β-Pinene and camphor based, pyrazole-tethered triarylphosphines as chiral P,N ligands for palladium

New, optically active, β -pinene and camphor-based pyrazole tethered phosphorus-nitrogen bidentate ligands were synthesized and their utility in asymmetric catalysis was explored with respect to asymmetric allylic alkylation, amination and Suzuki-Miyaura cross coupling reactions


Introduction
Asymmetric catalysis often requires design and synthesis of new chiral ligands to improve enantioselectivity in metal catalysed reactions.To date, a large number of P/N mixed donor ligands have been successfully used in asymmetric metal catalysed reactions [1][2][3][4][5][6][7][8][9][10] .These hemilabile ligands 11 are endowed with a soft phosphorus and a hard nitrogen donor atom.Phosphorus donor groups commonly include aryldialkyl or triarylphosphines, phosphites and heterophosphines where phosphorus is bound to nitrogen or any other hetero atom.Common nitrogen donors are pyridine, amine, imine, Schiff's base, azole etc.
Pyrazolate-bridged binuclear and polynuclear transition metal complexes have attracted special attention for many years [12][13] .Pyrazole being a π-excessive heterocycle is a poor π-acceptor.As a result, it has not been frequently used in catalysis with low-valent transition metals.However, pyrazole derivatives are relatively easy to synthesize, while substituents on the pyrazole nucleus permit electronic and steric control of the reactivity of metal centre 14 .We have used achiral chelating ligands derived from 1-arylpyrazoles to synthesize a variety of metal complexes, some of which has been successfully used in palladium catalysed coupling reactions [15][16] .It was considered worthwhile, therefore, to attach a chiral backbone to the pyrazole moiety so that enantiopure pyrazolyl ligands can be developed.There are a number of precedents where β-pinene [17][18][19][20][21] or camphor [22][23][24][25][26][27][28][29] moiety has been used as a chiral backbone to obtain optically pure ligands.

AUTHOR(S
In order to have an impression of the structure of the ligand L3 as well as a direct evidence of N, P coordination to the palladium centre, equimolar amount of ligand L3 was mixed with the palladium dimer [Pd(ƞ 3 -C 3 H 5 )Cl)] 2 in DCM in presence of AgSbF 6 .Unfortunately we were unable to grow X-ray grade crystals of the resulting complex.The 31 P NMR signal of the complex was shifted to 20.61 ppm from -15.48 ppm (in the ligand) indicating complex formation.The 1 H NMR spectrum was indeed very complex and ill-defined at room temperature.Although the ligand L3 has no signal in the region 2.5 to 6.0 ppm in the 1 H NMR spectrum, the spectrum of the complex featured seven peaks in this region, which may be attributed to π-C 3 H 5 moiety [33][34] .
When the ligand L3 was treated with Pd(CH 3 CN) 2 Cl 2 in 1:1 ratio in CHCl 3 , the resulting complex obtained, was crystallized from DCM/hexane.From the crystal structure depicted in Figure 1 it is evident that L3 acts as a bidentate, P, N-donor ligand for palladium where phosphorus and pyrazole occupy cis coordination sites.The Pd-Cl bond length trans to P-Pd and N-Pd bonds are 2.347 Å and 2.280 Å respectively, as would be expected from a π-acceptor and a σ-donor ligand 35 .The pseudo equatorial Ph ring has an angle of 119.49° with the P-Pd bond and is oriented nearly orthogonal to the plane of the complex.Such an orientation would offer a steric hindrance to diminish the formation of the diastereomer II (vide infra).A large number of chiral, P,N-ligands are known to be effective in Pd catalysed asymmetric allylic alkylation reactions.With the ligands L1, L2 and L3 in hand, we undertook evaluation of these ligands in palladium catalysed alkylation of rac-(E)-1,3-diphenylprop-2-enyl acetate with dimethyl malonate, a benchmark reaction that has been traditionally used to assess the efficacy of new ligands 36 .For allylic substitution reactions, the active chiral palladium(II) catalyst was generated by the reaction of 2 mol% of [Pd(ƞ 3-C 3 H 5 )Cl] 2 and 4 mol% of chiral N,P ligand in DCM at room temperature.The results obtained for each ligand under same condition have been displayed in Table-1.From the data it is clear that the catalyst system was highly active and produced the substitution product in high chemical yield.The camphor derived ligands L2 and L3 displayed higher enantioselectivities than the one derived from (-)-β-pinene, viz.L1.Introduction of -CF 3 group at the 3 position of pyrazole ring in L3 appears to have caused a slight decrease in the optical yield but afforded the product with a marginal increase in the chemical yield.The accepted mechanism for palladium catalysed allylic substitution postulates formation of a symmetrical Pd-allyl complex from the racemic substrate (Figure 2).The nucleophile can attack either of the two -allyl termini of two alternative diastereomeric π-allyl palladium complexes (I and II).That is, there are four possible reaction pathways of which two pathways (a and d) would lead to product 3a (with Sconfiguration) and the other two (b and c) would lead to product 3b (with R-configuration).
Figure 2. Symmetrical Pd-allyl complex from the racemic substrate (Figure 2 has been drawn maintaining a similarity with the crystal structure orientation).

AUTHOR(S)
With a ligand featuring two hetero-donor sites such as ours, overall stereochemistry of the ligand would determine the ratio of the two diastereomeric intermediates I and II at equilibrium 37 and the electronically different nature of the two donor centre would possibly direct the approach of the incoming nucleophiles.
It would be expected that the nucleophilic attack on these intermediates I and II could be regioselective since it is more likely to occur from a direction trans to phosphorus because phosphorus is a better π-acceptor than nitrogen.It can also be seen that the isomer II is more sterically encumbered than the corresponding I isomer: one of the terminal Ph ring of the allyl moiety can have unfavourable steric interaction with one Ph ring of the PPh 2 of the ligand.Therefore, in order to account for the observed R-configuration of the final major product 3b we can assume that the product arises from the intermediate I by attack of the nucleophile on to the allyl carbon trans to phosphorus (path-b).The enantio-differentiation, however, remained inadequate for a practical utility, presumably because the bicyclic core of the terpene was far removed from the metal-binding site.The gem-dimethyl-substituted bridge in ligand L1 is perhaps even farther removed for any effective chiral induction.
In ligand L3 where a 3-CF 3 group is present on pyrazole, a slight decrease of ee was observed compared to L2.The electron withdrawing group probably makes nitrogen a weaker donor and ends up lowering the energy differences between nucleophilic attack trans to P and N. Nevertheless, using L3, we obtained the highest yield of product.
In view of the superior yield, a variety of carbon and nitrogen nucleophiles were tested using L3 (Table 2).Though the yields are excellent with most of the nucleophiles, enatioselectivity was low to moderate.In terms of optical yield toluene was found to be a better solvent (entry-2, 4) than DCM though an increased amount of catalyst was required for high conversion.A primary amine like benzylamine however, gave poor yield 38 .The amination reaction required longer reaction time for completion and afforded products with inferior enantioselectivity.
Ligand L3 was also utilized in asymmetric Suzuki-Miyaura cross-coupling reaction to synthesize axially chiral biaryls.Axially chiral biaryls are important units in numerous natural products and constitute an important class of ligands for asymmetric catalysis.Examples of asymmetric Suzuki-Miyaura reaction are not numerous as it is usually difficult to couple two sterically hindered arenes.To the best of our knowledge, pyrazole tethered phosphine ligands have never been used in the asymmetric version of the Suzuki-Miyaura reaction.
The reaction of aryl bromide and aryl boronic acid with Pd catalyst and ligand L3 in presence of CsF in toluene afforded the coupling products in uniformly high yield (Table 3).A variety of 2-substituted aryl bromides were used.Enantioselectivity however, was far from satisfactory.When the methyl group is ortho to boronic acid instead of bromine (compare entry-4 with entry-5, Table 3), enantioselectivity was somewhat improved.

Conclusions
In summary, we synthesized chiral pyrazole tethered bidentate phosphorus-nitrogen ligands based on βpinene and camphor scaffold.The crystal structure of a representative complex established the bidentate mode of P, N-coordination to palladium.These ligands provided excellent chemical yields in asymmetric allylic substitution (with carbon and nitrogen nucleophiles) and Suzuki-Miyaura cross-coupling reactions but optical yields remained modest.The structural motifs are being modified further in our laboratory to improve the enantio-discrimination of such chiral ligands in these and other reactions.

Experimental Section
General.Unless otherwise noted all starting materials were obtained from commercial suppliers.Organic solvents were dried and distilled as described elsewhere.All moisture and air sensitive reactions were carried out in a oven-dried flask under argon atmosphere.Column chromatography was performed with silica gel 230 ~ 400 and 100 ~ 200 meshes.All 1 H NMR (300 and 500 MHz), 13 C NMR (75 and 125 MHz), 31 P NMR (212 MHz) and 19 F NMR (470.5 MHz) spectra were recorded in CDCl 3 solution and reported in ppm (δ).X-ray single crystal data were collected using MoKα (λ = 0.7107 Å) radiation on a SMART APEX II diffractometer equipped with CCD area detector.Data collection, data reduction, structure solution/refinement were carried out using the software package of SMART APEX.The structure was solved by Patterson method and refined in a routine manner.Non hydrogen atoms were treated anisotropically.The hydrogen atoms were geometrically fixed.

Preparation of bromo derivative (C) from keto aldehyde (B).
A solution of keto aldehyde (B) (2.2 g, 13.05 mmol) and 2-bromophenylhydrazine hydrochloride (3.0 g, 13.33 mmol) in dry methanol (140 mL) was refluxed for 9 h.Solvent was removed under reduced pressure and the resulting liquid was purified by flash column chromatography (silica gel, 8% acetone/ petroleum ether) to afford the bromo derivative (C) as yellow liquid (3.31

Preparation of bromo derivative (G) from 3-trifluoromethylhydroxymethylene-camphor (F).
To a solution of 3-trifluoromethylhydroxymethylene camphor F (1 g, 4 mmol) in dry methanol (20 mL), 2bromophenylhydrazine hydrochloride (0.89 g, 4 mmol) was added and heated under reflux for 18 h.Solvent was removed under reduced pressure, and resulting yellow solid was purified by column chromatography (silica gel, 3% ethyl acetate/petroleum ether) to give the desired product G as pale yellow solid (1.21

Preparation of L3 from bromo derivative (G).
To a solution of N-(2-bromophenyl)-pyrazole G (1 g, 2.5 mmol) in dry THF (8 ml), n-BuLi (1.9 mL, 1.6 M in hexane, 3.1 mmol) was added dropwise at -78 °C.The colour of the solution changed from yellow to dark brown.After stirring for 30 min, PPh 2 Cl (0.68 g, 3.1 mmol) was added dropwise at -78 °C and stirring was continued for 4 h.After work up, a pale yellow solid was obtained which was purified by flash column chromatography to give the pure crystalline product L3 (1.05 g, 83% yield) as pale yellow solid; mp 132-134 °C.

Figure 1 .
Figure 1.Synthesis and crystal structure of complex C3.