Gold-NHC-N -naphthamide complexes

Pure S a ,S and R a ,S atropoisomeric Au(I)- and Au(III)-NHC-complexes with benzimidazolyl N -(2-naphthamide) frameworks were prepared from appropriate axially chiral pre-ligands. The catalytic capacity of gold-NHC-N - naphthyl complexes was studied in cyclopropanation reactions. In contrast to corresponding unsuccessful Au(I)-NHC- N -naphthyl-oxazolyl complexes, all tested S a ,S and R a ,S diastereomers of Au(I) and Au(III)-NHC- N - naphthamide complexes were excellent catalysts to give both successful cyclopropanation (up to 99%, 15 min), as well as subsequent rapid in situ cis -to -trans isomerization. The results demonstrate that the new axially chiral Au(I)-/Au(III)-NHC-benzimidazolyl- N -naphthamide complexes represent an interesting group of gold catalysts with specific properties, affording fast cyclopropanation, excellent product yields and predictable trans -stereoselectivity (>99% yield; >99 % trans in 15 min).


Introduction
Gold catalysis has been a rapidly emerging field within transition metal catalysis in the past two decades to promote a great variety of organic scaffolds from unsaturated substrates. Gold has a high affinity towards carbon-carbon multiple bonds, especially alkynes, which may be activated towards nucleophilic attack. Coupled with high functional group tolerance, usually mild reaction conditions allow for diverse gold catalyzed transformations to be achieved, also including enantioselective reactions. 1 A broad review on the status of gold chemistry has lately been published as a thematic issue edited by Hashmi. 2 Gold(I) catalysis is by far more developed and understood compared to gold(III) catalysis, as evidenced by the large number of reported ligated gold(I) complexes. The last years have witnessed a revival of gold(III) chemistry. The challenge of gold(III) complexes is the possible reduction to gold(I) or gold(0) species. While the stability of gold(III) ions increases upon ligand coordination, stable gold(III) complexes generally show poorer catalytic activity. Thus, the aim for successful development of gold(III) catalysts is to find the balance between stability and catalytic activity. The development of gold(III) complexes and the applications of gold(III) in synthesis and catalysis, including mechanistic aspects, have recently been reviewed. 3 The interest toward the synthesis of chiral gold(III) complexes is also steadily growing and the progress achieved in the synthesis of well-defined chiral gold(III) complexes has lately been summarized. 4 We have previously reported studies on the gold(III) coordination ability of different ligands to give a variety of polydentate Au(III) complexes, [5][6][7][8][9][10][11][12][13] including oxazole and NHC based chiral Au(III) catalysts. The N,N-BOX-Au(III) complexes I (Scheme 1) based on bis-oxazoline ligands (BOX), were shown to represent an interesting group of Au(III) catalysts with specific catalytic properties. 5 Our experimental and theoretical studies on Au(III) bidentate coordination of a series of pyridine-oxazoline and quinoline-oxazoline II based ligands, concluded that the superior activity of the N,N-Au(III)-pyridine-oxazolyl complexes II is caused by decoordination of the pyridine-N ligand as a crucial step for efficient generation of catalytic activity. 6 Further mechanistic studies (NMR, X-ray, DFT) of Au(III)-bidentate pyridine-oxazoline II mediated alkoxycyclization, also demonstrated that de-coordination of the pyridine nitrogen is involved as the rate-limiting step. 7 Based on our successful synthesis of P,N-Au(III)-phosphine-oxazoline complexes III, 9 as well as the efficient coordination affinity of oxazoline-nitrogen to generate N,N-Au(III)-bis-heterocyclic structures II, we synthesized oxazoline functionalized Au(III)-NHC complexes IV. 12 These Au(III) complexes were too unstable for proper isolation and spectroscopic characterization, but selective 15 N NMR techniques provided valuable proofs of N-coordination by formation of N-ligated Au(III) complexes. The changes in 15 N-shift values ( 15 N), observed by 1 H, 15 N-HMBC 2D NMR studies, going from Au(I)Cl, via Au(III)Cl3 to the C,N-Au(III)-NHC-oxazolyl complexes, afforded important evidence that oxazoline-N-coordination to Au(III) took place. In particular, the huge up-field shift of the oxazoline-N ( 15 Noxaz) undoubtedly confirmed that the target six-membered bidentate C,N-Au(III)-NHC-oxazoline chelated complex IV was formed. Other studies showed that chiral alcohol functionalized Au(III)-NHC complexes V failed to generate bidentate C,O-Au(III)-NHC-alcohol complexes. 13 Several axially chiral Au(I) complexes, based on a binaphthyl scaffold have been synthesized. 14,15 Such Au(I) complexes, formed as atropoisomeric diastereomers with an axis of chirality, have also been designed as binaphthyl ligands connected to various N-heterocyclic carbene ligands. These synthesized axially chiral Au(I)-NHC complexes were, in general, reported to provide moderate to high catalytic activities and modest chiral inductions in asymmetric intramolecular cyclizations. [16][17][18][19][20][21][22][23][24][25] Scheme 1. Our a) previously reported N-oxazolyl or NHC Au(III) complexes I-V, [5][6][7]9,12,13 as well as b) the target Au(I)-and Au(III)-NHC-N-(2-naphthamide) (9)(10)(11)(12), and Au(I)-NHC-N-(2-naphthyl-oxazolyl complexes (13,14), 27 which are compared in the present study.
No examples of axially chiral gold(III) complexes with binaphthyl based ligands have been reported. Likewise, the use of stabilizing NHC ligands in gold(III) chemistry is much less developed than the NHC-gold(I) complexes, which have proved to be powerful catalysts. 26 By replacing one naphthyl group in the binaphthyl scaffold with a benzimidazolium unit, 27 the NHC functionality becomes incorporated in the axially chiral bisaryl structure (e.g. 9-14, Schemes 1 and 2). The aim of this proJect was to synthesize new axially chiral atropoisomeric NHC pre-ligands to allow the formation of modified axially chiral Au(I)/(III)-NHC complexes based on an N-naphthyl structure.
In addition to NMR ( 1 H, 13 C), IR and HRMS characterization of new N-naphthamide compounds 7-12, structures of selected products (5,9,11) shown with full assignment of 1 H and 13 C NMR data, based on 2D NMR studies (COSY, HSQC, HMBC), are available in the Supporting Information.

Catalytic properties of gold-NHC-N-naphthyl complexes in gold-catalyzed cyclopropanation
The catalytic potential of the Au(I) and Au(III)-NHC-N-naphthyl complexes 9-14 was studied in the goldcatalyzed cyclopropanation reaction of propargyl ester I and styrene II (Table 1). We have previously used this model reaction for evaluation of catalytic ability of other novel gold(I) and gold(III) catalysts, 5,6,11 hence, providing a solid and well-established background for comparison. The reactions were performed by addition of a silver salt (AgSbF6, 5 mol %) to a CH2Cl2 solution of the relevant Au complex (5 mol %, 9-14), to generate the catalytically active cationic gold species in the mixture of propargyl acetate I and styrene II. The yield and the stereoselectivity of the vinyl-cyclopropane product III were determined by 1 H NMR. The stereoselectivity, measured as the cis / trans ratio, was based on the ratio of the singlet integrals for the respective vinylic protons ( 5.90 and  6.08 ppm, structure III, Table 1).
In contrast to the initially established model for stereoselective cyclopropanations, 38 which explains a favored cis selectivity by steric interactions, we have previously shown that the stereoselective outcome is not fixed, since a more complex situation controls the stereochemistry of propargyl cyclopropanations. Our results demonstrated that the amounts of formed cis and trans isomers varies by cis-to-trans isomerization over time, and the stereoselective outcome is affected by the electronic properties, the bulkiness of substrates as well as the catalytic activity of the AuI or AuIII catalyst. 5 Some Au catalysts (e.g. BOX-Au(III) complexes I) (Scheme 1) were excellent for both fast cyclization into initial selective cis-cyclopropanes as well as subsequent complete in situ cis-to-trans isomerization. Thus, proper choice of Au catalyst successfully enabled highly selective formation of either cis or trans products (dr > 99%), and separate gold catalyzed (JohnPhosAu(I)SbF6) isomerization allowed the preparation of pure trans diastereomers (up to 98% yield) from corresponding pure cis substrates. The formation of trans isomers is proposed to proceed by Au-catalyzed ring-opening through different relevant intermediates, as shown in the suggested cis-to-trans isomerization pathways and a more detailed discussion presented in or previous work. 5 In the present studies, all the tested diastereomers of gold(I)-NHC (9, 10) and gold(III)-NHC (11, 12) N-naphthamide complexes were strong catalysts to afford cyclopropanation, as shown by complete conversion of propargyl substrate I into the target vinylcyclopropyl product III (100%, Table 1, entries 1-6) by vigorous stirring in 5-15 min. Also, predictable trans-stereoselectiviton p. 7)y of most reactions was excellent, as shown by up to 1:99 ratios of cis / trans cyclopropane III products (Table 1, entries 1-5). The results prove the combined ability of both the Au(I) and Au(III)-naphthamide complexes (Sa,S-and Ra,S-Au(I)-9; Sa,S-Au(I)-10, Sa,S-Au(III)-11 (R = Ph) and Sa,S-Au(III)-12 (R = iPr) to strongly activate for initial cyclopropanation as well as immediate in situ cis-to-trans-isomerization. The catalyst efficiency was demonstrated by the high-yielding preparation and isolation of the pure trans-isomer (91% isolated, >99% trans product III, Table 1, entry 1) provided by the Sa,S-Au(I)-9 (R = Ph) catalyst. The diastereomeric Ra,S-Au(III)-12 complex (R = iPr afforded slight isomerization and low trans-selectivity, in contrast to the Sa,S-12 isomer. The initial product mixture (cis / trans ratio 30:70, 15 min) did undergo slow isomerization over time (10:90 ratio, 2 h) to give the pure trans product (>99%) after 16 h (Table 1, 11 complex (R = Ph) could not be isolated from the oxidation reaction (Scheme 2) and was not available for comparison.
Despite the axially chiral nature of the applied new gold catalysts 9-12, no enantioselectivity was obtained (chiral HPLC) in the cyclopropanation reaction. The lack of stereocontrol is most likely due to the chiral environment provided by the ligand being too far from the reaction center.

Conclusions
A series of new pure Sa,S and Ra,S atropoisomeric diastereomers of Au(I) and Au(III)-NHC-N-naphthamide complexes (9-12; with iPr and Ph groups) were prepared from appropriate axially chiral pre-ligands. The catalytic capacity of gold-NHC-N-naphthyl complexes was studied in cyclopropanation. In contrast to the unsuccessful Sa,S-Au(I)-NHC-oxazole complexes 13 and 14, the Sa,S and Ra,S diastereomers of Au(I) and Au(III)-NHC-N-naphthamide complexes 9-11 and Sa,S-12 were excellent catalysts to give successful cyclopropanation (> 99% yield, 15 min). Additionally, subsequent rapid in situ cis-to-trans isomerization took place, thus affording predictable stereoselective trans-cyclopropyl products (trans > 99%) in 15 min. The Ra,S-Au(III)-12 (iPr) complex also gave immediate cyclopropanation, but afforded slow isomerization into pure trans (>99%, 16 h), indicating that the Ra,S-12 iPr-structure reduces the isomerization ability. No enantioselectivity was obtained in the reaction catalyzed with the axially chiral gold complexes.
The results demonstrate that the novel axially chiral Au(I)-as well as the Au(III)-NHC-N-naphthamide complexes (9)(10)(11)(12) represent an interesting group of gold catalysts with specific catalytic properties.

Experimental Section
General. Commercial grade reagents were used without any additional purification. Dry solvents were collected from a MB SPS-800 solvent purification system. Preparation of sensitive compounds was performed under dry conditions and inert atmosphere. All reactions were monitored by NMR and/or thin-layer chromatography (TLC) using silic a gel 60 F254 (0.25 mm thickness). TLC plates were developed using UV-light (254 nm) and/or phosphomolybdenic acid with heating. Flash chromatography was performed with Merck silica gel 60 (0.040-0.063 mm). 1 H and 13 C NMR spectra were recorded either a Bruker Avance DPX 400 MHz or a Bruker Avance III 600 MHz spectrometer. Chemical shift values for 1 H and 13 C NMR are reported in ppm (δ) down-field from tetramethylsilane (TMS) as an internal standard. Coupling constants (J) are reported in Hz. 1 H and 13 C NMR assignments of Sa,S-9, Ra,S-9 and Sa,S-11, based on 2D NMR studies (COSY, HSQC, HMBC) are available in Supp. Material. Accurate mass (HRMS) determination was performed on a "Synapt G2-S" Q-TOF instrument from Waters. Samples were ionized with an ESI probe with no chromatography separation performed before mass analysis. Calculated exact mass and spectra processing was done by Waters TM Software Masslynx V4.1 SCN871. IR spectra were recorded with a Bruker Alpha FT-IR spectrometer and OPUS V7.5 software was used for spectra analysis. Compounds 5,6,13,14 were prepared according to literature. 27