0.0355	R1 = 0.0456	R1 = 0.0408	R1 = 0.0406
wR2b [I>2σ(I)]	wR2 = 0.0734	wR2 = 0.0928	wR2 = 0.1055	wR2 = 0.1051

^aR₁ = Σ||F_o| - |F_c||/Σ|F_o|.

^bwR₂ = [Σ[w(F_o²-F_c²)²]/Σ[w(F_o²)²]^1/2.

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂, Ta(NP)₃Cl₂ (1).

Solid TaCl₅(THF) (287 mg, 0.667 mmol, 1 equiv) was dissolved in 5 mL benzene and Li(NP) (514 mg, 2.00 mmol, 3 equiv) was added as a solid. The solution immediately turned a yellow/orange and was stirred for 30 minutes. The mixture was filtered over celite and volatiles were removed in vacuo to yield 621 mg of 1 as a yellow-orange powder in 93% yield. Single crystals of 1 were grown by vapor diffusion of pentane into a concentrated benzene solution of 1. ¹H NMR (300 MHz, C 6D₆) δ, ppm: 5.95–5.99 (br, 2H), 6.33–6.38 (br, 2H), 6.50–6.55 (br, 2H), 6.65–6.70 (br, 1H, aryl), 6.8 to 7.10 (br, 19H, aryl), 7.20–7.24 (br, 1H), 7.32–7.38 (br, 1H), 7.55–7.63 (br, 4H), 7.71–7.82 (br, 8H, aryl), 8.46–8.52 (br, 1H, aryl). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: −17.75 (s, 2P), −8.39 (s, 1P). ¹³C NMR (126 MHz, THF-d₈) δ, ppm: 111.4 (d), 111.9, 118.1, 119.2, 128.3 (d), 128.5 (app m), 128.8 (app s), 130.3, 130.6 (d), 132.6, 132.7 (d), 133.4 (app s), 134.0, 134.4, 134.9 (d), 135.9 (app m),

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂Ni, Ta(NP)₃Cl₂Ni (2).

Solid TaCl₂NP₃ (300 mg, 0.299 mmol, 1 equiv) was dissolved in 4 mL benzene and Ni(COD)₂ (83 mg, 0.299 mmol, 1 equiv) was added. The solution was stirred for 30 minutes then reduced to half volume under vacuum and filtered over a fine frit to yield 240 mg of 2, as a dark powder in 75% yield. Single crystals of 2 were grown by vapor diffusion of pentane into a concentrated 50:50 benzene/THF solution of 2. ¹H NMR (300 MHz, 1:1 C₆D₆:THF) δ, ppm: 6.3 to 7.15 (br, 22H), 7.60 to 7.95 (br, 3H), 8.1 to 8.3 (br, 1H), 8.6–9.3 (br, 1H). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: −14.2 to −12.0 (br, 1P), −8.5 to −7.0 (br, 2P).

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂Pd, Ta(NP)₃Cl₂Pd (3).

Solid TaCl₂NP₃ (207 mg, 0.206 mmol, 1 equiv) was dissolved in 2.5 mL benzene and Pd(P^tBu₃)₂ (105 mg, 0.206 mmol, 1 equiv) was added. The solution was stirred for 30 minutes then 2.5 mL of THF was added. 3 was isolated as a dark powder by slow evaporation of pentane into the concentrated 50:50 benzene/THF solution of crude 3. The solution was filtered over a fine frit and washed with pentane to give 125 mg (54% yield) of 3 as a dark powder. Single crystals were grown by vapor diffusion of pentane into a concentrated benzene solution of 3. ¹H NMR (300 MHz, 1:1 C₆D₆:THF) δ, ppm: 6.05 (br, 2H), 6.18 (br, 2H), 6.29 (br, 2H), 6.65–7.10 (m, 30H), 7.55–7.65 (br, 4H), 7.77–7.83 (br, 2H), 8.52 – 8.56 (br, 1H). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: −6.48 (t, 1P, J_PP = 83.3 Hz), 2.84 (d, 2P, J_PP = 83.3 Hz).

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂RhCl, Ta(NP)₃Cl₂RhCl (4).

Solid TaCl₂NP₃ (147 mg, 0.146 mmol, 1 equiv) was dissolved in 2.5 mL benzene and [Rh(COD)Cl]₂ (36 mg, 0.073, 0.5 equiv) was added. The solution was stirred for 30 minutes, then 2.5 mL of THF was added. Single crystals were grown by slow evaporation of pentane into a concentrated 50:50 benzene/THF solution of crude 4. The resulting yellow/orange crystals were washed with pentane and dried to give 160 mg (89 % yield) of 4. ¹H NMR (300 MHz, 1:1 C₆D₆:THF) δ, ppm: 5.87–5.90 (s, 1H), 6.03–6.06 (br, 1H), 6.10–6.14 (br, 2H), 6.31–6.35 (br, 2H), 6.77–6.89 (br, 4H), 6.55–6.68 (br, 12H), 6.77–6.89 (m, 4H), 7.03–7.11 (br, 6H), 7.71–7.79 (br, 4H), 8.17–8.21 (br, 2H), 8.47–8.51 (br, 1H). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: 10.54 (dt, J_PP = 20.9 Hz, J_RhP = 140.4 Hz, 1P), 6.09 (dd, J_PP = 20.9 Hz, J_RhP = 105.8 Hz, 2P).

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂IrCl, Ta(NP)₃Cl₂IrCl (5).

Solid TaCl₂NP₃ (181 mg, 0.180 mmol, 1 equiv) was dissolved in 2.5 mL benzene and [Ir(COD)Cl]₂ (60.5 mg, 0.090, 0.5 equiv) was added. The solution was stirred for 30 minutes then 2.5 mL THF was added. Single crystals were grown by slow evaporation of pentane into a concentrated 50:50 benzene/THF solution of crude 5. The resulting yellow/orange crystals were washed with pentane and dried to give 100 mg (45 % yield) of 5. ¹H NMR (300 MHz, 1:1 C₆D₆:THF) δ, ppm: 5.91 (m, 1H), 6.10–6.13 (br, 3H), 6.32 (m, 2H) 6.37–6.45 (br, 3H), 6.51–6.60 (br, 4H), 6.70–6.60 (br, 8H), 6.84–6.92 (m, 5H), 7.08–7.13 (m, 7H), 7.64–7.69 (m, 4H), 8.19 (m, 2H), 8.49 (m, 1H). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: −1.25 (d, J_PP = 12.1 Hz, 2P), −18.03 (t, J_PP = 12.1 Hz, 1P).

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂Ni(CN(C₈H₉), Cl₂Ta(NP)₃Ni(CNXylyl) (6).

Solid 2 (20 mg, 0.019 mmol, 1 equiv) was dissolved in 1 mL of a 50:50 THF/C₆D₆ solution and 2,6-xylylNC was added (2.5 mg, 0.019 mmol, 1 equiv). The solution was layered with pentane and allowed to crystallize. The resulting crystals were washed with pentane and dried to give crystalline 6 quantitatively. Dissolution of the crystals results in 3 different isomers, as determined by ³¹P NMR. ¹H NMR (300 MHz, 1:1 C₆D₆) δ, ppm: 1.50 (s, 2H), 1.72 (s, 6H), 1.79 (s, 5H), 5.40–5.60 (m, 2H), 5.80–5.95 (br, 4H), 6.20–7.10 (m, 50H), 7.20–8.10 (m, 33H), 8.15–8.40 (br, 6H), 8.40–8.60 (br, 2H), 9.18–9.30 (br, 1H). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: Species A: −45.45 (t, J_PP = 9.6 Hz, 1P), −2.05 (d, J_PP = 9.6 Hz, 2P). Species B: −19.84 (d, J_PP = 25.0 Hz, 1P), −1.26 (d, J_PP = 74.4P), 10.28 (dd, J_PP = 74.4, 25.0 Hz, 1P). Species C: −37.33 (d, J_PP = 4.2 Hz, 1P), −0.59 (d, J_PP = 95.9 Hz, 1P), 6.39 (dd, J_PP = 95.9, 4.2 Hz, 1P).

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂Pd(CN(C₈H₉), Cl₂Ta(NP)₃Pd(CNXylyl) (7).

Solid 3 (20 mg, 0.018 mmol, 1 equiv) was dissolved in 1 mL of a 50:50 THF/C₆D₆ solution and 2,6-xylylNC was added (2.36 mg, 0.018 mmol, 1 equiv). The solution was layered with pentane and allowed to crystallize. The resulting crystals were washed with pentane and dried to give crystalline 7 quantitatively. ³¹P NMR (121 MHz, 1:1 C₆D₆:THF) δ, ppm: −6 to −4 (br, 2P), 9.66 (d, J_PP = 66.8 Hz, 1P).

Synthesis of Ta((2-Ph₂P)C₄H₃N)₃Cl₂IrCl(CN(C₈H₉), Cl₂Ta(NP)₃IrCl(CNXylyl) (8).

Solid 5 (20 mg, 0.016 mmol, 1 equiv) was dissolved in 1 mL of a 50:50 THF/C₆D₆ solution and 2,6-xylylNC was added (2.13 mg, 0.016 mmol, 1 equiv). The solution was layered with pentane and allowed to crystallize. The resulting crystals were washed with pentane and dried to give crystalline 8 quantitatively. ¹H NMR (300 MHz, 1:1 C₆D₆:THF) δ, ppm: 1.85 (s, 6H), 6.01–6.03 (m, 1H), 6.11–6.15 (m, 3H), 6.18–6.21 (m, 2H), 6.57–6.60 (m, 6H), 6.69–6.73 (m, 4H), 6.75–6.85 (m, 12H), 7.18–7.23 (br, 4H), 7.28–7.41 (br, 3H), 7.89–7.95 (m, 4H), 8.19–8.22 (m, 2H), 8.49–8.50 (m, 1H). ³¹P NMR (121 MHz, C₆D₆) δ, ppm: −35.26 (t, J_PP = 21.5 Hz, 1P), −32.25 to −31.0 (br, 2P).

Results and Discussion

Salt metathesis of TaCl₅(THF) with 3 equiv Li(NP) (NP = 2-(diphenylphosphino)pyrrolide) gives Ta(κ²-NP)₃Cl₂ (1) in 93% yield (Figure 2). TaCl₅(THF) is more soluble than TaCl₅ and easier to obtain and store in pure form[52], and as a result was a much better precursor for salt metathesis. The solid-state structure of 1 reveals an 8-coordinate Ta atom containing three NP ligands that are bound κ² to the metal center. At room temperature, the ³¹P NMR consists of two resonances in a 2:1 ratio, consistent with the solid-state structure where two phosphines are equivalent and related by a mirror plane. No P-P coupling is observed at room temperature, similar to previous group 4 NP metalloligands that have labile M-P bonds[46]. However, The Ta-P distances in 1 (Ta-P = 2.667(1), 2.768(1), and 2.749(1) Å) are significantly shorter than that of the similar group 4 complex, Hf(NP)₄ (Average Hf-P = 2.93 Å). While both 1 and Hf(NP)₄ are 8-coordinate in the solid state, reduced steric pressure around the metal center in 1 likely enables shorter Ta-P bonds.

Figure 2:

Synthesis of Ta(κ²-NP)₃Cl₂ (1) and 50% thermal ellipsoid drawing of 1. Hydrogen atoms and solvent omitted for clarity.

Treatment of 1 with Ni(COD)₂ (COD = 1,5-cyclooctadiene) or Pd(P^tBu₃)₂ affords the heterobimetallic complexes Cl₂Ta(μ₂-NP)₃M (M = Ni (2), Pd (3)) in good yield (Figure 3, left). Similar metalation attempts with Pt(P^tBu₃)₂ resulted in decomposition of the Pt starting material. The solid-state structures of 2 and 3 are presented in Figure 4 and relevant bond distances and angles can be found in Table 3. 2 and 3 are isostructural, containing three phosphinopyrrolide ligands that bridge between the two metals. The geometry about the group 5 metal is pseudooctahedral, while the geometry about the late TM is a distorted trigonal pyramidal wherein the group 5 metal occupies the axial site of the pyramid. Complexes 2 and 3 are d¹⁰ and diamagnetic, and at room temperature display two ³¹P NMR signals in a 2:1 ratio, consistent with the crystal structures. Complex 2 features two broad ³¹P NMR signals, consistent with some degree of fluxionality, while 3 is more resolved, exhibiting a doublet and triplet corresponding to two equivalent trans phosphines and a cis phosphine, respectively. Single crystal diffraction reveals close contacts between metal centers: (2) Ta-Ni: 2.2979(5) Å (FSR = 0.92) and (3) Ta-Pd 2.4189(6) Å (FSR = 0.92)[34]. Complexes 2 and 3 are d⁰/d¹⁰ and diamagnetic and contain significant metal-metal interactions arising from dative, late to early bonding through a metal-metal σ bond, akin to what we have observed in group 4 – group 10 complexes[47, 49]. Group 4 (NP)M(NP)₃Ni (M = Ti, Zr, and Hf) complexes had weak (Ti) to nonexistent (Zr, Hf) π bonds, indicating that π bonding in such disparate metal combinations was not a significant contributor to overall bond order[47]. Given the similar metal-metal distances and FSRs seen in these group 5 based bimetallics it is unlikely that 2 or 3 contain significant π interactions.

Figure 3:

Synthesis of Ta – group 9/10 bimetallic complexes 2-5.

Figure 4:

50% thermal ellipsoid drawings of 2 (top left), 3 (top right), 4 (bottom left) and 5 (bottom right). Phenyl groups have been truncated to the ipso carbon, and hydrogen and solvent atoms have been removed for clarity. Bond distances and angles can be found in Table 3.

Table 3.

Relevant bond lengths (Å) and angles (°) for 2–5.

	2 Ta-Ni	3 Ta-Pd	4 Ta-Rh	5 Ta-Ir
Ta-M (Å)	2.2979(5)	2.4189(6)	2.4238(5)	2.4366(2)
FSRa	0.922	0.923	0.936	0.936
M-P (Å)	2.249(1)	2.389(2)	2.346(2)	2.362(2)
	2.239(1)	2.416(2)	2.302(1)	2.402(1)
	2.221(1)	2.385(3)	2.406(2)	2.305(1)
P-M-P (°)	123.13	112.98	91.84	91.95
	106.81	115.49	92.81	93.56
	129.17	131.53
Ta-N (Å)	2.111(3)	2.110(6)	2.141(4)	2.136(4)
	2.140(4)	2.10(1)	2.128(3)	2.128(3)
	2.135(4)	2.132(6)	2.108(5)	2.142(3)
Ta-Cl (Å)	2.375(1)b	2.358(3)b	2.356(1)b	2.370(1)b
	2.396(1)c	2.369(2)c	2.408(2)c	2.408(1)c
M-Cl (Å)	-	-	2.320(1)	2.338(1)
Torsional Angle	22.98	30.82	25.64	19.69
(P-M-Ta-N) (°)	11.60	4.43	25.62	22.49
	29.08	21.14	26.61	23.74

^aFSR = M_D/Σ(M_c): M_D = Metal-metal distance, M_c= Covalent Radii.

^bEquatorial Cl.

^cAxial Cl

Next, analogous metalations with group 9 starting materials were performed. Treatment of 1 with 0.5 equiv of [M’Cl(COD)]₂ (M’ = Rh, Ir) afforded the heterobimetallic complexes Cl₂Ta(μ₂-NP)₃M’Cl (M’ = Rh (4), Ir (5)) in good yield (Figure 3, right). The solid-state structures of 4 and 5 are presented in Figure 4 and relevant bond distances and angles can be found in Table 3. 4 and 5 are isostructural, containing three phosphinopyrrolide ligands that bridge between the two metals, resulting in a P₃MCl square-pyramidal d⁸ late transition metal center and a pseudooctahedral d⁰ Ta center. Similar to the group 10 complexes 2 and 3, these species are diamagnetic, and display two ³¹P NMR signals: a doublet and a triplet in a 2:1 ratio (J_PP = 20.2 Hz for 4, J_PP = 12.7 Hz for 5). In the case of 4, further Rh coupling is also observed (J_RhP = 140.9 Hz, 105.0 Hz). Both complexes contain short metal-metal bond distances: (4) Ta-Rh: 2.4238(5) Å (FSR = 0.93) and (5) Ta-Ir: 2.4366(2) Å (FSR = 0.94) consistent with the formation of a metal-metal bond. Compared to 2 and 3, 4 and 5 should contain stronger bonds if π bonding is present due to better orbital overlap. However, both the distances and FSRs are nearly identical, which leads us to conclude that the bonding in 4 and 5 is, again, dominated by a dative σ bond.

Next, the coordination chemistry of 2-5 with the π-acceptor isocyanide 2,6-xylylNC was examined. Previously, we had found that coordination of the π-acceptor CO to the group 4 heterobimetallic Ti(NP)₄Ni resulted in complete loss of Ti-Ni bonding, demonstrating that Ni->CO backbonding outcompeted Ni->Ti dative bonding[44]. Addition of 1 equiv of 2,6-xylylNC to complexes 2, 3, and 5 resulted in coordination of the isocyanide to the late TM center and quantitative formation of complexes 6-8 (Figure 5). Interestingly, all 3 isocyanide adducts exhibit different coordination geometries in the solid state (Figure 6).

Figure 5:

Synthesis of isocyanide adducts 6-8.

Figure 6.

50% thermal ellipsoid drawings of 6 (left), 7 (center), and 8 (right). Hydrogen atoms and solvent have been removed and phenyl groups have been truncated to the ipso carbon for clarity. Selected bond lengths and angles can be found in Table 4.

In the case of 6, the isocyanide coordinates to Ni and induces a ligand rearrangement where one NP ligand has migrated to the Ta center giving (κ²-NP)ClTa(μ₂-NP)₂(μ₂-Cl)Ni(CNXylyl) (6). The Ni center is distorted trigonal bipyramidal, and bonded to two phosphines, the bridging chloride, and capped by the isocyanide and Ta in the axial positions. The Ta center is formally 7 coordinate and adopts a pentagonal bipyramid where the two bridging pyrrolides occupy the axial positions. The metal-metal distance has elongated by ~0.23 Å from 2.2979(5) to 2.5354(3) Å (FSR = 1.02). The Ta-Cl bonds have elongated slightly from 2.396(1) and 2.375(1) to 2.4371(5) and 2.4439(6) Å, while the Ta-N distances do not vary significantly. The Ni-P distances have contracted slightly by ~0.02 Å. As mentioned, coordination of CO to Ti(NP)₄Ni resulted in loss of metal-metal bonding as shown by ~0.4 Å bond length increase. CNXylyl coordination results in a similar, but smaller, increase in metal-metal bond length in 6, in addition to ligand rearrangements. We attribute this rearrangement to the presence of the chloride ligand that can act as a single atom bridge between the two metal centers, which also attenuates the disruption of the metal-metal bond length by forcing close contact between the two metal centers.

In solution, 6 exists as 3 isomers in a 4 : 2 :1 ratio as measured by ³¹P NMR spectroscopy (Figure 7). Each isomer contains a single resonance shifted upfield, consistent with a Ta bound phosphine that is weakly coupling to the other phosphines on Ni giving a doublet and triplet (J_PP =9.6 Hz), a doublet of doublets and two doublets (J_PP = 74.4 and 25.0 Hz) and a doublet of doublets and two doublets (J_PP = 95.9 and 4.2 Hz), respectively (Figure 8). We have observed similar, weak P-P coupling through M-M bonds in Group 4–10 heterobimetallics complexes in low temperature ³¹P NMR experiments[47]. The nature of these closely related species is ambiguous but could arise from twisting/canting of the NP ligands or the relative position of the Ta-bound NP ligand and bridging chloride.

Figure 7:

³¹P{¹H} NMR of a crystalline sample of 6 showing three different isomers in solution.

In contrast to 6, the solid-state structures of 7 and 8 maintain the lantern-like core of the parent complexes 3 and 5 with short intermetallic distances (Ta-Pd = 2.5509(6) Å (FSR = 0.973) in 7; Ta-Ir = 2.5316(3) Å (FSR = 0.973) in 8) but coordinate the isocyanide in different positions. The slightly elongated Ta-M bonds in 7 and 8 result from reduced electron density at the late transition metal center resulting from M->CNR π backbonding. Unlike the similar (NP)Ti(NP)₃Ni(CO) complex[44], 7 and 8 maintain M-M bonding; this could be a function of relative halophilicities of the late transition metals, or better orbital overlap between Ta/Pd and Ta/Ir as compared to Ti/Ni.

Complex 7 adopts a distorted trigonal bipyramidal Pd geometry with the RNC ligand and Ta occupying the axial positions, in direct analogy to (NP)Ti(NP)₃Ni(CO). In contrast, 8 adopts a pseudoctahedral Ir geometry where the RNC ligand is in plane with the P₃ framework. This geometry is counter to what may be electronically favorable, where the strong trans RNC ligand would occupy the position trans to the (presumably) weak Ta ligand, but is likely sterically preferred as the bulky 2,6-xylylNC only contains two neighboring phosphine ligands when coordinated in the equatorial position, as opposed to three neighboring phosphine ligands, which would be the case with apical coordination. Interestingly, the location of the RNC ligand either trans (as in 7) or cis (as in 8) has little impact on the degree of Ta-M bonding as measured by FSR, further indicating that the amount of π interaction between the two metals is limited.

Conclusions

Synthesis of a tantalum metalloligand, Ta(NP)₃Cl₂, allowed for formation of a suite of bimetallic complexes with group 9 and 10 metals. Reaction with zero-valent group 10 starting materials resulted in complexes of the form Cl₂Ta(NP)₃M (M = Ni (2), and Pd (3)), while reaction with monovalent group 9 starting materials resulted in Cl₂Ta(NP)₃MCl (M = Rh (4), and Ir (5)) complexes. All complexes reported are isostructural, diamagnetic, and contain short metal-metal distances (FSR < 1.00), resulting from dative, late-to-early metal-metal bonds. Addition of isocyanides to TaNi (2), TaPd (3), and TaIr (5) revealed differing behaviors where the Ta-Ni framework underwent ligand rearrangement, while Ta-Pd and Ti-Ir frameworks did not. The Ta-M bond distances in this series of complexes are significantly shorter than those previously reported, demonstrating the utility of the 2-(diphenylphosphino)pyrrolide framework in supporting strong metal-metal interactions.

Table 4.

Relevant bond lengths (Å) and angles (°) for 6–8.

	6	7	8
	Ta-Ni(CNXylyl)	Ta-Pd(CNXylyl)b	Ta-Ir(CNXylyl)
M-M(Å)	2.5354(3)	2.5509(6)	2.5316(3)
FSRa	1.017	0.973	0.973
P-M (Å)	2.2068(6)	2.390(2)	2.383(2)
	2.1917(7)	2.382(2)	2.372(1)
		2.379(2)	2.405(1)
P-M-P (°)	143.11	110.69	91.43
		112.02	95.25
		135.04
M-N (Å)	2.101(2)	2.126(5)	2.114(4)
	2.096(2)	2.132(5)	2.133(6)
	2.149(2)	2.093(5)	2.134(4)
Ta-Cl (Å)	2.4371(5)	2.363(1)c	2.381 (2)c
	2.4439(6)d	2.434(2)e	2.452(1)e
M-Cl (Å)	2.3518(5)	-	2.609(1)
Torsional Angle	20.32	31.13	24.05
(P-M-M-N) (°)	20.59	7.10	17.65
		19.80	21.96
M-C (Å)	1.874(3)	2.164(6)	2.006(7)
C-N (Å)	1.159(3)	1.176(8)	1.150(9)

^aFSR = M_D/Σ(M_c): M_D = Metal-metal distance, M_c= Covalent Radii.

^bvalues given for one of two molecules in the asymmetric unit.

^Cequatorial Cl.

^dbridging Cl.

^eaxial Cl.