Synthesis of an extremely sterically shielding N -heterocyclic carbene ligand

The N-heterocyclic carbene (NHC) ligand IPr** features substituents of unprecedented steric demand (IPr** = 1,3-bis[2,6-bis[(4-tert -butylphenyl)methyl]-4-methylphenyl]imidazol-2-ylidene). The NHC structure is an advanced derivative of the IPr system, and of the IPr* by Berthon-Gelloz and Markó. In the IPr** ancillary ligand, two para -methyl are introduced and eight methyl groups of IPr are formally replaced by a 4-tert -butylphenyl substituent, respectively, thereby sterically shielding both a coordinated metal and its second ligand. Favorable features of ligand and late transition metal complexes include high solubility, aesthetic NMR spectra, and a tendency towards crystallization.


Introduction
N-Heterocyclic carbenes (NHC) have been increasingly used as ligands, particularly in transition metal complexes for homogenous catalysis. 1The main advantages of these ligands are their ability to kinetically stabilize highly reactive low-valent transition metal atoms with low coordination number. 2 A decisive feature of NHC ligands with ortho-substituted aromatic substituents is their steric shielding of the carbene ligand atom and the metal atom.IPr and its saturated derivative SIPr are typical representatives.Several factors allow the widespread use of the IPr ligand in homogenous catalysis: 1 its efficient preparation, 3 the ease of handling of the imidazolium precursor and its tendency to stabilize reactive transition metal intermediates. 4he Cavallo group developed the "buried volume" concept to quantify the steric demand of N-heterocyclic carbenes. 5The buried volume (in % Vbur) gives a measure of the space occupied by the NHC ligand in the first coordination sphere of the metal center (see Figure 1).The greater the steric bulk of the NHC, the bigger is its so-called "buried volume".It is important to notice that for the comparison of different buried volumes it is necessary to have the same calculation parameters as the sphere radii and the distance between the metal and the coordination carbene atom.The most cited values for buried volumes were calculated from (NHC)AuCl complexes with a sphere radius "r" of 3.5 Å and a distance "d" between the metal and the carbene of 2.0 Å (Figure 2). 6,715a Some prominent sterically demanding NHC are displayed in Figure 2. IBiox has been prepared by the Glorius group. 8Ligands of the CAAC type have been introduced by the Bertrand group. 9,10These ligands stand out due to their "flexible steric bulk", which enables them to influence transition metal catalysts' activities. 11,1215a Analogously, copper and silver complexes of a chiral IPr derivative with four 1-phenylethyl substituents have been prepared in the Gawley group.15b In this manuscript, we present the synthesis of a new and even more sterically demanding NHC ligand, IPr**, 1,3-bis[2,6-bis[bis-4-tert-butylphenyl)methyl]-4-methylphenyl]imidazol-2ylidene (Figure 3).

Results and Discussion
A ligand system tailored for the isolation of reactive intermediates Structurally, the IPr** ligand is only an octa-tert-butyl derivative of the IPr* system.Its novelty lies in the so far unique combination of steric demand, favorable spectroscopic features, high solubility, and tendency towards crystallization.The steric bulk of IPr** was tailored to stabilize reactive intermediates, and make possible their synthesis and isolation.Therefore, four of the eight 4-tert-butylphenyl substituents surround the coordinated metal center, and the metal's other ligands.The tert-butyl groups lead to intense signals in 1 H-NMR spectroscopy.The mulitplet signals in the aromatic region of IPr* are significantly simplified: Thus, mixtures of IPr** complexes are easier to analyze.Furthermore, the introduction of tert-butyl groups leads to a higher solubility in organic solvents, rendering possible low-temperature NMR spectroscopy of reactive IPr** complexes in inert solvents.Despite the high solubility, the tendency towards crystallization remains intact, which is relevant for purification and single-crystal X-ray structure analyses.

Synthesis
The synthesis of the imidazolium chloride precursor salt of IPr** consists of several steps, starting with inexpensive tert-butylbenzene (Scheme 1).The first step is the bromination of tert-butylbenzene to 1-brom-4-tert-butylbenzene. 16The latter is transformed to the Grignard reagent which reacts in situ with ethyl formate to a benzhydrol derivative. 17The following step is the dialkylation of p-toluidine with this benzhydrol mediated by concentrated HCl and ZnCl2. 14The aniline 1 was obtained in moderate yield on a multigram scale.The further conversion to the diimine 2 was more challenging.Using aqueous glyoxal as a reagent, MgSO4 as a water abstracting agent and dichloromethane as the solvent, only mixtures of the toluidine substrate 1 and the product 2 were be observed. 14As separation failed, both the solvent and the reaction conditions were modified: The method of Plenio et al. using formic acid as a catalyst and aqueous glyoxal as a reagent in a 1:1 mixture of methyl tert-butyl ether and ethanol initially failed to give access to the desired diimine 2. 18 However, after heating the reaction mixture to 58 ºC for ten days, the diimine precipitated as yellow solid.The diimine product 2 precipitated only in a solvent mixture of tert-butyl ether and ethanol, whereas the diimine precursor for IPr* precipitated in dichloromethane. 14The reaction time for the diimine precursor of IPr* amounted to only four days.Apparently, the greater steric bulk of the eight additional tert-butyl groups lead to a significant decrease in the reaction rate.Single-crystal structure analyses were performed for both the toluidine 1 and the diimine 2 (Figure 4).The final cyclization step towards IPr** .HCl (3) was again accomplished by a modified method of Berthon-Gelloz et al. (Scheme 1 and Figure 5). 14The number of signals in the 1 H-NMR spectrum of IPr** .HCl ( 3) is limited due to its high symmetry (Figure 6), which is C2v on the NMR time scale (Figure 6).The chemical shift of the imidazolium proton of 12.9 ppm is very close to the respective chemical shift in IPr* of 13.0 ppm. 14The imidazolium proton of IPr .HCl (HIm = 10.0 ppm, CDCl3) 13 features a highfield shift of nearly 3 ppm, presumably due to a less stable hydrogen bond.A dedicated C-H-Cl hydrogen bond is apparent in the single-crystal X-ray structure analysis of IPr** .HCl.The eight tert-butyl groups are divided into two sets of diastereotopic substituents.There is one 1 H NMR signal for four tert-butyl groups at the NCHN side of the imidazolium ring, and a second signal for the four tert-butyl groups at the NCH=CHN side.Of course, the analogous observation also applies to the aromatic signals of the eight tert-butylphenyl groups.We here also report the synthesis of the IPr** palladium complex 4. Imidazolium deprotonation and palladium coordination were successfully achieved by a modified literature protocol. 21 The obtained palladium complex was completely characterized, including a single-crystal Xray analysis.

Catalytic applicability
The complex (IPr**)Pd(py)Cl2 displays low catalytic activity in the copper-free Sonogashira coupling of phenylacetylene with iodobenzene (Scheme 3).Under standard reaction conditions reported in the literature, 21 an isolated yield of 27% diphenylacetylene product suggests that the Sonogashira coupling does not benefit from the steric demand of the IPr** system.No catalytic activity was observed in the Suzuki-Miyaura coupling between paratolylboronic acid and 4-chloroanisole (Scheme 4). 22Only traces of the desired product were detected by 1 H-NMR spectroscopy and EI-MS spectrometry.However, 4,4'-dimethylbiphenyl (2 %) was isolated by column chromatography, exactly matching the added amount of palladium(II) complex 4. According to the proposed general catalytic cycle for Pd-PEPPSI-IPr, this homocoupling product originates from the activation of the catalyst. 23Formation of a black precipitate was observed, presumably palladium metal.

Discussion of buried volumes
A buried volume 5,24 of the X-ray structure (IPr**)Pd(py)Cl2 was determined to 46.2 %.The same structure, used as input for a DFT structure optimization, resulted in a buried volume of 44.3 %.The BP86/LACVP** level of theory as implemented in the Jaguar program was used. 25,26For (IPr)Pd(py)Cl2, also obtained by DFT structure optimization, the corresponding buried volume of 34.1 % was much smaller.However, the computed (IPr*)Pd(py)Cl2 structure leads to a value of 44.6 %, indistinguishable to the IPr** value within the error margin of the computational approach.Apparently, the tert-butyl groups are too distant from the palladium atom in order to influence its buried volume.
However, the computed structures of (IPr**)AuCl (Vbur = 53 %) and of (IPr**)AgCl (Vbur = 54.5 %) feature the highest buried volumes reported so far.As in the palladium complex 4, experimental values are again expected to be even higher, since the density functionals do not reproduce van-der-Waals attractions that favor the proximity of IPr** substituents and the metal fragment.
The buried volume values derived from the (NHC)AgCl complexes are about 3 percentage points higher than the corresponding values of (NHC)AuCl complexes.The Pd(py)Cl2 fragment is significantly larger, the substituents of the NHC are pushed away from the metal, thus resulting in low buried volumes.Thus, the following order of buried volume values is derived: Vbur (NHC)AgCl > Vbur(NHC)AuCl >> Vbur(NHC)Pd(py)Cl2 The large difference of IPr**'s buried volumes for different metal fragments indicates a considerable ligand flexibility due to partial rotation of the four substituted benzhydryl groups.
A modified one-dimensional approach for the quantification of steric shielding As a modified model, we were interested in the steric shielding of a second ligand in trans position by the bulky ancillary NHC ligand.Taking the pyridine's nitrogen ligand atom as center of the sphere (Figure 7), a modified buried volume of 30.9 % was obtained.The addition of tert-butyl groups to the IPr* system does not result in higher buried volume values.Apparently, the tert-butyl groups are located outside of the sphere with the r' radius around the ligand atom.Thus, the IPr** ancillary ligand will provide steric shielding for a second ligand against bulky reactants.A two-dimensional representation for the quantification of steric shielding Chemical structures of NHC ligands are defined by the three space dimensions.The concept of buried volume is a one-dimensional approach.We present a less simplifying model that uses two dimensions for the reproduction of steric demand.The quantification of steric shielding of the central atom by the ancillary ligand is based on two parameters: (1) The distance of an atom X of the NHC ancillary ligand to the metal center.
(2) The angle between the carbene atom, the metal center and the respective atom X within the NHC ancillary ligand.For the 3-chloropyridine complex (IPr)Pd(3-ClC5H4N)Cl2, 27 steric shielding is defined by maximum  values of 82° for X = carbon atoms and 92.5° for X = hydrogen atoms (Figure 9).Thus, IPr's steric bulk is concentrated in an area up to 83° at metal substituent distances between 3 Å and 7 Å.For (IPr**)Pd(py)Cl2, only the atoms at the NCN side of the imidazolylidene have been considered.Figure 10 shows IPr**'s steric shielding up to carbene-palladium-X angles  of 105° for carbon atoms and 108.6° for hydrogen atoms at close range (Pd-X distance between 2 Å and 4 Å).At medium range (Pd-X distance between 4 Å to 6 Å), angles of up to 117.6° for carbon atoms and 126° for hydrogen atoms are covered.Without tert-butyl groups, the diagram equals that of the IPr* ligand core.In a more distant Pd-X range (> 6 Å), the tert-butyl groups result in shielding up to angles of 134° for X carbon atoms and 139° for hydrogen atoms.
While the standard approach to the calculation of buried volumes recovers the additional shielding of IPr** compared to IPr* only for some metal fragments, the two-dimensional approach delivers detailed insight into the steric consequences of the additional substituents.

Conclusions
The N-heterocyclic carbene system IPr** has been prepared and characterized.For the computed AuCl complex, a new record for steric shielding of late transition metals by an NHC ligand has been established.High solubility in organic solvents, aesthetic NMR spectra, and a high tendency towards crystallization have been demonstrated for the ligand precursor and a derived palladium(II) complex.The complex (IPr**)Pd(py)Cl2 has low catalytic activity in the Sonogashira reaction and no activity in the Suzuki-Myaura cross-coupling.A two-dimensional representation for the quantification of steric demand of bulky ancillary has been put forward, applied, and discussed.The additional steric shielding of the IPr** system will help to reject large substrates that would otherwise react with highly electrophilic or nucleophilic metal fragments.The unique steric demand of IPr** establishes an opportunity for the stabilization of reactive metal fragments that are proposed as intermediates in catalytic cycles.Palladium(0) arene complexes and cationic gold(I) carbene complexes are future targets for the detection by NMR spectroscopy, and isolation by crystallization.

Figure 1 .Figure 2 .
Figure 1.Representation of the sphere for the calculation of buried volume values (here with an IPr complex).

Figure 8 .
Figure 8. Modified model with the trans ligand L as center of the sphere.