The Suzuki Reaction in Aqueous Media Promoted by P, N Ligands

The synthesis and structure of palladium complexes of trisubstituted PTA derivatives, PTAR3, are described. Water-soluble phosphine ligands 1,3,5-triaza-7-phosphaadmantane (PTA), tris(aminomethyl)phosphine trihydrobromide, tri(aminomethyl) phosphine, 3,7-dimethyl-1,5,7-triaza-3-phosphabicyclo[3,3,1]nonane (RO-PTA), 3,7-diacetyl-1,3,7-triaza-5-phosphabicyclo[3.3.1]nonane (DAPTA), lithium 1,3,5-triaza-7-phosphaadamantane-6-carboxylate (PTA-CO2Li), 2,4,6-triphenyl-1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane, and 2,4,6-triphenyl-1,3,5-triaza-7-phosphatricyclo[3.3.1.1]decane were used as ligands for palladium catalyzed Suzuki reactions in aqueous media. RO-PTA in combination with palladium acetate or palladium chloride was the most active catalyst for Suzuki cross coupling of aryl bromides and phenylboronic acid at 80 °C in 1:1 water:acetonitrile. The activity of Pd(II) complexes of RO-PTA is comparable to PPh2(m-C6H4SO3Na) (TPPMS) and P(m-C6H4SO3Na)3 (TPPTS) and less active than tri(4,6-dimethyl-3-sulfonatophenyl)phosphine trisodium salt (TXPTS). Activated, deactivated, and sterically hindered aryl bromides were examined, with yields ranging from 50% to 90% in 6 h with 5% palladium precatalyst loading. X-ray crystal structures of (RO-PTA)PdCl2, (PTAR3)2PdCl2 (R = Ph, p-tert-butylC6H5), and PTAR3 (R = p-tert-butylC6H5) are reported.


Results and Discussion
Palladium complex 1 was obtained as a yellow solid in 78% yield by reacting RO-PTA and PdCl 2 (COD) in methylene chloride at room temperature for one hour (Scheme 1). Compound 1 is slightly soluble in organic solvents and soluble in water; however, aqueous solutions of 1 form black precipitates over several hours indicating decomposition. The square planar palladium center of 1, Figure 2, contains similar bond lengths and bond angles as the analogous palladium acetate derivative synthesized by Peruzzini and coworkers [53]. The 31 P-NMR spectrum of 1 contains a singlet at −26.6 ppm in D 2 O, downfield from the cis-palladium acetate derivative (−46.6 ppm) in CD 2 Cl 2 [53]. Peruzzini and coworkers obtained the 31  The PTA R3 ligands were synthesized by the reaction of P(CH 2 NH 2 ) 3 and either benzaldehyde or p-tertbutylbenzylaldehyde under acidic conditions (Scheme 2). X-ray quality crystals of 2 were grown by slow diffusion of diethyl ether into a methylene chloride solution of 2, Figure 3. Palladium complexes of the PTA R3 ligands were prepared by stirring two equivalents of PTA R3 and PdCl 2 (COD) in chloroform at room temperature overnight (Scheme 2). The products were obtained as yellow powders in 78% yield for 3 and 83% yield for 4. The 31 P-NMR spectrum of 4 contained the expected singlet resonance at −51.4 ppm in CDCl 3 . Compound 3 was obtained as a mixture of cis and trans isomers as seen by 31 P-NMR spectroscopy which contained a resonance at −51.5 for the trans isomer and one at −34.0 ppm for the cis isomer in CDCl 3 . X-ray quality crystals of the trans isomers of 3 and 4 were obtained and the structures (Figures 4 and 5) are similar to the previously reported trans-PdCl 2 (PTA PhOMe3 ) 2 [30].

Suzuki-Miyaura Coupling
Initial studies on the catalytic activity of Pd(II) complexes of the water soluble ligands described were performed using phenylboronic acid and bromobenzene in 1:1 H 2 O:CH 3 CN with sodium carbonate as a base and 5% Pd(OAc) 2 loading. The combination of PTA and Pd(OAc) 2 produced a modestly active catalyst at 80 °C over 72 h (66%, Table 1, entry 3). No product was observed at room temperature or at 50 °C. The catalyst decomposed quickly to a black precipitate. Formation of black precipitate is not surprising because PTA, with a cone angle of 103°, is not large enough to support the coordinatively unsaturated active catalyst. The addition of mercury to the reactions essentially shut down catalysis (Table 1, entries 4, 5, 7), indicating that colloidal palladium likely was involved. PTA derivatives were then employed as ligands in the palladium catalyzed Suzuki coupling ( Table 2). All the PTA derivatives explored here resulted in higher yields, 40%-91% than PTA in less time.  The DAPTA/Pd(OAc) 2 system provided only slightly better yields, 40%-42%, than PTA ( Table 2, entries 1 and 2). Catalysts generated from PTA R3 /Pd(OAc) 2 provided the product in moderate yields, 56%-65% ( Table 2, entries 3-6). The potentially chelating ligand PTA-CO 2 Li with Pd(OAc) 2 was slightly more effective with 60%-76% yields depending on the number of equivalents of ligand added (Table 2, entries 7-9). Catalysts derived from P(CH 2 NH 3 Br) 3 or P(CH 2 NH 2 ) 3 and Pd(OAc) 2 were more active with yields ranging from 74%-80% (Table 2, entries [10][11][12][13][14]. The most active system studied was RO-PTA with Pd(II) salts. The in situ catalyst (RO-PTA/Pd(II)) showed very good activity with yields between 82% to 89% depending on ration of ligand to Pd(II) ( Table 2, entry 15,16,18,19). The preformed catalyst (1) was the most active with yields of 91% obtained for the Suzuki coupling of bromobenzene and phenylboronic acid ( Table 2, entry 20).
The ratio of ligand to palladium also affected the amount of product produced. The largest change in yield was observed with PTA-CO 2 Li, as the L:Pd ratio increased from 1:1 to 2:1 the yield increased from 60 to 76%. Increasing the ratio to 3:1 ligand to palladium had little effect on catalysis ( Table 2, entry 7-9). Increasing the L:Pd ratio for P(CH 2 NH 3 Br) 3 (Table 2, entry 10-12) and P(CH 2 NH 2 ) 3 (Table 2, entry 13,14) from 1:1 to 2:1 resulted in only a small increased yield. Increasing the ratio of RO-PTA:Pd(II) from 1:2 to 1:3 resulted in a slight decrease in the coupling product (Table 2, entry 15,16,18,19). Changing the palladium source from Pd(OAc) 2 to PdCl 2 with RO-PTA resulted in a small difference in yield with PdCl 2 being slightly more active (Table 2, entry 15,16,18,19). It is important to note that the reaction remained clear during catalysis for entries 7-16 and 18-19; unlike the PTA/Pd(OAc) 2 system where palladium black was clearly visible. Addition of Hg to the reaction catalyzed by RO-PTA and Pd(OAc) 2 resulted in a decrease in yield from 86 to 52% (Table 2, entry 17) indicating that the reaction is mainly homogeneous. RO-PTA and P(CH 2 NH 2 ) 3 ligand are potential (P, N) bidentate ligands and due to the hemilabile functionality [54] the catalyst can be stabilized, reducing the amount of palladium black formed. Ligand free coupling reactions with Pd(OAc) 2 and PdCl 2 were explored under the above conditions as a control (

Catalyst Scope
Pd(OAc) 2 , RO-PTA, and sodium carbonate were used to study the catalyst scope under optimized reaction conditions. It is well-established that electron-deficient aryl bromides are good substrates in palladium catalyzed cross coupling reactions. With optimal reaction conditions in hand, the scope of the catalyst system was explored with a range of aryl bromides. Suzuki coupling yields were affected by the steric and electronic parameters of the aryl halides (Table 3). Electron neutral aryl bromides such as 4-bromotoluene (Table 3, entry 1) and electron deficient aryl bromides such as 4-bromo-benzonitrile (Table 3, entry 5) coupled well under the conditions described above. Electron donating aryl bromides such as 1-bromo-4-methoxybenzene (Table 3, entry 4) and sterically demanding aryl bromides such as 2-bromotoluene (Table 3, entry 2) resulted in decreased coupling. No catalytic turnover was observed with the sterically demanding 2-bromo-m-xylene (Table 3, entry 3). The sterically demanding election donating 2-bromoanisole resulted in a modest yield (Table 3, entry 6) comparable with electron donating aryl bromide 1-bromo-4-methoxybenzene (Table 3, entry 4). The sterically demanding electron withdrawing 1-bromo-2-nitrobenzene resulted in good but lower yield (Table 3, entry 7) than the electron withdrawing sterically unhindered 4-bromobenzonitrile (Table 3, entry 5). The catalytic activity of RO-PTA/Pd(OAc) 2 is comparable to Suzuki coupling utilizing water soluble phosphines such as TPPMS and TPPTS [19][20][21]. When compared to water soluble TXPTS and palladacylces developed by Shaughnessy et al. [22][23]27] or water soluble diamine ligands [25,26] catalysis by RO-PTA was much less effective.

General
Standard Schlenk and drybox techniques were used for all reactions unless noted. Prior to use, solvents were distilled under nitrogen from the appropriate drying agent (sodium/benzophenone for tetrahydrofuran, calcium hydride for hexanes; magnesium/iodine for methanol). Water (deionized) and acetonitrile were deoxygented by sparging with nitrogen. Deuterated NMR solvents were purchased from commercial sources and used as received. All NMR spectra were recorded on either a Varian NMR System 400 or Varian Unity Plus 500 FT-NMR spectrometer. 1 H-and 13 C-NMR spectra were referenced to a residual solvent relative to tetramethylsilane. Phosphorus chemical shifts are relative to an external reference of 85% phosphoric acid in D 2 O with positive values downfield of the reference. Tetrakis(hydroxymethyl)phosphonium chloride was obtained from Cytec and used without further purification. PTA [44,56], PTA Ph3 [30], P(CH 2 NH 3 Br) 3 [38], P(CH 2 NH 2 ) 3 [30], [Me-PTA] + I − [35], DAPTA [36,37], PTA-CO 2 Li [33], and PdCl 2 (COD) [57] were synthesized according to previously reported methods. The synthesis of ROPTA was performed by a modification of a method reported by Schmidbaur [35]. Palladium chloride and palladium acetate were purchased from Strem and stored in a drybox. Aryl bromides and phenylboronic acids were purchased from Acros Organics and used without further purification. [35]. To a mixture of condensed liquid ammonia (80 mL) and [Me-PTA] + I − (9.00 g, 30.2 mmol) was added sodium metal (878 mg, 38.2 mmol) at −78 °C until the color turned dark blue. Stirring was continued for 20 min at −78 °C. The ammonia was slowly evaporated at room temperature. To the residue was added hexanes (200 mL) and the resulting mixture vigorously stirred for several minutes before filtering under nitrogen. The hexane was removed under reduced pressure resulting in a white, crystalline solid (1.50 g, 29%). Spectral data were identical to previously reported data [35].

cis-(3,7-Dimethyl-1,5,7-triaza-3-phosphabicyclo[3.3.1]nonane)dichloro palladium (II) (1):
To a solution of PdCl 2 (COD) (142.8 mg, 0.5 mmol) in methylene chloride (15 mL) was added RO-PTA (86.5 mg, 0.5 mmol). Precipitation was observed after 10 min, but the reaction was stirred for another hour. The precipitate was filtered off, washed with methylene chloride (2 × 10 mL), collected and dried in vacuo to give the product as a yellow solid (136 mg, 78%). 1  at room temperature. Methanol was removed under reduced pressure. The resultant white residue was dissolved in methylene chloride (100 mL) and sodium bromide was extracted by water (2 × 50 mL). The organic layer was dried over anhydrous potassium carbonate, filtered, and the methylene chloride removed under reduced pressure. The residue was dissolved in methylene chloride (5 mL), absolute ethanol (120 mL) was added, and the flask was set in the freezer overnight. The precipitate was filtered off, washed with ethanol (2 × 15 mL) and dried in vacuo to give the product as a white, crystalline solid (1.063, 64%). 1

General Procedure for the Suzuki Coupling Reaction Aryl Halides and Arylboronic Acids
A round bottom flask equipped with stir bar in the drybox was charged with palladium chloride or palladium acetate (11.2 mg, 0.05 mmol), an appropriate amount of ligand, sodium carbonate (212 mg, 2.0 mmol) and phenylboronic acid (183 mg, 1.5 mmol). Deoxygenated 1:1 H 2 O:CH 3 CN (5 mL) and aryl halide (1.0 mmol) were added and the reaction was stirred at 80 °C for 6 h unless noted. The reaction was cooled to room temperature, saturated sodium bicarbonate (20 mL) was added, and the organics were extracted with ethyl acetate (3 × 30 mL). The combined ethyl acetate extracts were dried (MgSO 4 ) and the solvent was removed under reduced pressure. The crude material was flash chromatographed on a short silica gel column.

Mercury Experiment
A Schlenk flask with stir bar was charged with palladium chloride or palladium acetate (8.8 mg or 11.2 mg respectively, 0.05 mmol), PTA (17.3 mg, 0.10 mmol), and sodium carbonate (212.0 mg, 2.0 mmol) under N 2 atmosphere. Five mL of deoxygenated H 2 O was then added via syringe and the catalyst solution was stirred for 1.5 h. A 25 mL round bottom was charged with phenylboronic acid (183 mg, 1.5 mmol) and bromobenzene (105 μL, 1.00 mmol), equipped with a condenser under N 2 atmosphere. Deoxygenated CH 3 CN (5 mL) and catalyst solution (5 mL) were added via syringe. The reaction mixture was stirred for 15 min followed by the addition of a few drops of mercury before heating to 80 °C. After 24 or 48 h the reaction was allowed to cool to room temperature. The mixture was then extracted with dichloromethane (3 × 5 mL). The combined organic extracts were dried over Na 2 SO 4 and the solvent removed under reduced pressure. The crude product was purified on a short column of silica gel.

X-ray Crystallography
X-ray crystallographic data were obtained on a Bruker APEX CCD diffractometer. The structures were solved by direct methods and refined using SHELXTL, version 6.10 [62,63]. Crystallographic data and data collection parameters may be found in Table 4. CCDC 827744-827747 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html.

Conclusions
We have reported here the synthesis and structure of palladium(II) complexes of RO-PTA (1) and PTA R3 (3)(4). The air stable, water-soluble, and potentially hemilabile P,N ligand RO-PTA was successfully used for the Suzuki reaction in aqueous media. The combination of RO-PTA and palladium acetate generated an effective catalyst for the Suzuki coupling reaction. Electron neutral and electron deficient aryl bromide substrates coupled well with phenylboronic acid in good yields. The catalytic system was modestly effective in the Suzuki coupling reaction for electron-rich and sterically bulky aryl bromides. Catalytic activity of RO-PTA is comparable to TPPMS and TPPTS and less active than water-soluble diamines and phosphines like TXPTS.