Open Access Article
Ł.
Banach
,
P. A.
Guńka
and
W.
Buchowicz
*
Faculty of Chemistry, Warsaw University of Technology, Noakowskiego 3, 00-664 Warsaw, Poland. E-mail: wbuch@ch.pw.edu.pl
First published on 28th January 2016
The general synthesis of [Ni(Cp)(X)(NHC)] complexes from a nickel halide, CpLi, and a carbene solution is reported. This procedure yields unprecedented complexes with ring-expanded NHC ligands (RE-NHC) of six- (1a, 1b), seven- (1c), and eight-membered (1d) heterocycles. The NMR spectra of 1a–1d are consistent with the hindered rotation of Ni–Ccarbene and N–CMes bonds, while X-ray analyses of 1b, 1c, and 1d reveal a pronounced trans influence of the RE-NHC ligands. Complexes 1a–1e are efficient pre-catalysts in Kumada–Tamao–Corriu coupling with the maximum efficiency observed for complexes bearing the six-membered NHC.
Owing to their catalytic activity in many organic transformations, these air- and moisture-stable complexes are currently the object of intensive research.2–5 Efforts for tuning the properties of [Ni(Cp)(X)(NHC)] are focused on appropriate ligand selection within the scope of synthetic methodologies presented in Scheme 1, i.e. for five-membered NHCs only. Since the halide1a,3b–d,4 or Cp ligands5 can be readily displaced from the title complexes, NHC is the ligand of choice for modifications aimed at enhancing their catalytic activity. Recently, it has been shown that increasing the steric demand of NHCs by means of bulky substituents on the nitrogen atoms improved the efficiency of the arylamination of aromatic halides catalysed by [Ni(Cp)(X)(NHC)].3i Another way to enhance spatial requirements as well as the electron-donating strength of NHC ligands is the expansion of the carbenic heterocycle.6 Both of these properties are presumed to influence the reactivity of complexes by facilitating the formation of a catalytically active intermediate and increasing its stability. On the other hand, ring-expanded NHC (RE-NHC) may enable an intramolecular C–H activation, which can be detrimental to catalytic performance.7
Although RE-NHC nickel complexes are known, these studies are limited to only a few papers.7,8 To the best of our knowledge, to date there have been no reports on [Ni(Cp)(X)(RE-NHC)]. Driven by scientific curiosity to see if [Ni(Cp)(X)(RE-NHC)] complexes are of a greater catalytic activity than their congeners with five-membered “classical” NHCs, we decided to develop the synthesis of such complexes.
Our initial attempts to prepare the desired RE-NHC nickel complexes according to the procedure shown in Scheme 1a, i.e. from nickelocene and an appropriate carbene precursor, failed. Reactions involving the use of [Ni(acac)2] as the starting material were also unsuccessful (Scheme 1b). A probable reason for this lack of reactivity is the weak acidity of carbene precursors9 and steric congestion around the reacting centre.6 Inspired by the synthesis of related complexes with Cp*-functionalized NHC ligands,3g we decided to test a similar approach involving the addition of the two ligands (Cp and RE-NHC) to a nickel halide. Despite the plausible formation of different products, e.g. nickelocene and [Ni(X)2(RE-NHC)2], the desired [Ni(Cp)(X)(RE-NHC)] complexes were obtained in remarkably good yields as the only isolable products. More specifically, [Ni(DME)(X)2] (X = Br or Cl, DME = 1,2-dimethoxyethane) was treated first with a CpLi solution in THF to form what we believe is a dinuclear complex [Ni(Cp)(μ-X)]2.10 Subsequently, a freshly prepared RE-NHC solution, obtained by deprotonation of the carbene precursor with BuLi, was injected. After stirring overnight and conventional purification complexes 1a–1d were obtained in 32–41% yields (two steps from the Ni halide) as deep purple solids (Scheme 2).11
Moreover, to further expand the scope of our method, we have applied it to prepare a complex with a five-membered NHC. Thus, [Ni(Cp)(Br)(SIMes)]3c (SIMes = 1,3-bis(2,4,6-trimethylphenyl)-4,5-dihydroimidazol-2-ylidene) (1e) was obtained via this simple procedure in 61% yield (two steps from [Ni(DME)Br2]).
The NMR spectra of complexes 1a–1d were conventionally recorded in CDCl3 at ambient temperature. Signals originating from NHC ligands imply restricted Ni–Ccarbene and N–CMes bond rotation. Owing to the inequivalence of hydrogen atoms in each NCH2 group their signals appear in the 1H NMR spectra as unresolved multiplets. Moreover, as a result of the hindered rotation, all complexes 1a–1d have a characteristic set of signals for mesityl wingtips, showing two separated equal singlets (2
:
2 ratio) for meta hydrogens from phenyl rings and three singlets (6
:
6
:
6 ratio) for methyl groups, i.e. ortho methyls and meta hydrogens are nonequivalent within each mesityl moiety. Another interesting feature of 1H NMR spectra concerns the Cp group. While for [Ni(Cp)(X)(NHC)] with five-membered NHCs the Cp singlet appears from 5.2 to 4.6 ppm, its chemical shift for 1a–1d is unusual and strongly influenced by the size of the NHC as it varies from 4.21 ppm for 1b to 3.68 ppm for 1d. Moreover, all signals in the 1H NMR spectrum of 1d are strongly broadened. We tentatively interpret these phenomena in terms of spin equilibrium,4f,12i.e. increasing content of the high-spin form of the complex which is correlated to the NHC's ring size.
Crystal structure analysis of complexes 1b–1d revealed that they crystallize in the monoclinic crystal system and their molecular geometries are typical for nickel(II) NHC complexes (Fig. 1–3).13 The Ni–CCp bond lengths exhibit a pronounced trans influence on the RE-NHC ligands while the C–C bond lengths in Cp (see figure captions for selected bond lengths) show variations characteristic of the so-called “diene distortion”.14,15
Complex [Ni(Cp)(Cl)(IMes)] (IMes = 1,3-bis(2,4,6-trimethyl-phenyl)imidazol-2-ylidene) was reported as an active catalyst in the Kumada–Tamao–Corriu (KTC) coupling between p-tolylmagnesium bromide and phenyl O-sulfamate.3a To our surprise, further applications of [Ni(Cp)(X)(NHC)] complexes in this reaction are limited.3l,p,16 In order to get an insight into the influence of NHC ring expansion on the activity of [Ni(Cp)(X)(NHC)] in KTC coupling, we tested complexes 1a–1e in reactions of PhMgBr with halotoluenes (3-bromotoluene, 4-chlorotoluene). The expansion of the NHC ligand from a five- to six-membered ring proved to be beneficial for the catalytic activity (Table 1, entries 1–3, 9 and 10). A catalyst loading of only 1 mol% of 1a or 1b is sufficient to achieve conversions over 95% with very good selectivity at room temperature in 90 minutes.17 Also, the nature of the halogen ligand in the nickel complex appeared to have no influence on the catalytic performance (Table 1, entries 2 and 3). Further expansion of the NHC ligand to seven- or eight-membered rings resulted in diminished activity. However, high conversions could be achieved at longer reaction times (Table 1, entries 4–7). Selectivity is another parameter affected by ring expansion beyond the six-membered system (Table 1, entries 4, 5 and 7).
a
| Entry | Catalyst | n | X′ | Conversionb (%) | Ratio A : B : Cb |
|---|---|---|---|---|---|
| a All reactions in duplicate; reaction conditions: THF (2.8 mL), halotoluene (0.66 mmol), phenylmagnesium bromide solution in THF (0.81 mL, 1.00 mmol, 1.24 mol L−1), nickel complex (6.7 μmol, 1 mol%), room temperature, 90 min; see the ESI for details. b Determined with GC. c Reaction time = 21 h. | |||||
| 1 | 1e | 0 | 3-Br | 78 | 91 : 6 : 3 |
| 2 | 1a | 1 | 3-Br | 97 | 92 : 6 : 2 |
| 3 | 1b | 1 | 3-Br | 98 | 92 : 7 : 1 |
| 4 | 1c | 2 | 3-Br | 36 | 89 : 11 : 0 |
| 5c | 1c | 2 | 3-Br | 86 | 85 : 10 : 5 |
| 6 | 1d | 3 | 3-Br | Traces | — |
| 7c | 1d | 3 | 3-Br | 97 | 70 : 20 : 10 |
| 8 | 1a | 1 | 4-Cl | Traces | — |
| 9c | 1a | 1 | 4-Cl | 68 | 65 : 22 : 13 |
| 10c | 1e | 0 | 4-Cl | 42 | 54 : 36 : 10 |
Footnote |
| † Electronic supplementary information (ESI) available. CCDC 1429974–1429976. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5dt04663g |
| This journal is © The Royal Society of Chemistry 2016 |