Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2021 Aug 4;27(52):13221–13234. doi: 10.1002/chem.202101484

[Ni(NHC)2] as a Scaffold for Structurally Characterized trans [H−Ni−PR2] and trans [R2P−Ni−PR2] Complexes

Sara Sabater 1,+, David Schmidt 2,+, Heidi (née Schneider) Schmidt 2, Maximilian W Kuntze‐Fechner 2, Thomas Zell 2, Connie J Isaac 1, Nasir A Rajabi 3, Harry Grieve 1, William J M Blackaby 1, John P Lowe 1, Stuart A Macgregor 3,✉, Mary F Mahon 1, Udo Radius 2,✉, Michael K Whittlesey 1,✉
PMCID: PMC8518396  PMID: 34190374

Abstract

The addition of PPh2H, PPhMeH, PPhH2, P(para‐Tol)H2, PMesH2 and PH3 to the two‐coordinate Ni0 N‐heterocyclic carbene species [Ni(NHC)2] (NHC=IiPr2, IMe4, IEt2Me2) affords a series of mononuclear, terminal phosphido nickel complexes. Structural characterisation of nine of these compounds shows that they have unusual trans [H−Ni−PR2] or novel trans [R2P−Ni−PR2] geometries. The bis‐phosphido complexes are more accessible when smaller NHCs (IMe4>IEt2Me2>IiPr2) and phosphines are employed. P−P activation of the diphosphines R2P−PR2 (R2=Ph2, PhMe) provides an alternative route to some of the [Ni(NHC)2(PR2)2] complexes. DFT calculations capture these trends with P−H bond activation proceeding from unconventional phosphine adducts in which the H substituent bridges the Ni−P bond. P−P bond activation from [Ni(NHC)2(Ph2P−PPh2)] adducts proceeds with computed barriers below 10 kcal mol−1. The ability of the [Ni(NHC)2] moiety to afford isolable terminal phosphido products reflects the stability of the Ni−NHC bond that prevents ligand dissociation and onward reaction.

Keywords: carbene ligands, density functional calculations, hydride ligands, nickel, phosphido ligands

Whole lotta terminal phosphidos: A combined experimental and computational study of the reactivity of [Ni(NHC)2] (NHC=N‐heterocyclic carbene) towards secondary and primary phosphines has been undertaken. P−H bond activation, as well as P−P activation of diphosphines, yields an extensive series of structurally characterised terminal phosphido complexes that exhibit unusual trans‐phosphido hydrido and novel trans‐bis‐phosphido geometries.

graphic file with name CHEM-27-13221-g006.jpg

Introduction

Isolation of the 14‐electron nickel(0) bis‐N‐heterocyclic carbene (NHC) complex, [Ni(IMes)2] (IMes=1,3‐bis(2,4,6‐trimethylphenyl)imidazolin‐2‐ylidene), by Arduengo and co‐workers in 1994 provided an early illustration of the ability of NHCs to stabilize coordinatively unsaturated transition metal complexes. [1] In the case of [Ni(IMes)2] and analogues containing even more bulky N‐aryl substituents,[ 2 , 3 ] however, a price is paid for this stability in the form of somewhat altered and limited reactivity.[ 4 , 5 ] [Ni(NHC)2] systems with higher reactivity towards small molecules [6] are accessible through the use of N‐alkyl substituted carbenes, although with the exception of [Ni(ItBu2)2] (ItBu2=1,3‐di‐tert‐butylimidazolin‐2‐ylidene)[ 2b , 7 ] these can only be generated in situ. Particular success has been found with [Ni2(IiPr2)4(COD)] (IiPr2=1,3‐di‐isopropylimidazolin‐2‐ylidene; [8] and related molecules) [9] as a labile source of the N‐isopropyl substituted carbene complex [Ni(IiPr2)2] (A, Scheme 1), which is able to activate a wide range of C−X (X=F, Cl, Br),[ 10 , 11 ] S−X (X=C, S) [12] and E−H (E=Si, S) bonds.[ 12 , 13 ] In a number of cases, these processes have been made catalytic, allowing the Suzuki‐Miyaura cross‐coupling of aryl fluorides, chlorides and bromides,[ 11 , 14 ] hydrodefluorination of aryl fluorides [15] and silane dehydropolymerisation. [16] An even more impressive transformation builds upon the ability of A to cleave strained C−C bonds, [17] allowing the coupling of 2,2′‐biphenylene with PhC≡CPh to yield 9,10‐diphenylphenanthrene. [8]

Scheme 1.

Scheme 1

Open in a new tab

[Ni2(IiPr2)4(COD)] as a labile source of [Ni(IiPr2)2] (A).

Very recently, one of us reported isolation of the Ni(II) terminal phosphido phenyl complex, trans‐[Ni(IMe4)2(PPh2)Ph], (1, Scheme 2) from the reaction of the N‐methyl‐substituted NHC IMe4 (IMe4=1,3,4,5‐tetramethylimidazolin‐2‐ylidene) with the 16‐electron Ni(0) complex [Ni(6‐Mes)(PPh3)2] (6‐Mes=1,3‐bis(2,4,6‐trimethylphenyl)‐3,4,5,6‐tetrahydropyrimidin‐2‐ylidene). The formation of 1, most likely through P−C activation of PPh3 by [Ni(IMe4)2] (B), [18] is surprising given that P−C cleavage reactions of aryl phosphines typically yield multi‐metallic products with bridging phosphido ligands, rather than mononuclear complexes with terminal phosphido groups. [19] Moreover, the isolation and structural characterization of 1 contrasts with the instability of the bis‐triethylphosphine analogue, trans‐[Ni(PEt3)2(PPh2)Ph], which has been reported as being too labile even to allow spectroscopic characterization. [20]

Scheme 2.

Scheme 2

Open in a new tab

Syntheses of the [Ni(NHC)2(PPh2)Ph] complexes 1 (NHC = IMe4) and 2 (NHC = IEt2Me2).

Our mutual interests in the bond activation reactions of A and B have now led us to conduct a joint investigation into their reactivity, together with that of the N−Et‐substituted derivative [Ni(IEt2Me2)2] (C; IEt2Me2=1,3‐diethyl‐4,5‐dimethylimidazolin‐2‐ylidene), towards P−H bonds of secondary and primary phosphines. At the outset of this work, we were aware of only a few examples of mononuclear nickel complexes with terminal phosphido ligands,[ 21 , 22 ] and a single case of an isolated Ni phosphido hydrido complex, [Ni(dtbpe){P(Dmp)H}H)] (dtbpe=tBu2PCH2CH2PtBu2; Dmp=2,6‐Mes2C6H3). [23] Herein, we describe the use of A–C to generate terminal phosphido hydrido and terminal bis‐phosphido complexes, nine of which have been structurally characterized. [24] Among these are the first examples of terminal bis‐phosphido complexes with trans R2P−M−PR2 geometries reported for any transition metal. DFT calculations have also been used to rationalize the formation of the terminal phosphido hydrido and terminal bis‐phosphido complexes as a function of the NHCs involved.

Results and Discussion

In‐situ generation of [Ni(IMe4)2] (B) and [Ni(IEt2Me2)2] (C)

Given the atom inefficiency associated with the formation of 1 from [Ni(6‐Mes)(PPh3)2] and IMe4 (i. e., loss of both 6‐Mes and PPh3 during reaction), a more economical pathway was afforded by simply mixing [Ni(COD)2] with IMe4 and PPh3 in a 1 : 2 : 1 molar ratio at room temperature. This gave 1 (Scheme 2a) in about 90 % yield based on NMR spectroscopy. The N‐ethyl carbene derivative, trans‐[Ni(IEt2Me2)2(PPh2)Ph] (2), could be prepared similarly (Scheme 2b), again presumably through activation of a P−Ph group by [Ni(IEt2Me2)2] formed in situ. The combination of the N‐iPr substituted NHC IiPr2Me2 (IiPr2Me2=1,3‐di‐isopropyl‐4,5‐dimethylimidazol‐2‐ylidene) with [Ni(COD)2] and PPh3 gave a mixture of uncharacterized products. The formation of 1 and 2 was quite slow, taking days to reach completion. In the case of 2, raising the temperature to 50 °C reduced the reaction time to just 3 h. Complex 2, which was isolated in 45 % yield, was readily characterizable by comparison of spectroscopic data to those of 1, with a high frequency 13C{1H} NMR resonance at δ=166 ppm (d, 2 J(C,P)=40 Hz) diagnostic of the ipso‐C of the diphenylphosphido ligand.

The molecular structure of 2 (Figure 1) confirmed the presence of a pyramidal phosphido ligand (Σ(angles at P)=325.4°; cf. 328.5° for a tetrahedral P) and a trans Ph−Ni−PPh2 geometry. In comparison to 1, there was a slight elongation of both the Ni−Cphenyl and Ni−PPh2 distances (2: 1.960(3) Å, 2.2851(8) Å; 1: 1.9371(15), 2.2520(5) Å). We attribute this to increased crowding at the nickel centre resulting from the less coplanar arrangement of the two NHC ligands in 2 (N−C−C−N dihedral angles: 2: 142.3°, 161.6°; 1: 159.2°, 156.7°). As will become obvious over the course of the manuscript, sterics plays a key role not only in defining the products formed between [Ni(NHC)2] and specific phosphines, but also their geometries.

Figure 1.

Figure 1

Open in a new tab

Molecular structure of trans‐[Ni(IEt2Me2)(PPh2)Ph] (2). Ellipsoids are shown at 30 % probability with all hydrogen atoms removed for clarity.

[Ni(NHC)2(PR2)H] formation with secondary phosphines

Upon addition of the secondary aryl phosphine PPh2H to [Ni(IiPr2)2] (A) or [Ni(IEt2Me2)2] (C), rapid, room temperature P−H activation ensued to afford the phosphido hydrido complexes trans‐[Ni(IiPr2)2(PPh2)H] (3) and trans‐[Ni(IEt2Me2)2(PPh2)H] (4), respectively (Scheme 3). The trans‐R2P−M−H geometry is rare for any transition metal, [25] and contrasts with the cis‐R2P−Ni−H arrangement in [Ni(dtbpe){P(Dmp)H}H)] resulting from the presence of the chelating phosphine ligand. [23]

Scheme 3.

Scheme 3

Open in a new tab

Formation of the phosphido hydrido complexes 3 and 4.

The trans Ph2P−Ni−H arrangement in both 3 and 4 was established by X‐ray crystallography (Figure 2) and multinuclear NMR spectroscopy. The change of ligand trans to phosphido group from Ph (2) to H (4) resulted in a slight shortening of the Ni−PPh2 distance, despite the higher trans influence of the latter (2: 2.2851(8) Å; 4: 2.2783(4) Å). This may reflect the narrowing of the CNHC−Ni−CNHC bond angle (2: 175.26(12)°; 4: 169.16(5)°) that permits a closer approach of the phosphido ligand on steric grounds. Ni−H resonances for 3 and 4 were found at δ −11.3 ppm and δ −11.7 ppm, respectively, both with a doublet 2 J(H,P) splitting of 73 Hz. Identification as a trans‐coupling by comparing it with the cis‐splitting in [Ni(dtbpe){P(Dmp)H}H] [23a] was prevented by the absence of any resolvable couplings on the Ni−H signal of the latter (even at low temperature). Although the cis phosphido‐Ni−H splitting was likewise unresolved in the related complex [Ni(dtbpe){PMes*H}H] (Mes*=2,4,6‐tBu3C6H2), [23b] this did show a coupling of comparable large magnitude (104 Hz) for J(trans phosphine‐Ni−H). [26] The best comparison involves the iridium bis‐PH2 hydrido complex, [Ir(PEt3)2(PH2)2(CO)H], [25c] which shows large trans (42 Hz) and much smaller cis (7 Hz) values of 2 J(PH2‐Ir−H). Final confirmation of geometry in 3 and 4 was provided by the appearance of a small (11 Hz) cis NHC−Ni−PPh2 2 J(C,P) doublet splitting on each of the high frequency 13C carbene resonances of 3 and 4.

Figure 2.

Figure 2

Open in a new tab

Molecular structures of left: trans‐[Ni(IiPr2)2(PPh2)H] (3) and right: trans‐[Ni(IEt2Me2)2(PPh2)H] (4). Ellipsoids are shown at 30 % probability with all hydrogen atoms, except Ni−H, removed for clarity.

Formation of trans‐[Ni(NHC)2(PAr2)2] complexes

In contrast to the clean formation of phosphido hydrido complexes 3 and 4, [27] [Ni(IMe4)2] (B) reacted with PPh2H to give a mixture of both the phosphido hydrido complex trans‐[Ni(IMe4)2(PPh2)H] (5) and the bis‐phosphido species, trans‐[Ni(IMe4)2(PPh2)2] (6, Scheme 4). Conversion through to just 6 took place over about 24 h at room temperature. [28]

Scheme 4.

Scheme 4

Open in a new tab

Phosphido hydrido and bis‐phosphido complexes from B and PPh2H.

Despite this difference in initial products from A and B (Schemes 3 and 4), we found that addition of PPh2H to [Ni(IEt2Me2)2(PPh2)H] (4) did bring about conversion through to the bis‐phosphido complex, trans‐[Ni(IEt2Me2)2(PPh2)2] (7, Scheme 5), [27] although the process was slow, requiring over ca. 1 month at room temperature to reach completion. [29] Steric factors likely contribute significantly to the difficulty of this transformation; indeed, exposure of the bulkier IiPr2 precursor A to even a large excess of PPh2H failed to generate any of the bis‐PPh2 complex, trans‐[Ni(IiPr2)2(PPh2)2] (8), although the less bulky methylphenylphosphido derivative, trans‐[Ni(IiPr2)2(PPhMe)2] (9), could be made this way (Scheme 6).

Scheme 5.

Scheme 5

Open in a new tab

Routes to [Ni(IEt2Me2)2(PPh2)2] 7.

Scheme 6.

Scheme 6

Open in a new tab

Formation of bis‐phosphido complexes from A.

Unexpectedly, 8 could be formed through P−P bond activation of Ph2P−PPh2 in the presence of A (Scheme 6), which also reacted with PhMeP−PMePh to give an alternative route to 9. In fact, P−P activation proved to be a highly efficient route to bis‐phosphido complexes; formation of 7 was complete within just 30 min when Ph2P−PPh2 was added to C and within a few hours upon addition to 4. Although P−P activation of P4 and other cyclic/oligomeric P x species is well‐known, [30] activation (or even simple coordination) of diphosphines at a transition metal (or p‐block) fragment is extremely rare.[ 31 , 32 , 33 ] Indeed, Ph2P−PPh2 is most commonly encountered as an unreactive elimination product in catalytic dehydrocoupling reactions of secondary phosphines. [34]

Figure 3 shows the X‐ray crystal structures of 6 and 7, which we believe represent the first structurally characterized examples of mononuclear metal complexes with trans terminal phosphido ligands.[ 35 , 36 ] The Ni atoms are coincident with crystallographic inversion centres in both structures, resulting in a trans disposition of the two sets of phenyl substituents about the P−Ni−P axis in 6 and the two sets of N−Et groups about the C−Ni−C axis in 7. The Ni−P distance in 7 (2.2668(3) Å; 2.2628(5) Å in 6) was shorter than in both [Ni(IEt2Me2)2(PPh2)Ph] (2) and [Ni(IEt2Me2)2(PPh2)H] (4), and there was a minor increase in the sum of the angles at the pyramidal phosphorus atoms (2, 325.4°; 4, 324.3°; 7, 329.6°).

The isolation of the NHC‐containing bis‐phosphido complexes 6 and 7 is in marked contrast to the stabilities of the bis(triethylphosphine) derivatives, trans‐[M(PEt3)2(P(SiMe3)2)2]. [37] The M=Pt derivative survives long enough to allow a trans‐R2P−Pt−PR2 arrangement to be confirmed spectroscopically, whereas the M=Ni analogue proved to be unstable above −70 °C. The difference in stability imparted by NHCs underlines the comparison made earlier between 1 and [Ni(PEt3)2(PPh2)Ph]. [20]

Unequivocal spectroscopic evidence for the presence of trans R2P−Ni−PR2 arrangements in the four bis‐phosphido complexes 6–9 was provided by i) the small (2 J(C,P)=15–17 Hz) triplet splitting of the 13C carbenic carbon resonance and ii) the appearance in all the spectra of a virtually coupled signal at about δ=150 ppm for the magnetically inequivalent ipso‐C of the two trans‐phosphido ligands.

Reactivity of [Ni(NHC)2] with primary aryl phosphines

P−H bond activation of the primary aryl phosphine PPhH2 by B and C (Scheme 7) was also facile, although both reactions gave a mixture of the primary phosphido hydrido complexes trans‐[Ni(NHC)2(PPhH)H] (NHC=IMe4 10, IEt2Me2 12) and bis‐phosphido products, trans‐[Ni(NHC)2(PPhH)2] (NHC=IMe4 11, IEt2Me2 13). As shown in Scheme 8, A displayed even more complex reactivity with 1–4 equiv. PPhH2 (as well as P(para‐Tol)H2) giving a mixture of three products, even when the reactions were conducted at −30 °C. [38] The formation of trans‐[Ni(IiPr2)2(PArH)H] (Ar=Ph 14, para‐Tol 15) was signified by the appearance of low frequency signals (see below) in both the 1H and 31P NMR spectra, although these complexes defied all efforts at isolation. Repeating the reactions in the presence of an alkene as a hydrogen acceptor allowed isolation and characterisation of the second component of the reactions, namely the bis‐phosphido complexes, trans‐[Ni(IiPr2)2(PArH)2] (Ar=Ph 16, para‐Tol 17). The diphosphene complexes, [Ni(IiPr2)2(η2‐ArP=PAr)] (Ar=Ph 18, para‐Tol 19) were identified as being the third component upon subsequent preparation using more forcing conditions (see below).

Scheme 7.

Scheme 7

Open in a new tab

Phosphido hydrido and bis‐phosphido complexes from PPhH2.

Scheme 8.

Scheme 8

Open in a new tab

Reactivity of A towards primary aryl phosphines.

Increasing the phosphine bulk proved one way to direct the reactivity of A. Thus treatment with mesitylphosphine (PMesH2) gave around 95 % formation (based on NMR spectroscopy) of only the phosphido hydrido complex, trans‐[Ni(IiPr2)2(PMesH)H] (20, Scheme 9). As for 10 and 12, this was readily characterised from the 1H NMR spectrum through the presence of a low frequency Ni−H resonance (ca. δ=−11 ppm) with a 2 J(H,P) doublet splitting (61 Hz) again indicative of the hydride being trans to the phosphido ligand. An additional smaller coupling of 2 Hz to the PMesH proton was observed. The P−H resonance itself appeared at δ=2.9 ppm with a doublet 1 J(H,P) coupling of 200 Hz; as noted by others, the magnitude of this splitting is similar to that in the free primary phosphine. [39] In the proton coupled 31P NMR spectrum, the phosphido ligand appeared as a very low frequency doublet of doublets (δ=−91 ppm), while the 13C{1H} NMR spectrum displayed the high frequency (δ=191 ppm) carbenic carbon resonance as the anticipated doublet with a cis‐[NHC−Ni−PR2] 2 J(C,P) coupling of 9 Hz.

Scheme 9.

Scheme 9

Open in a new tab

Reactivity of A with primary phosphines.

Given the similarity of the NMR spectra of 20 to those of 10, 12, 14 and 15, it is reasonable to assume that these four compounds also displayed trans‐[H−Ni−PArH] geometries; they therefore provide a direct comparison of spectroscopic data for isostructural primary and secondary phosphido hydrido complexes. Thus, the Ni−H resonances of the former resonated at slightly higher frequency and showed smaller 2 J(H,P) coupling constants (10: δ=−10.04 ppm, 2 J(H,P)=58.0 Hz c.f. 5: δ=−10.97 ppm, 2 J(H,P)=68.0 Hz; 14: δ=−10.41 ppm, 2 J(H,P)=62.1 Hz c.f. 3: δ=−11.31 ppm, 2 J(H,P)=70.0 Hz), while the 31P NMR signals appeared at significantly lower frequencies (10: δ=−43.8 ppm; 12: δ=−38.1 ppm; 14: δ=−41.2 ppm cf. 5: δ=19.3 ppm; 4: δ=18.1 ppm; 3: δ=25.1 ppm; cf. Table 1). [40]

Table 1.

Comparison of bond distances [Å] and angles [°] and selected NMR spectroscopic data (δ (ppm), J [Hz]) for the bis‐phosphido complexes 6, 7, 11, 16, 17, 21 and 22.

 

6

7

11

16

17

21

22

r(Ni−P)

2.2628(5)

2.2668(3)

2.2349(3)

2.2399(13)

2.2396(13)

2.2671(8)

2.2241(9)

 

 

 

 

2.2414(13)

 

2.2550(8)

 

r(Ni−CNHC)

1.8893(18)

1.8905(12)

1.8917(12)

1.896(4)

1.8852(19)

1.885(3)

1.890(3)

 

 

 

 

1.888(4)

 

1.899(3)

 

∠(P−Ni−P)

179.999(16)

180.00

180.00

175.42(6)

180.000(17)

172.12(3)

180.00

Σ(angles at P)

327.7

329.6

307.1

311.7/313.2

314.2

320.2/321.3

306.8

δ 1HPH

–

–

3.33

3.11

3.11

2.94

0.90

31P

31.2

47.1

−40.4

−29.5

−31.0

−71.2

−156.3

1 J(H,P)

–

–

–

198

198

200

171

δ 13CNHC

182.7

190.3

188.1

186.3

190.0

185.9

190.0

2 J(CNHC,P)

16

13

13

14

13

13

11
Open in a new tab

Characterisation of trans‐[Ni(NHC)2(PArH)2] species

The bis‐phosphido complexes 11 and 13 (Scheme 7) formed as the ultimate products of the reactions of B and C with PPhH2 over several days at room temperature. Similarly, despite the bulk of PMesH2, 20 also slowly converted to trans‐[Ni(IiPr2)2(PMesH)2] (21, Scheme 9) upon addition of further PMesH2, albeit in only low yield. At the opposite end of the steric spectrum to 21 was trans‐[Ni(IiPr2)2(PH2)2] (22), which formed as the major product of the reaction of A with excess PH3 (Scheme 9). [41]

Complexes 11, 16, 17, 21 and 22 were structurally characterized as shown in Figure 4. They represent not only rare cases of structurally characterized metal primary phosphido complexes[ 21c , 40 , 42 ] but are the first examples of trans‐RHP−M‐PRH complexes. Table 1 summarises selected solid‐state and spectroscopic metrics for these compounds, together with the two bis‐PPh2 complexes 6 and 7 for comparison. In general, the Ni−PArH distances are slightly shorter than those in the Ni−PAr2 species (with the exception of the sterically crowded bis‐P(Mes)H complex 21). Although the trans PR2−Ni−PR2 geometry may limit comparisons to other systems, Glueck has reported that the Pd−PR2 distance in [Pt(dppe)(PMes2)Me]>[Pt(dppe)(P(Mes)H)Me] and that the trans influence of PPh2>PHPh for the [Pt(dppe)(PR2)Me] complexes. [39] The bis‐PH2 complex 22 is notable in displaying significant upfield shifts of both the 1H and 31P signals of the PH2 ligands, as well as a much reduced 1 J(H,P) coupling constant. [43]

Figure 4.

Figure 4

Open in a new tab

Molecular structures (from top left to bottom right) of trans‐[Ni(IMe4)2(PPhH)2] (11; i 1− x, 1−y, 1−z symmetry operation), trans‐[Ni(IiPr2)2(PPhH)2] (16), trans‐[Ni(IiPr2)2(P{para‐Tol)H}2] (17; i 1−x, 1−y, −z symmetry operation), trans‐[Ni(IiPr2)2(PMesH)2] (21) and trans‐[Ni(IiPr2)2(PH2)2] (22; i −x, 2−y, 1−z symmetry operation). Ellipsoids are represented at 30 % probability, and hydrogens (with the exception of those bonded to phosphorus) have been omitted for clarity. Also omitted is the minor disordered component in 16 and lattice solvent in 21. Only one of the two molecules in the asymmetric unit of 22 is shown.

Figure 3.

Figure 3

Open in a new tab

Molecular structures of trans‐[Ni(IMe4)2(PPh2)2] (6) and trans‐[Ni(IEt2Me2)(PPh2)2] (7). Ellipsoids are shown at 30 % probability and all hydrogen atoms are removed for clarity (6: i 1−x, −y, 1−z symmetry operation; 7: i 1−x, 1−y, 1−z symmetry operation).

Isolation and characterization of [Ni(IiPr2)2(η2‐ArP=PAr)]

The diphosphene complexes [Ni(IiPr2)2(η2‐ArP=PAr)] (Ar=Ph 18, para‐Tol 19) were isolated in about 50 % yield by heating A with excess PPhH2 or P(para‐Tol)H2 at 140 °C. While diphosphene complexes have been synthesised previously using a range of approaches (ligand exchange of stable diphosphenes with labile M−L complexes, salt metathesis and degradation of oligophosphines),[ 30e , 44 ] their preparation by the formal dehydrocoupling of a primary phosphine in the coordination sphere of a transition metal as observed here is unusual. [45]

X‐ray crystallography (Figure 5) showed, in both cases, a trans disposition of the two aryl substituents in the diphosphene ligand. The P−P bond lengths (18: 2.1379(7) Å; 19: 2.1342(7) Å) were in good agreement with those in [Ni(PEt3)2(trans‐MesP=PMes)] (2.1355(9) Å), [Ni(PMe2Ph)2(trans‐MesP=PMes)] (2.137(1) Å), and [Pd(dppe)(trans‐PhP=PPh)] (2.121(4) Å; dppe=1,2‐bis(diphenylphosphino)ethane), respectively.[ 44 , 46 ] The angles C1−Ni−C11 (18, 104.62(6)°; 19, 98.85(8)°) were larger than 90°, as expected for a pseudo three‐coordinated complex with small P−Ni−P angles (18, 56.76(2)°; 19, 56.89(2)°).

Figure 5.

Figure 5

Open in a new tab

Molecular structures of [Ni(IiPr2)2(η2‐PhP=PPh)] (18) and [Ni(IiPr2)2{η2‐(para‐Tol)P=P(para‐Tol)}] (19). Ellipsoids are represented at 30 % probability, and hydrogens (along with lattice solvent in 19) have been omitted for clarity. For 19: i 1−x, y, 1/2 −z symmetry operation.

The 1H NMR resonances of the IiPr2 ligands in both 18 and 19 were broadened at room temperature, consistent with hindered rotation of the ArP=PAr ligand around the Ni‐(PP)centroid vector and/or of the NHC ligands around the Ni−C bonds. Cooling to −50 °C resolved the spectra into four iPr methyl and two iPr methine resonances, as typically found for a C 2v type structure. At 90 °C, two iPr methyl resonances and a single iPr methine signal were observed, which were still broadened. Both complexes exhibited singlets in their 31P{1H} NMR spectra (18, δ=−41 ppm; 19, δ=−39 ppm) to lower frequency of the Pd and Pt derivatives [M(dppe)(trans‐PhP=PPh)] (M=Pd, δ=−34 ppm; M=Pt, δ=−24 ppm). [44]

Computational Studies

i) P−H Activation of PPh2H at [Ni(NHC)2]

The mechanisms of these processes were investigated using DFT calculations for the reactions of both [Ni(IiPr2)2], A, and [Ni(IMe4)2], B. Geometry optimisation and frequency calculations were performed with the BP86 functional using SDD pseudopotentials and basis sets on Ni and P (with d‐orbital polarization on the latter) and 6–31G** basis sets on all other atoms (collectively termed BS1). Electronic energies were then recomputed with B97D including corrections for benzene solvent and a def2‐TZVP basis set (BS2). These were then combined with the BP86 thermochemical corrections to supply the relative free energies reported in the text. The choice of B97D was motivated by the computed barriers for the reactions of A and B with PPh2H, for which this functional gave values qualitatively consistent with the reactivities observed at room temperature (i. e., B: rapid formation of trans‐[Ni(IMe4)2(PPh2)H], 5, followed by slower conversion to trans‐[Ni(IMe4)2(PPh2)2], 6; A: rapid formation of trans‐[Ni(IiPr2)2(PPh2)H], 3, but no onward reaction to form bis‐phosphido complex 8). The computed energetics were found to be very functional dependent, in a similar way that phosphine ligand addition energies are highly sensitive to functional choice and, in particular, the treatment of dispersion effects.[ 47 , 48 ] This appears to be exacerbated here due to the sequential addition of two PPh2H ligands to two [Ni(NHC)2] systems of varying steric bulk. The data below therefore present our best proposed mechanisms in the face of these challenges. [49] Full computational details, references and functional testing are provided in the Supporting Information.

The free energy profile computed with B97D for the reaction of B with PPh2H to form 5 is shown in Figure 6. Addition of PPh2H to B is exergonic by 11.6 kcal mol−1 and forms [Ni(IMe4)2(PPh2H)], I, the energy of which is set to 0.0 kcal mol−1. [50] P−H activation then proceeds via an initial rearrangement to intermediate II, in which the H substituent bridges the Ni−P bond (Ni−H=1.62 Å, P−H=1.59 Å, Ni−P=2.18 Å). The P−H bond is therefore significantly weakened and readily undergoes cleavage to form the cis isomer of 5, denoted cis‐III in the computational study. Formation of cis‐III is slightly endergonic, and hence reversible, allowing the trans isomer to be accessed via TS(II‐trans‐III) at +19.6 kcal mol−1 (see Figure 7 for the computed geometries of rate‐limiting transition states). The much higher barrier associated with TS(II‐trans‐III) arises from the coupling of P−H bond cleavage with a cis‐trans isomerisation process. The triplet form of [Ni(IMe4)2(PPh2)H], 3 III, has a computed energy of +23.7 kcal mol−1 and so is not implicated in this isomerisation. [51] IMe4 dissociation from cis‐III was also ruled out as this incurs a free energy penalty of 25.7 kcal mol−1. Overall, trans‐III is formed with an overall barrier of 19.6 kcal mol−1 in a process that is exergonic by 2.4 kcal mol−1 relative to I.

Figure 6.

Figure 6

Open in a new tab

Computed free‐energy profile [kcal mol−1, B97D(benzene, BS2)//BP86(BS1)] for the reaction of [Ni(IMe4)2] (B) with PPh2H to form trans‐[Ni(IMe4)2(PPh2)H] 5. The triplet form of [Ni(IMe4)2(PPh2)H], 3 III, is computed at +23.7 kcal mol−1.

Figure 7.

Figure 7

Open in a new tab

Computed geometries for the rate‐limiting transition states for the first and second P−H activation of PPh2H at B. Selected distances are shown in Å, with IMe4 and phenyl hydrogens omitted for clarity.

The lowest energy pathway for the onward reaction of 5 with a second molecule of PPh2H to form 6 is shown in Figure 8. [52] This involves an initial P−H reductive coupling to reform II which can then add a second PPh2H molecule in the unconventional H‐bridged binding mode to give IV (+10.7 kcal mol−1). Both P−H bonds in IV are significantly elongated (ca. 1.63 Å) and a P−H activation transition state, TS(IV–V), was located at +21.9 kcal mol−1 in which both P−H bonds are cleaved (Figure 8). This occurs with concomitant H2 elimination to give cis‐V directly. [53] cis‐V can then isomerise to trans‐V (i. e., 6) at −19.9 kcal mol−1 via TS(cis‐trans‐V) at +4.9 kcal mol−1 on the singlet surface, although in this case the triplet species 3 V lies 3.8 kcal mol−1 below TS(cis‐trans‐V) and so isomerisation may involve a spin crossover process. [54] These isomerisation pathways do not affect the overall energy span of 24.3 kcal mol−1 for the second P−H bond activation, as this is associated with TS(IV–V) as the highest point on the profile. The formation of 6 from 5 is strongly exergonic (−17.5 kcal mol−1).

Figure 8.

Figure 8

Open in a new tab

Computed free‐energy profile [kcal mol−1, B97D(benzene, BS2)//BP86(BS1)] for the reaction of 5 with PPh2H to form trans‐[Ni(IMe4)2(PPh2)2] 6.

Overall, the first P−H activation at B to form 5 is more kinetically accessible (ΔG ≠=19.6 kcal mol−1) than the second to form 6 (ΔG ≠=+24.3 kcal mol−1) and this is qualitatively consistent with the rapid formation of 5 at room temperature and its subsequent slower conversion to 6. Calculations for the reactions of A with PPh2H followed similar pathways, with a barrier of 20.2 kcal mol−1 for the first P−H activation to form 3 at −0.8 kcal mol−1. In this case the potential second P−H bond activation entails a higher overall barrier of 26.6 kcal mol−1, consistent with the formation of trans‐[Ni(IiPr2)2(PPh2)2], 8, being much less accessible at room temperature, at least by this route (see Figures S74 and S75 in the Supporting Information). [55]

As well as the observed P−H bond activations, the alternative P−Ph bond activation of PPh2H at [Ni(NHC)]2 is a potentially competing process (see Figure 9 for the IMe4 system). This proceeds with an initial computed barrier of 13.4 kcal mol−1 to give cis‐[Ni(IMe4)2(PPhH)Ph], cis‐IIIPh , at −1.6 kcal mol−1 from which cis–trans isomerisation entails a barrier of 25.6 kcal mol−1 to give trans‐IIIPh at −6.2 kcal mol−1. Therefore, in the absence of a second equivalent of PPh2H, P−H bond activation to form 6 is kinetically favoured over trans‐IIIPh (ΔG ≠=+19.6 kcal mol−1, ΔG=−2.4 kcal mol−1) but as this could be reversible, this could in principle allow access to trans‐IIIPh at −6.2 kcal mol−1 as the thermodynamically more stable product of P−C bond activation. This latter process, however, would have a significant barrier of 26.4 kcal mol−1 relative to 6. Moreover, in the presence of excess PPh2H, a second P−H activation to form 6 is favoured both kinetically (ΔG ≠=+24.3 kcal mol−1) and thermodynamically (ΔG=−19.9 kcal mol−1). The pattern of a kinetically favoured P−H bond activation versus a thermodynamically favoured P−C bond activation has been reported for the reaction of [Pt(dppe)(trans‐stilbene)] with PMes2H.[ 56 , 57 ] Attempts to realise this pattern experimentally through the reaction of [Ni(COD)2] with 2 equiv of IMe4 and only one equiv. of PPh2H were unsuccessful with a complex mixture of products being formed.

Figure 9.

Figure 9

Open in a new tab

Computed free‐energy profile [kcal mol−1, B97D(benzene, BS2)//BP86(BS1)] for competing P−C bond activation in [Ni(IMe4)2(PPh2H)], I. The triplet form of [Ni(IMe4)2(PPh2)Ph], 3 IVPh , is computed at +25.6 kcal mol−1.

ii) P−P Activation of Ph2P−PPh2 at [Ni(NHC)2]

The computed profile for the alternative route to trans‐[Ni(IMe4)2(PPh2)2], 6, through reaction of Ph2P−PPh2 with B is shown in Figure 10, with the geometries of key transition states in Figure 11. Ph2P−PPh2 addition to B is exergonic by 17.8 kcal mol−1 (cf. 11.6 kcal mol−1 for PPh2H) and gives trigonal planar [Ni(IMe4)2(Ph2PPPh2)] (VI, Ni−P=2.35 Å), the energy of which is set to 0.0 kcal mol−1. P−P bond activation then proceeds via a facile transfer of one PPh2 group onto Ni via TS(VI‐cis‐V) at +9.0 kcal mol−1. This forms cis‐V at −5.5 kcal mol−1 that, as discussed above, can undergo cis–trans isomerisation on the singlet surface via TS(cis‐trans‐VI) at +10.3 kcal mol−1 or potentially via a lower energy spin crossover mechanism via 3 V at +6.5 kcal mol−1. [58] Cis–trans isomerisation on the singlet surface represents the largest barrier along the profile for the formation of 6, but at only 15.8 kcal mol−1, the implication is that the overall reaction to form 6 should be readily accessible, consistent with the experiment observation of this process at room temperature.

Figure 10.

Figure 10

Open in a new tab

Computed free energy profiles [kcal mol−1, B97D(benzene, BS2)//BP86(BS1)] for the reaction of B with Ph2P‐PPh2 to form 6.

Figure 11.

Figure 11

Open in a new tab

Computed geometries for key transition states in the P−P bond activation of Ph2PPPh2 at B. Selected distances are shown in Å, with IMe4 and phenyl hydrogens omitted for clarity.

P−P bond activation in the analogous reaction of Ph2P−PPh2 with A is shown in Figure 12. This now involves a two‐step process from the initial Ph2P−PPh2 adduct, IV i Pr , via an intermediate Int(VI‐cis‐V) i Pr at −0.3 kcal mol−1 that features a bridging PPh2 ligand, followed by facile P−P bond cleavage to give cis‐V i Pr at −5.8 kcal mol−1. Cis–trans isomerisation on the singlet surface then proceeds via TS(cis‐trans‐V) i Pr at +11.8 kcal mol−1. The details of this process are complicated by the existence of numerous conformations along the pathway and only the highest‐lying transition state is shown in Figure 12 (see Figure S78 for more details). The triplet form, 3 V i Pr , is again accessible at +4.3 kcal mol−1. As with the analogous reaction with B, all barriers are readily accessible, the largest energy span being only 17.6 kcal mol−1 (corresponding to isomerisation on the singlet surface) and so the formation of 8 should be readily accessible, as is the case experimentally.

Figure 12.

Figure 12

Open in a new tab

Computed free‐energy reaction profile [kcal mol−1, B97D(benzene, BS2)//BP86(BS1)] for the reaction of A with Ph2P−PPh2 to form trans‐[Ni(IiPr)2(PPh2)2], 8. [a] Only the rate‐limiting transition state for cis–trans isomerisation is indicated (see Figure S78 for more details).

Conclusions

Facile P−H activation of secondary and primary phosphines by the 14‐electron fragments [Ni(IiPr2)2] (A), [Ni(IMe4)2] (B) and [Ni(IEt2Me2)2] (C) provides access to a unprecedented series of stable terminal phosphido complexes with trans H−Ni−PR2 and trans‐R2P−Ni−PR2 structures. Within the family of [Ni(NHC)2] fragments that we have considered, clear evidence for the stereoelectronic influence of the N substituents is apparent. This is best evidenced in the initial reactions of A, B and C with PPh2H. With bulkier A and C, the initial room temperature reaction yields isolable phosphido hydrido complexes, trans‐[Ni(IiPr2)2(PPh2)H] (3) and trans‐[Ni(IEt2Me2)2(PPh2)H] (4), respectively. With B, a mixture of the equivalent phosphido hydrido trans‐[Ni(IMe4)2(PPh2)H] (5) and bis‐phosphido trans‐[Ni(IMe4)2(PPh2)2] (6) complexes is first seen that then converts fully to 6 over 24 h. Thus B, that features the smallest of the carbenes, IMe4 can most easily accommodate two P−H bond activations to form a bis‐phosphido directly. Further reaction of 4 with PPh2H to form trans‐[Ni(IEt2Me2)2(PPh2)2] (7) does occur but is very slow, requiring weeks at room temperature. No further reaction between 3 and PPh2H is seen and so [Ni(IiPr2)2(PPh2)2] (8) is inaccessible by this route. We were able to prepare 8, as well as 7 and the mixed aryl/alkyl bis‐phosphido complex [Ni(IiPr2)2(PPhMe2)2] (9) by the unexpected P−P cleavage of the diphosphines R2P−PR2, which like the P−H activation reactions above, takes place under very mild conditions.

These trends were captured by DFT calculations on the reactions of A and B, although the computed energetics of these processes were found to be very functional dependent. With PPh2H low barriers for the first P−H activation (A: 20.2 kcal mol−1; B: 19.6 kcal mol−1, B97D functional) contrasted with increased barriers for the second P−H activation (A: 26.6 kcal mol−1; B: 24.3 kcal mol−1). Each P−H bond activation proceeds from an unconventional phosphine adduct in which the H substituent bridges the Ni−P bond. The rate limiting transition states then correspond to P−H bond cleavage; for the first P−H activation, this is coupled to cis–trans isomerisation to form trans‐[Ni(NHC)2(PPh2)H], whereas for the second P−H activation, a novel transition state involving dual P−H bond cleavage of two PPh2H ligands is located. P−P bond activation from adducts of the type [Ni(NHC)2(Ph2P−PPh2)] (NHC=IiPr2, IMe4,) is computed to be an extremely facile process, with barriers below 10 kcal mol−1 to form cis‐[Ni(NHC)2(PPh2)2]. In this case cis–trans isomerisation has a higher barrier but remains readily accessible.

The ability of the [Ni(NHC)2] moiety to afford isolable terminal phosphido products is in marked contrast to previous efforts to prepare analogues based on a [Ni(PR3)2] core, all of which have been reported to be unstable even at low temperatures. The latter often react on to form bridging phosphido dimers and higher nuclearity species, [20] suggesting phosphine loss as a key process in this onward reactivity. The higher binding energies of NHC ligands [59] may therefore contribute to the greater stabilities exhibited here. To quantify this hypothesis, calculations on ligand dissociation and subsequent dimer formation were performed on trans‐[NiL2(PH2)2] (L=PH3, IMe2; IMe2=1,3‐dimethylimidazolin‐2‐ylidene) where these simple model systems were employed to probe electronic effects directly.

As shown in Scheme 10, phosphine dissociation from trans‐[Ni(PH3)2(PH2)2] is significantly more accessible than NHC loss from trans‐[Ni(IMe2)2(PH2)2]. Moreover, subsequent dimer formation is exergonic for the phosphine system, but strongly endergonic with the NHC analogue. This provides some insight into the remarkable ability of the [Ni(NHC)2] fragment to not only effect a wide range of novel bond activation processes, but also to stabilize the resultant products.

Scheme 10.

Scheme 10

Open in a new tab

Computed free energies [kcal mol−1, B97D(benzene, BS2)//BP86(BS1)] for ligand loss and dimerization in trans‐[NiL2(PH2)2] (L=PH3, IMe2).

Conflict of interest

The authors declare no conflict of interest.

Supporting information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Click here for additional data file. (15.6MB, pdf)

Acknowledgements

We thank the Royal Society (Newton Fellowship to SS), EPSRC, the Deutsche Forschungsgemeinschaft (DFG; RA720/7), the Julius‐Maximilians‐Universität Würzburg and Heriot–Watt University for financial support.

S. Sabater, D. Schmidt, H. (. S. Schmidt, M. W. Kuntze-Fechner, T. Zell, C. J. Isaac, N. A. Rajabi, H. Grieve, W. J. M. Blackaby, J. P. Lowe, S. A. Macgregor, M. F. Mahon, U. Radius, M. K. Whittlesey, Chem. Eur. J. 2021, 27, 13221.

Contributor Information

Prof. Stuart A. Macgregor, Email: S.A.Macgregor@hw.ac.uk.

Prof. Dr. Udo Radius, Email: u.radius@uni-wuerzburg.de.

Prof. Michael K. Whittlesey, Email: m.k.whittlesey@bath.ac.uk.

References

  • 1. Arduengo A. J., Gamper S. F., Calabrese J. C., Davidson F., J. Am. Chem. Soc. 1994, 116, 4391–4394. [Google Scholar]
  • 2. 
  • 2a. Matsubara K., Miyazaki S., Koga Y., Nibu Y., Hashimura T., Matsumoto T., Organometallics 2008, 27, 6020–6024; [Google Scholar]
  • 2b. Danopoulos A. A., Pugh D., Dalton Trans. 2008, 30–31; [DOI] [PubMed] [Google Scholar]
  • 2c. Poulten R. C., Page M. J., Algarra A. G., Le Roy J. J., López I., Carter E., Llobet A., Macgregor S. A., Mahon M. F., Murphy D. M., Murugesu M., Whittlesey M. K., J. Am. Chem. Soc. 2013, 135, 13640–13643; [DOI] [PubMed] [Google Scholar]
  • 2d. Hoshimoto Y., Hayashi Y., Suzuki H., Ohashi M., Ogoshi S., Organometallics 2014, 33, 1276–1282. [Google Scholar]
  • 3.For an example with a cyclic (alkyl)(amino)(carbene) in which the oxidation state lies between Ni0 and NiI, see: Mondal K. C., Samuel P. P., Li Y., Roesky H. W., Roy S., Ackermann L., Sidhu N. S., Sheldrick G. M., Carl E., Demeshko S., De S., Parameswaran P., Ungur L., Chibotaru L. F., Andrada D. M., Eur. J. Inorg. Chem. 2014, 818–823. [Google Scholar]
  • 4. 
  • 4a. Clement N. D., Cavell K. J., Jones C., Elsevier C. J., Angew. Chem. Int. Ed. 2004, 43, 1277–1279; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2004, 116, 1297–1299; [Google Scholar]
  • 4b. Clement N. D., Cavell K. J., Ooi L., Organometallics 2006, 25, 4155–4165; [Google Scholar]
  • 4c. Miyazaki S., Koga Y., Matsumoto T., Matsubara K., Chem. Commun. 2010, 46, 1932–1934; [DOI] [PubMed] [Google Scholar]
  • 4d. Zhang K., Conda-Sheridan M., Cooke S. R., Louie J., Organometallics 2011, 30, 2546–2552; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4e. Lee C. H., Cook T. R., Nocera D. G., Inorg. Chem. 2011, 50, 714–716; [DOI] [PubMed] [Google Scholar]
  • 4f. Harrold N. D., Corcos A. R., Hillhouse G. L., J. Organomet. Chem. 2016, 813, 46–54; [Google Scholar]
  • 4g. Nelson D. J., Maseras F., Chem. Commun. 2018, 54, 10646–10649; [DOI] [PubMed] [Google Scholar]
  • 4h. Tendera L., Schaub T., Krahfuss M. J., Kuntze-Fechner M. W., Radius U., Eur. J. Inorg. Chem. 2020, 3194–3207. [Google Scholar]
  • 5.The high reactivity of [Ni(IMes)2] towards C−F bonds of fluoroarenes and a comparison with the reactivity of [Ni2(IiPr2)4(COD)] in C−F bond activation has very recently been reported.
  • 5a. Kuntze-Fechner M. W., Verplancke H., Tendera L., Diefenbach M., Krummenacher I., Braunschweig H., Marder T. B., Holthausen M. C., Radius U., Chem. Sci. 2020, 11, 11009–11023. For reports of C−F, C−Cl, and C−H borylation using [Ni(IMes)2] as a catalyst see; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5b. Zhou J., Kuntze-Fechner M., Bertermann R., Paul U. S. D., Berthel J. H. J., Friedrich A., Du Z., Marder T. B., Radius U., J. Am. Chem. Soc. 2016, 138, 5250–5253; [DOI] [PubMed] [Google Scholar]
  • 5c. Tian Y., Guo X., Kuntze-Fechner M., Krummenacher I., Braunschweig H., Radius U., Steffen A., Marder T. B., J. Am. Chem. Soc. 2018, 140, 17612–17623; [DOI] [PubMed] [Google Scholar]
  • 5d. Tian Y., Guo X., Wu Z., Friedrich A., Braunschweig H., Radius U., Marder T. B., J. Am. Chem. Soc. 2020, 142, 13136–13144; [DOI] [PubMed] [Google Scholar]
  • 5e. Tian Y., Guo X., Krummenacher I., Wu Z., Nitsch J., Braunschweig H., Radius U., Marder T. B., J. Am. Chem. Soc. 2020, 142, 18231–18242. [DOI] [PubMed] [Google Scholar]
  • 6.For a recent overview, see: Danopoulos A. A., Simler T., Braunstein P., Chem. Rev. 2019, 119, 3730–3961. [DOI] [PubMed] [Google Scholar]
  • 7. Arnold P. L., Cloke F. G. N., Geldbach T., Hitchcock P. B., Organometallics 1999, 18, 3228–3233. [Google Scholar]
  • 8. 
  • 8a. Schaub T., Radius U., Chem. Eur. J. 2005, 11, 5024–5030; [DOI] [PubMed] [Google Scholar]
  • 8b. Schaub T., Backes M., Radius U., Organometallics 2006, 25, 4196–4206. [Google Scholar]
  • 9. 
  • 9a. Fischer P., Linder T., Radius U., Z. Anorg. Allg. Chem. 2012, 638, 1491–1496; [Google Scholar]
  • 9b. Berthel J. H. J., Tendera L., Kuntze-Fechner M. W., Kuehn L., Radius U., Eur. J. Inorg. Chem. 2019, 3061–3072; [Google Scholar]
  • 9c. Kuehn L., Jammal D. G., Lubitz K., Marder T. B., Radius U., Chem. Eur. J. 2019, 25, 9514–9521. [DOI] [PubMed] [Google Scholar]
  • 10. 
  • 10a. Schaub T., Fischer P., Steffen A., Braun T., Radius U., Mix A., J. Am. Chem. Soc. 2008, 130, 9304–9317; [DOI] [PubMed] [Google Scholar]
  • 10b. Schaub T., Backes M., Radius U., Eur. J. Inorg. Chem. 2008, 2680–2690; [Google Scholar]
  • 10c. Schaub T., Fischer P., Meins T., Radius U., Eur. J. Inorg. Chem. 2011, 3122–3126; [Google Scholar]
  • 10d. Iwamoto H., Imiya H., Ohashi M., Ogoshi S., J. Am. Chem. Soc. 2020, 142, 19360–19367. [DOI] [PubMed] [Google Scholar]
  • 11. 
  • 11a. Zell T., Feierabend M., Halfter B., Radius U., J. Organomet. Chem. 2011, 696, 1380–1387; [Google Scholar]
  • 11b. Zell T., Fischer P., Schmidt D., Radius U., Organometallics 2012, 31, 5065–5073; [Google Scholar]
  • 11c. Zell T., Radius U., Z. Anorg. Allg. Chem. 2013, 639, 334–339. [Google Scholar]
  • 12. 
  • 12a. Schaub T., Backes M., Radius U., Chem. Commun. 2007, 2037–2039; [DOI] [PubMed] [Google Scholar]
  • 12b. Schaub T., Backes M., Plietzsch O., Radius U., Dalton Trans. 2009, 7071–7079. [DOI] [PubMed] [Google Scholar]
  • 13. 
  • 13a. Zell T., Schaub T., Radacki K., Radius U., Dalton Trans. 2011, 40, 1852–1854; [DOI] [PubMed] [Google Scholar]
  • 13b. Hauf C., Barquera-Lozada J. E., Meixner P., Eickerling G., Altmannshofer S., Stalke D., Zell T., Schmidt D., Radius U., Scherer W., Z. Anorg. Allg. Chem. 2013, 639, 1996–2004. [Google Scholar]
  • 14. 
  • 14a. Schaub T., Backes M., Radius U., J. Am. Chem. Soc. 2006, 128, 15964–15965; [DOI] [PubMed] [Google Scholar]
  • 14b. Zhou J., Berthel J. H. J., Kuntze-Fechner M. W., Friedrich A., Marder T. B., Radius U., J. Org. Chem. 2016, 81, 5789–5794; [DOI] [PubMed] [Google Scholar]
  • 14c. Kuntze-Fechner M. W., Kerpen C., Schmidt D., Häring M., Radius U., Eur. J. Inorg. Chem. 2019, 1767–1775. [Google Scholar]
  • 15. Fischer P., Götz K., Eichhorn A., Radius U., Organometallics 2012, 31, 1374–1383. [Google Scholar]
  • 16. Schmidt D., Zell T., Schaub T., Radius U., Dalton Trans. 2014, 43, 10816–10827. [DOI] [PubMed] [Google Scholar]
  • 17. Schaub T., Döring C., Radius U., Dalton Trans. 2007, 1993–2002. [DOI] [PubMed] [Google Scholar]
  • 18. Sabater S., Page M. J., Mahon M. F., Whittlesey M. K., Organometallics 2017, 36, 1776–1783. [Google Scholar]
  • 19. 
  • 19a. Chatt J., Davidson J. M., J. Chem. Soc. 1964, 2433–2445; [Google Scholar]
  • 19b. Bender R., Braunstein P., Dedieu A., Ellis P. D., Huggins B., Harvey P. D., Sappa E., Tiripicchio A., Inorg. Chem. 1996, 35, 1223–1234; [DOI] [PubMed] [Google Scholar]
  • 19c. Scriban C., Wicht D. K., Glueck D. S., Zakharov L. N., Golen J. A., Rheingold A. L., Organometallics 2006, 25, 3370–3378; [Google Scholar]
  • 19d. Mastrorilli P., Eur. J. Inorg. Chem. 2008, 4835–4850; [Google Scholar]
  • 19e. Beck R., Shoshani M., Krasinkiewicz J., Hatnean J. A., Johnson S. A., Dalton Trans. 2013, 42, 1461–1475; [DOI] [PubMed] [Google Scholar]
  • 19f. Varela-Izquierdo V., Geer A. M., de Bruin B., López J. A., Ciriano M. A., Tejel C., Chem. Eur. J. 2019, 25, 15915–15928; [DOI] [PubMed] [Google Scholar]
  • 19g. Berkefeld A., Reimann M., Hörner G., Kaupp M., Schubert H., Organometallics 2020, 39, 443–452. [Google Scholar]
  • 20. Fahey D. R., Mahan J. E., J. Am. Chem. Soc. 1976, 98, 4499–4503. [Google Scholar]
  • 21. 
  • 21a. Schäfer H., Z. Anorg. Allg. Chem. 1979, 459, 157–169; [Google Scholar]
  • 21b. Schäfer H., Binder D., Deppisch B., Mattern G., Z. Anorg. Allg. Chem. 1987, 546, 79–98; [Google Scholar]
  • 21c. Melenkivitz R., Mindiola D. J., Hillhouse G. L., J. Am. Chem. Soc. 2002, 124, 3846–3847; [DOI] [PubMed] [Google Scholar]
  • 21d. Madadi M., Najafabadi B. K., Fard M. A., Corrigan J. F., Eur. J. Inorg. Chem. 2015, 3094–3101; [Google Scholar]
  • 21e. Kessler J. A., Iluc V. M., Dalton Trans. 2017, 46, 12125–12131. [DOI] [PubMed] [Google Scholar]
  • 22.For very recent examples, see:
  • 22a. Hadlington T. J., Szilvási T., Driess M., J. Am. Chem. Soc. 2019, 141, 3304–3314; [DOI] [PubMed] [Google Scholar]
  • 22b. Lee K., Moore C. E., Thomas C. M., Organometallics 2020, 39, 2053–2056. [Google Scholar]
  • 23. 
  • 23a. Ganushevich Y. S., Miluykov V. A., Polyancev F. M., Latypov S. K., Loennecke P., Hey-Hawkins E., Yakhvarov D. G., Sinyashin O. G., Organometallics 2013, 32, 3914–3919; [Google Scholar]
  • 23b. Latypov S. K., Polyancev F. M., Ganushevich Y. S., Miluykov V. A., Sinyashin O. G., Dalton Trans. 2016, 45, 2053–2059. [DOI] [PubMed] [Google Scholar]
  • 24.Deposition Numbers 2009924 (for 2), 2009925 (for 3), 2009926 (for 4), 2009927 (for 6), 2009928 (for 7), 2009929 (for 11), 2009930 (for 16), 2009931 (for 17), 2009932 (for 18), 2009933 (for 19), 2009934 (for 21), 2009935 (for 22), 2009936 (for [Ni(IEt2Me2)]2(μ-PPh2)2, Supporting Information only) contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service.
  • 25. 
  • 25a. Ir: R. A. Schunn, Inorg. Chem. 1973, 12, 1573–1579; [Google Scholar]
  • 25b. Ir: E. A. V. Ebsworth, R. A. Mayo, Angew. Chem. Int. Ed. 1985, 24, 68–70; [Google Scholar]; Angew. Chem. 1985, 97, 65–66; [Google Scholar]
  • 25c. Ir: E. A. V. Ebsworth, R. A. Mayo, J. Chem. Soc. Dalton Trans. 1988, 477–484; [Google Scholar]
  • 25d. W: L. D. Field, N. G. Jones, P. Turner, J. Organomet. Chem. 1998, 571, 195–199; [Google Scholar]
  • 25e. Ru: V. S. Chan, I. C. Stewart, R. G. Bergman, F. D. Toste, J. Am. Chem. Soc. 2006, 128, 2786–2787. [DOI] [PubMed] [Google Scholar]
  • 26.This contrasts with cis phosphine-Ni−H splittings of ca. 65–77 Hz. Darensbourg M. Y., Ludwig M., Riordan C. G., Inorg. Chem. 1989, 28, 1630–1634. [Google Scholar]
  • 27. 
  • 27a.As reported in the Supporting Information, the phosphido hydrido complex 4 was crystallised from Et2O/pentane and the bis-phosphido complex 7 from benzene/hexane. Small amounts of the green NiI bridging-phosphido dimer, [Ni(IEt2Me2)]2(μ-PPh2)2, could be isolated if either 4 or 7 were left in the presence of Et2O for extended periods of time. This diamagnetic, dimeric compound was spectroscopically and structurally characterised (Supporting Information);
  • 27b.For an analogue bearing PiPr3 in place of the NHC, see ref. [19e].
  • 28.Even a single equivalent of PPh2H still gave a mixture of 6 and 7. Although this precluded isolation of 6, NMR spectra (measured at low temperature to slow the conversion through to 7) indicated it was isostructural with 3–5.
  • 29.Simply leaving 5 to stand in solution in the absence of any PPh2H failed to give 6, showing that bis-phosphido complex formation is not via two molecules of 5 combining, followed by ligand redistribution.
  • 30. 
  • 30a. Figueroa J. S., Cummins C. C., J. Am. Chem. Soc. 2003, 125, 4020–4021; [DOI] [PubMed] [Google Scholar]
  • 30b. Bai G., Wei P., Das A. K., Stephan D. W., Dalton Trans. 2006, 1141–1146; [DOI] [PubMed] [Google Scholar]
  • 30c. Yakhvarov D., Barbaro P., Carpio S. M., Midollini S., Orlandini A., Peruzzini M., Sinyashin O., Zanobini F., Angew. Chem. Int. Ed. 2006, 45, 4182–4185; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2006, 118, 4288–4291; [Google Scholar]
  • 30d. Greenberg S., Stephan D. W., Chem. Soc. Rev. 2008, 37, 1482–1489; [DOI] [PubMed] [Google Scholar]
  • 30e. Gómez-Ruiz S., Zahn S., Kirchner B., Böhlmann W., Hey-Hawkins E., Chem. Eur. J. 2008, 14, 8980–8985; [DOI] [PubMed] [Google Scholar]
  • 30f. Geier S. J., Stephan D. W., Chem. Commun. 2008, 2779–2781; [DOI] [PubMed] [Google Scholar]
  • 30g. Arisawa M., Sawahata K., Yamad T., Sarkar D., Yamaguchi M., Org. Lett. 2018, 20, 938–941; [DOI] [PubMed] [Google Scholar]
  • 30h. Modl M., Heinl S., Balázs G., Calvo F. D., Caporali M., Manca G., Keilwerth M., Meyer K., Peruzzini M., Scheer M., Chem. Eur. J. 2019, 25, 6300–6305. [DOI] [PubMed] [Google Scholar]
  • 31. 
  • 31a. Geier S. J., Stephan D. W., Chem. Commun. 2008, 99–101; [DOI] [PubMed] [Google Scholar]
  • 31b. Waterman R., Curr. Org. Chem. 2008, 12, 1322–1339. [Google Scholar]
  • 32. Chu T., Boyko Y., Korobkov I., Kuzmina L. G., Howard J. A. K., Nikonov G. I., Inorg. Chem. 2016, 55, 9099–9104. [DOI] [PubMed] [Google Scholar]
  • 33.Activation of the P−P bond of an ortho-carborane-substituted 1,2-diphosphetane by the cobalt(-I) species [K(thf)0.2][Co(η4-cod)2] has been described by Hey-Hawkins, Wolf and co-workers. Coburger P., Demeshko S., Rödl C., Hey-Hawkins E., Wolf R., Angew. Chem. Int. Ed. 2017, 56, 15871–15875; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2017, 129, 16087–16091. [Google Scholar]
  • 34.For representative examples, see:
  • 34a. Böhm V. P. W., Brookhart M., Angew. Chem. Int. Ed. 2001, 40, 4694–4696; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2011, 113, 4832–4834; [Google Scholar]
  • 34b. Han L.-B., Tilley T. D., J. Am. Chem. Soc. 2006, 128, 13698–13699; [DOI] [PubMed] [Google Scholar]
  • 34c. Geer A. M., Serrano Á. L., de Bruin B., Ciriano M. A., Tejel C., Angew. Chem. Int. Ed. 2015, 54, 472–475; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2015, 127, 482–485. [Google Scholar]
  • 35.Structures of the homoleptic terminal phosphido species [Hf(PCy2)5]− and [Mo(PCy2)4] have been reported. Baker R. T., Krusic P. J., Tulip T. H., Calabrese J. C., Wreford S. S., J. Am. Chem. Soc. 1983, 105, 6763–6765. [Google Scholar]
  • 36.For examples of bis-phosphido complexes with a cis-[R2P−M-PR2] arrangement, see:
  • 36a. Baker R. T., Whitney J. F., Wreford S. S., Orgametallics 1983, 2, 1049–1051; [Google Scholar]
  • 36b. Weber L., Meine G., Boese R., Augart N., Organometallics 1987, 6, 2484–2488; [Google Scholar]
  • 36c. Scriban C., Glueck D. S., DiPasquale A. G., Rheingold A. L., Organometallics 2006, 25, 5435–5448; [Google Scholar]
  • 36d. Chapp T. W., Schoenfeld A. J., Sanderson M. D., Chang C. L., Glueck D. S., Golen J. A., Moore C. E., Rheingold A. L., Organometallics 2012, 31, 5573–5585. [Google Scholar]
  • 37. Schäfer H., Binder D., Z. Anorg. Allg. Chem. 1988, 560, 65–79. [Google Scholar]
  • 38.The same species still formed even with up to 4 equiv. PArH2 and independent of the reaction temperature (−78 to 100 °C).
  • 39. 
  • 39a. Wicht D. K., Paisner S. N., Lew B. M., Glueck D. S., Yap G. P. A., Liable-Sands L. M., Rheingold A. L., Haar C. M., Nolan S. P., Organometallics 1998, 17, 652–660. For NMR data on the free phosphines, see: [Google Scholar]
  • 39b. Akitt J. W., Cragg R. H., Greenwood N. N., Chem. Commun. 1966, 5, 134–135. [Google Scholar]
  • 40. Bohle D. S., Clark G. R., Rickard C. E. F., Roper W. R., Taylor M. J., J. Organomet. Chem. 1988, 348, 385–409. [Google Scholar]
  • 41.The minor product was the dinuclear, P2-bridged complex [{Ni(IiPr2)2}2(μ,η2:2-P2)] which was reported previously from the reaction of [Ni2(IiPr2)4(COD)] with P4. Zarzycki B., Zell T., Schmidt D., Radius U., Eur. J. Inorg. Chem. 2013, 2051–2058. [Google Scholar]
  • 42. 
  • 42a. Bohle D. S., Jones T. C., Rickard C. E. F., Roper W. R., Organometallics 1986, 5, 1612–1619; [Google Scholar]
  • 42b. Deeming A. J., Doherty S., Marshall J. E., Powell J. L., Senior A. M., J. Chem. Soc. Dalton Trans. 1993, 1093–1100; [Google Scholar]
  • 42c. Driess M., Pritzkow H., Winkler U., J. Organomet. Chem. 1997, 529, 313–321; [Google Scholar]
  • 42d. Hickey A. K., Muñoz I., Lutz S. B. S. A., Pink M., Chen C. H., Smith J. M., Chem. Commun. 2017, 53, 412–415; [DOI] [PubMed] [Google Scholar]
  • 42e. Garner M. E., Arnold J., Organometallics 2017, 36, 4511–4514; [Google Scholar]
  • 42f. Wang D., Chen Q., Leng X., Deng L., Inorg. Chem. 2018, 57, 15600–15609; [DOI] [PubMed] [Google Scholar]
  • 42g. Rookes T. M., Wildman E. P., Balazs G., Gardner B. M., Wooles A. J., Gregson M., Tuna F., Scheer M., Liddle S. T., Angew. Chem. Int. Ed. 2018, 57, 1332–1336; [DOI] [PMC free article] [PubMed] [Google Scholar]; Angew. Chem. 2018, 130, 1346–1350. [Google Scholar]
  • 43. Bohle D. S., Clark G. R., Rickard C. E. F., Roper W. R., J. Organomet. Chem. 1990, 393, 243–285. [Google Scholar]
  • 44. Gómez-Ruiz S., Hey-Hawkins E., Dalton Trans. 2007, 5678–5683. [DOI] [PubMed] [Google Scholar]
  • 45. Ho J. W., Breen T. L., Ozarowski A., Stephan D. W., Inorg. Chem. 1994, 33, 865–870. [Google Scholar]
  • 46. Chatt J., Hitchcock P. B., Pidcock A., Warrens C. P., Dixon K. R., J. Chem. Soc. Dalton Trans. 1984, 2237–2244. [Google Scholar]
  • 47.For computational studies that consider phosphine dissociation, see:
  • 47a. Zhao Y., Truhlar D. G., Org. Lett. 2007, 9, 1967–1970; [DOI] [PubMed] [Google Scholar]
  • 47b. Minenkov Y., Occhipinti G., Jensen V. R., J. Phys. Chem. A 2009, 113, 11833–11844; [DOI] [PubMed] [Google Scholar]
  • 47c. Benitez D., Tkatchouk E., Goddard W. A., Organometallics 2009, 28, 2643–2645; [Google Scholar]
  • 47d. Sieffert N., Bühl M., Inorg. Chem. 2009, 48, 4622–4624; [DOI] [PubMed] [Google Scholar]
  • 47e. McMullin C. L., Jover J., Harvey J. N., Fey N., Dalton Trans. 2010, 39, 10833–10836; [DOI] [PubMed] [Google Scholar]
  • 47f. Ahlquist M. S. G., Norrby P.-O., Angew. Chem. Int. Ed. 2011, 50, 11794–11797; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2011, 123, 11998–12001. [Google Scholar]
  • 48.For recent related examples at nickel see:
  • 48a. Hooker L. V., Neufeldt S. R., Tetrahedron 2018, 74, 6717–6725; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48b. Entz E. D., Russell J. E. A., Hooker L. V., Neufeldt S. R., J. Am. Chem. Soc. 2020, 142, 15454–15463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.For recent reviews on the challenge of accuracy in computational chemistry, see:
  • 49a. Perrin L., Carr K. J. T., McKay D., McMullin C. L., Macgregor S. A., Eisenstein O., Struct. Bond. 2016, 167, 1–37; [Google Scholar]
  • 49b. Grimme S., Schreiner P. R., Angew. Chem. Int. Ed. 2018, 57, 4170–4176; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2004, 130, 4241–4248; [Google Scholar]
  • 49c. Harvey J. N., Himo F., Maseras F., Perrin L., ACS Catal. 2019, 9, 6803–6813. [Google Scholar]
  • 50. 
  • 50a.A second PPh2H ligand can add to I to give [Ni(IMe4)2(PPh2H)2], IVb, at −5.5 kcal mol−1. As this lies below the energy of 5, the observed product of P−H activation, the relative stability of IVb appears to be over-estimated with B97D and for the purposes of the discussion we have disregarded this. This energy difference is very functional dependent, but those functionals that place IVb higher in energy than 5 also gave too high a barrier for the P−H activation reactions. Identifying one functional that can capture all the features of the experimental system is highly problematic.
  • 50b.For a rare example of a [Ni(NHC)2(PR3)2] species, see: Ostericher A. L., Waldie K. M., Kubiak C. P., ACS Catal. 2018, 8, 9596–9603. [Google Scholar]
  • 51.The energy of this triplet species is again functional dependent but the behaviour of the B97D functional is generally similar to that of TPSS that was shown in a recent study to produce reliable spin-state energetics in a range of CoIII complexes. Neale S. E., Pantazis D. A., Macgregor S. A., Dalton Trans. 2020, 49, 6478–6487. See the Supporting Information for details. [DOI] [PubMed] [Google Scholar]
  • 52.The direct reaction of trans-III with PPh2H is possible but was found to be less accessible, proceeding through a transition state at +27.9 kcal mol−1 (Figure S79).
  • 53.A trigonal bipyramidal η2-H2 complex, Int(IV–V) and a corresponding H2 loss transition state, TS(IV–V)2 were located on the SCF surface, but TS(IV–V)2 falls below Int(IV–V) upon applying the energy corrections (Figure S71).
  • 54.The free energy of dissociation of IMe4 from cis-V is 20.5 kcal mol−1, ruling out a dissociative isomerisation pathway.
  • 55.Possible proton, H atom or hydride transfers from PPh2H to 5 were also assessed for the second P−H activation, but these proved significantly higher in energy (Figure S80).
  • 56. Kourkine I. V., Sargent M. D., Glueck D. S., Organometallics 1998, 17, 125–127. [Google Scholar]
  • 57.P−C bond activation in [Ni(IMe4)2(PPh3)] to form 1 was also computed and found to proceed with an overall barrier of 23.1 kcal mol−1 (Figure S77).
  • 58.The triplet form of VI lies at +18.5 kcal mol−1 and a transition state for P−H activation on the triplet surface was located at +21.3 kcal mol−1. This leads directly to 3 V, but this process is significantly higher than the reaction on the singlet surface.
  • 59.For experimental determinations, see:
  • 59a. Ru: L. Jafarpour, S. P. Nolan, J. Organomet. Chem. 2001, 617–618, 17–27; [Google Scholar]
  • 59b. Ru: A. C. Hillier, W. J. Sommer, B. S. Yong, J. L. Petersen, L. Cavallo, S. P. Nolan, Organometallics 2003, 22, 4322–4326; [Google Scholar]
  • 59c. Ni: R. Dorta, E. D. Stevens, N. M. Scott, C. Costabile, L. Cavallo, C. D. Hoff, S. P. Nolan, J. Am. Chem. Soc. 2005, 127, 2485–2495. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supporting Information

Click here for additional data file. (15.6MB, pdf)

Articles from Chemistry (Weinheim an Der Bergstrasse, Germany) are provided here courtesy of Wiley

RESOURCES