Palladium-Catalyzed Cross-Coupling Reactions of Perfluoro Organic Compounds

In this review, we summarize our recent development of palladium(0)-catalyzed cross-coupling reactions of perfluoro organic compounds with organometallic reagents. The oxidative addition of a C–F bond of tetrafluoroethylene (TFE) to palladium(0) was promoted by the addition of lithium iodide, affording a trifluorovinyl palladium(II) iodide. Based on this finding, the first palladium-catalyzed cross-coupling reaction of TFE with diarylzinc was developed in the presence of lithium iodide, affording α,β,β-trifluorostyrene derivatives in excellent yield. This coupling reaction was expanded to the novel Pd(0)/PR3-catalyzed cross-coupling reaction of TFE with arylboronates. In this reaction, the trifluorovinyl palladium(II) fluoride was a key reaction intermediate that required neither an extraneous base to enhance the reactivity of organoboronates nor a Lewis acid additive to promote the oxidative addition of a C–F bond. In addition, our strategy utilizing the synergetic effect of Pd(0) and lithium iodide could be applied to the C–F bond cleavage of unreactive hexafluorobenzene (C6F6), leading to the first Pd(0)-catalyzed cross-coupling reaction of C6F6 with diarylzinc compounds.

This review is the first report of the formation, structure and reactivity of a trifluorovinyl palladium(II) complex from the oxidative addition of the C-F bond of 1 to palladium(0) in the presence of LiI.The first palladium-catalyzed cross-coupling reaction of 1 with aryl zinc compounds in the presence of LiI is also discussed [41,57].We then discuss the development of the active Pd(0)/PR 3 species that enabled the oxidative addition of the C-F bond of 1 using no additives.By employing the Pd(0)/PR 3 species as a catalytic precursor, a Suzuki-Miyaura type of a cross-coupling reaction of 1 with arylboronates was successfully achieved [58].This cross-coupling reaction required neither an extraneous base to enhance the reactivity of organoboron reagents nor a Lewis acid to promote the oxidative addition of a C-F bond.
The transformation of perfluoroarenes into highly functionalized perfluoroaryl-substituted compounds is also an efficient and economical strategy.Radius et al. reported a coupling reaction of octafluorotoluene (C 7 F 8 ) and decafluorobiphenyl (C 12 F 10 ) with arylboronic acid in the presence of a catalytic amount of NHC-nickel(0) catalyst (where NHC represents N-Heterocyclic carbene) [59].This group also demonstrated the usefulness of a NHC-nickel(0) complex for the C-F bond activation of hexafluorobenzene (C 6 F 6 ), and the Ni(0)/NHC complex did indeed show catalytic activity toward the hydrodefluorination of C 6 F 6 [60][61][62].However, an efficient catalytic transformation of C 6 F 6 involving a C-C bond formation is very rare.To the best of our knowledge, only two examples of transition metal-catalyzed C-C bond formation reactions using C 6 F 6 to give biaryls have been reported [63,64].Yoshikai and Nakamura reported the coupling reaction of multi-fluorinated benzenes with aryl zinc catalyzed by a nickel catalyst ligated with alkoxydiphosphine, and that group also achieved the selective activation of a C-F bond [64].From a practical point of view, nonetheless, there remains no easily accessible catalyst system that is applicable for a coupling reaction that could introduce a perfluorinated aryl group to a certain position of arene compounds.We know the synergistic effect of Pd(0) and lithium iodide has been successfully applied to the C-F bond cleavage of C 6 F 6 , and the first development of the Pd(0)-catalyzed cross-coupling reaction of C 6 F 6 with diarylzinc has been achieved [65].This revies introduces a possible reaction path based on certain stoichiometric reactions and on the robustness of trans-(PCy 3 ) 2 Pd(I)(C 6 F 5 ) formed by the oxidative addition of C 6 F 6 to Pd(PCy 3 ) 2 in the presence of LiI.

Pd(0)-Catalyzed Cross-Coupling Reactions of Tetrafluoroethylene with Diarylzinc Reagents
The treatment of LiI with (η 2 -CF 2 =CF 2 )Pd(PPh 3 ) 2 (2a) in THF at room temperature promoted the oxidative addition of a C-F bond of THF to give a trifluorovinyl palladium(II) iodide (3; Scheme 1).In contrast to the known platinum analog, (η 2 -CF 2 =CF 2 )Pt(PPh 3 ) 2 [54], the C-F bond cleavage on palladium took place with no heating of the reaction mixture.An attempt to cleave the carbon-fluorine bond in 2a at 100 °C in the absence of LiI resulted in the decomposition of 2a along with the liberation of a TFE molecule and the precipitation of Pd black.Thus, cleavage of the C-F bond, generating 3, required LiI to as a Lewis acid to enhance the elimination ability of fluorine.The formation of a strong Li-F bond might also be important for oxidative addition at room temperature.The ORTEP drawing of 3 definitely shows that the palladium in 3 adopted a square-planar coordination geometry and was coordinated with two PPh 3 ligands in a trans manner (Figure 1).Complex 3 is the first example of a mononuclear trifluorovinyl complex generated by the carbon-fluorine bond cleavage of 1 with a well-defined structure [53,66].Complex 3 seemed to be a promising reaction intermediate for the preparation of various trifluorovinyl compounds via cross-coupling reactions of TFE with organometallic reagents.In particular, each reaction step in the cross-coupling of TFE with arylmetal reagents to give trifluorovinylarenes had to occur at a relatively low temperature, since the undesired side-reactions of the resultant trifluorovinylarenes gave a complex mixture.In fact, the [2 + 2] cyclodimerization of (α,β,β-trifluoro)styrene occurred in a head-to head manner at 80 °C to give a mixture of cis and trans isomers [27,67].Therefore, 3 was reacted with a stoichiometric amount of ZnPh 2 (4a) to determine if the expected reaction would occur at room temperature to give (α,β,β-trifluoro)styrene (5a).Although the reaction of 3 with 0.5 equiv of 4a in THF resulted in the formation of a complicated mixture that contained a small amount of the expected compound 5a, in the presence of LiI and DBA (trans,trans-dibenzylideneacetone), the reaction of 3 with 4a proceeded smoothly to give 5a in 87% yield (Scheme 2).Both LiI and DBA are potential additives in the cross-coupling reaction, since the reaction of TFE with Pd 2 (dba) 3 and PPh 3 in the presence of LiI, giving 3, simultaneously yielded an uncoordinated DBA.The role of lithium iodide in this reaction was the formation of reactive zincates such as Li[ArZnXI] (X = Ar or I, vide infra) [68].By contrast, platinum is an unlikely catalyst for the cross-coupling reaction, because the oxidative addition of TFE to platinum(0) requires both a much higher temperature and a longer reaction time (at 95 °C for 24 h) [54].
A logical extension of this reaction scheme was to conduct a Pd-catalyzed coupling reaction of TFE with diarylzinc in the presence of LiI, and the results are summarized in Table 1.In the presence of 2.5 mol% of Pd 2 (dba) 3 and 10 mol % of PPh 3 , the coupling reaction of 1 with 4a, which was prepared by treating ZnCl 2 with 2 equiv of PhMgBr in situ, took place at room temperature.The desired product 5a was obtained in 48% yield (entry 1).Under the same reaction conditions, the reaction with isolated ZnPh 2 occurred somewhat slowly compared with ZnPh 2 prepared in situ (entry 2).As expected from the stoichiometric reactions, the addition of lithium iodide was essential for the Pd-catalyzed coupling reaction (entry 3).Although either elongation of the reaction time or elevation of the reaction temperature was required, even with reduced catalyst loading (0.01 mol% of Pd 2 (dba) 3 ), the catalytic reaction proceeded smoothly at 40 °C to give 5a in 72% yield (entry 4).The rate of the coupling reaction was remarkably enhanced by the omission of PPh 3 from the catalytic system, and 5a was obtained in 73% yield (entry 5).By contrast, in the absence of Pd(0), the reactions of 1 with 4a were negligible, indicating that Pd(0) catalyzed the coupling reaction with or without lithium iodide (entries 6 and 7).Scheme 2. Reactions of 3 with ZnPh 2 (4a) in the presence of additives.
Table 1.Optimization of the reaction conditions for the Pd(0)-catalyzed cross-coupling reaction of 1 with 4a.General conditions: solvent; 0.5 mL.All reactions were conducted in a pressure-tight NMR tube.Yields, based on aryl group, were determined by 19 F NMR analysis of the crude product using α,α,α-trifluorotoluene as an internal standard.The scope of diarylzinc reagents was investigated using the optimized reaction conditions (Scheme 3).The treatment of 1 with 4a, which was prepared by the reaction of ZnCl 2 with PhMgCl, gave 5a in 81% yield, which marked the highest reactivity from among the arylzinc reagents (TON = 8100, entry 9).In addition, the reactions with Zn(4-Me-C 6 H 4 ) 2 (4b) and Zn(3-Me-C 6 H 4 ) 2 (4c) gave the monoaryl-substituted products 5b and 5c in 75% and 72% yields, respectively, while the reaction with Zn(2-Me-C 6 H 4 ) 2 (4d) gave only 57% yield of 5d.The reactions with fluoro-substituted aryl zinc reagents (4e and 4f) yielded the corresponding products (5e and 5f) in 53% and 55% yields, respectively.The reactions with p-substituted arylzinc reagents, such as Zn(4-MeO-C 6 H 4 ) 2 (4g) and Zn(4-styryl) 2 (4h), also afforded the corresponding products (5g and 5h) in good yields.By contrast, Zn(4-CF 3 -C 6 H 4 ) 2 (4i), Zn(4-Cl-C 6 H 4 ) 2 (4j) and Zn(4-MeS-C 6 H 4 ) 2 (4k) required prolonged reaction times to yield the corresponding products (5i-k) in moderate yields.In addition, the reaction with Zn(4-Me 2 N-C 6 H 4 ) 2 (4l) was terminated within 2 hours, and as a consequence, the yield of the desired product (5l) remained at 30%.Use of Zn(2-thienyl) 2 (4m) allowed the reaction with 1 to proceed to give 5m, although much longer reaction time was required and the product yield was low.This catalytic system was also successfully applied to Zn(2-naphthyl) 2 (4n), which gave the corresponding product (5n) in 61% yield.The reaction products were isolated as a THF solution due to the occurrence of the cyclodimerization to give hexafluoro-cyclobutane derivatives at a higher concentration [69].Relatively lower isolated yields were caused either by high volatility even at room temperature or by cyclodimerization.Scheme 3. Pd(0)-Catalyzed Coupling Reaction of TFE (1) with ZnAr 2 (4).General conditions: 1 (3.5 atm, >0.30 mmol, estimated from an equation of state), 4 (0.100 mmol, in situ prepared by treating of ZnCl 2 with 2 equiv of ArMgBr), solvent; 0.5 mL.All reactions were conducted in a pressure-tight NMR tube.Yields, based on aryl group, were determined by 19 F NMR analysis of the crude product using α,α,α-trifluorotoluene as an internal standard.The values in parentheses are of isolated yield.
Based on the results described above, the Pd-catalyzed monoaryl substitution reaction of 1 might proceed via the mechanism depicted in Scheme 4. Coordination of a TFE molecule to Pd(0) would take place to generate an η 2 -TFE species (B).Then oxidative addition of a C-F bond to Pd(0) is promoted by lithium iodide, generating a trifluorovinyl palladium(II) intermediate (C).Transmetalation of C with Li[ArZnXI] would yield a transient aryl palladium intermediate (D), which would undergo reductive elimination to afford (α,β,β-trifluoro)styrene derivative 5 along with regeneration of the Pd(0) species.The addition of lithium iodide is essential not only for accelerating cleavage of the carbon-fluorine bond, but also for enhancing the reactivity of arylzinc reagents via the formation of zincates, such as Li[ArZnXI].Scheme 4. A plausible reaction mechanism.

Pd(0)-Catalyzed Cross-Coupling Reactions of Tetrafluoroethylene with Arylboronates
Our next concern was to apply the C(sp 2 )-F bond activation methodology to a Suzuki-Miyaura type C-C bond formation reaction that generally offers the advantages of tolerance across a broad range of functional groups [70][71][72][73][74].Most of the reported Suzuki-Miyaura type cross-coupling reactions via C-F bond cleavage, employing either highly electron-deficient organofluorine compounds or those bearing a directing group, have traditionally been conducted in the presence of a base [59,[75][76][77][78][79][80][81][82][83][84][85], whereas fluoride anion itself is regarded as a good activator for neutral organoboron reagents.The role of a base in a Suzuki-Miyaura coupling reaction is generally considered to follow one of two patterns; either converting a neutral organoboron compound into a nucleophilic boronate, or converting a palladium halide intermediate into an active palladium species via a ligand exchange reaction by the base [86,87].In fact, Widdowson pointed out the possibility that the use of an extraneous base should be, in principle, catalytic [75].Such a reaction, however, has not been developed.Some coupling reactions with organoboran reagents are known to proceed under neutral conditions, in which such an active species as palladium alkoxy or acyl complex is generated in situ via the oxidative addition of a C-O bond [75,88,89].We speculated that if the transition-metal fluorides generated via C-F bond cleavage would have reactivity sufficiently high so as to act as a fluoride donor, the development of base-free C-C couplings with organoboron reagents would bring a significant concept to the Suzuki-Miyaura coupling reaction with organofluorine compounds.Thus, we started developing a base-free C-C bond formation reaction of 1 with arylboronates in the presence of a Pd(0) catalyst.
We began by seeking an active species that could cleave the C-F bond of 1 without additives, because our original protocol using LiI eliminated the chance to generate a transition-metal fluoride intermediate in return for efficient C-F bond cleavage [41,[54][55][56]90].As a result, the thermolysis of (η 2 -CF 2 =CF 2 )Pd(PCy 3 ) 2 (2b) in THF at 100 °C under a N 2 atmosphere underwent a C-F bond activation of 1 to give an expected trifluorovinylpalladium(II) fluoride ( 6) in 45% yield (Scheme 5).NMR observation revealed the concomitant formation of a palladium 2-perfluorobutenyl species (7) as well as Pd(PCy 3 ) 2 .Complex 7 was identified on the basis of the similarity of the 19 F NMR patterns observed in perfluoro2-butenyl zinc species, CF 3 (ZnX)C=CFCF 3 [91].The recovery of Pd(PCy 3 ) 2 (26%) indicated the existence of a coordination-dissociation equilibrium of 1 to palladium under the reaction conditions.Therefore, this reaction was carried out under a TFE atmosphere (1 atm), leading to an improvement in the yield of 6. B contrast, as already mentioned above, the corresponding palladium fluoride analog could not be generated at all by heating the PPh 3 analog 2a [41].In the 19 F NMR spectrum of 6, characteristic upfield-shifted resonance attributable to a fluorine adjacent to palladium appeared at −317.9 ppm.To the best of our knowledge, the examples of fluoropalladium complexes generated via the oxidative addition of a C-F bond are very rare [82,[92][93][94].In addition, complex 6 marked the first example of a structurally well-defined oxidative addition product of 1 on a transition metal without the use of Lewis acid additives.We next examined the reaction of 6 with a stoichiometric amount of 5,5-dimethyl-2-phenyl-1,3,2dioxaborinane (8a) to evaluate the degree of its reactivity toward organoborane reagents.The treatment of 6 with 4 equiv of 8a in the presence of DBA at 100 °C for 2 h afforded 5a in 75% yield (Scheme 6).In contrast, no C-C bond formation occurred, even for a prolonged reaction time, when 8a was treated with the corresponding palladium iodide (9a) that reacts with 4a.In addition, neither the corresponding palladium bromide (9b) nor chloride (9c) underwent a coupling reaction with 8a.These observations clearly show that the C-C bond formation with organoboron reagents is unique to palladium fluoride 6 among corresponding palladium halides.In fact, the Pd(0)-catalyzed coupling reaction of chlorotrifluoroethylene with 8a in the absence of a base gave no coupling products, probably due to the generation of the unreactive trifluorovinylpalladium chloride intermediate.As mentioned above, the Pd-catalyzed coupling reaction of chlorotrifluoroethylene with arylboronic acids in the presence of a base has been reported [36,37].Scheme 6. Reactivity of palladium trifluorovinyl halides towards 8a.
It seemed logical to apply this reaction scheme to a palladium-catalyzed cross-coupling reaction of 1 with 8a.In the presence of 10 mol % Pd(dba) 2 and 20 mol % PCy 3 , the coupling reaction of 1 with 8a proceeded at 100 °C, in the absence of any additive, to afford 5a in 66% yield.This result pointed out that the reaction takes place even in the absence of a base, while the use of a base is generally indispensible for the Suzuki-Miyaura coupling reaction to enhance the reactivity of organoboron reagents.The addition of CsCO 3 did not affect the yield of 5a.Further optimization of the cross-coupling reaction of 1 with 8a was carried out, and as a result, the reaction conducted at 100 °C in the presence of Pd(dba) 2 /P i Pr 3 in THF led to the formation of the desired product 5a in 83% yield [95].
The optimized reaction conditions were used to investigate the scope of the cross-coupling reaction with arylboronates (Scheme 7).The reactions with 4-anisyl, 4-vinylphenyl, and 4-trifluoromethylphenyl boronates (8g-i) also afforded the corresponding trifluorostyrene derivatives (5g-i) in good to moderate yields.Of the 4-halogenophenyl boronates employed, 4-fluorophenyl and 4-chlorophenyl boronates yielded p-fluoroand p-chloro-substituted (α,β,β-trifluoro)styrenes (5f and 5j) in 74% and 76% yields, respectively.In contrast, no coupling product was generated by employing 4-bromophenyl boronate, probably due to the occurrence of an undesired oxidative addition of the C-Br bond.In addition, the reactions with 2-and 1-naphthyl boronates (8n and 8o) gave 5n and 5o in 73% and 88% yields, respectively.Furthermore, the reaction with 1-pyrenyl boronates (8p) gave 5p in moderate yield.The reactions with 2-benzofulyl boronates (8q) yielded the corresponding products (5q) in moderate yield.Although this catalytic reaction leaves much to be desired regarding the catalyst loading and the product yield, it is of great significance in preparing substituted trifluorostyrenes bearing nitro, aldehyde, ester, and cyano groups (5r-u).These functional groups can easily react with Grignard reagents that are required for the in situ preparation of organozinc reagents, and therefore, products 5r-u were difficult to synthesize from a coupling reaction with organozinc reagents.In addition, bis-boronate reagents, such as 4,4'-biphenyl diboronate (8v), can be used to prepare monotrifluorovinyl compounds, for which the unreacted boronate moiety was applied in a further cross-coupling reaction to synthesize highly-functionalized derivatives.Scheme 7. Pd(0)-catalyzed base-free cross-coupling reaction of 1 with arylboronates (8).General conditions: 8 (1.00 mmol), solvent (10.0 mL), TFE (3.5 atm).Yields were determined by 19 F NMR analysis of the crude product using α,α,α-trifluorotoluene as an internal standard.The values in brackets are of isolated yields.a Using PnBu 3 instead of PiPr 3 ; b Reaction conditions: 8v (0.30 mmol), solvent (1.5 mL), TFE (30 mg, 0.30 mmol).NMR analysis revealed that 29% of 8v was remained and the bistrifluorovinyl compound was generated in 7% yield; c Reaction conditions: 8v (0.95 mmol), solvent (10.0 mL), TFE (100 mg, 1.00 mmol).After the isolation procedure, 130 mg (36%) of 8v was recovered and the bistrifluorovinyl compound was isolated in 10% yield.This base-free cross-coupling reaction with arylboronates can be successfully expanded to other organofluorine molecules.The reaction of vinylidene fluoride with 8o proceeded in the presence of Pd(0)/P i Pr 3 catalyst, to give 1-(1-fluorovinyl)naphthalene (10o) in 86% yield (Scheme 8a).In addition, the corresponding reaction with hexafluoropropylene gave a mixture of regioisomers (11o) (Scheme 8b), while Dmowski reported the reaction of CF 3 CF=CF 2 with PhMgBr to give an E/Z mixture of CF 3 CF=CFPh (E/Z = 83/17) [96].However, a Pd(0) catalyst was ineffective in a base-free coupling reaction of fluoroarenes.Scheme 8. Pd(0)-catalyzed base-free cross-coupling reactions of (a) vinylidene fluoride or (b) hexafluoropropylene with arylboronates 8o.Yields were determined by 19 F NMR analysis of the crude product using α,α,α-trifluorotoluene as an internal standard.The values in brackets are of isolated yields.
The base-free Pd-catalyzed monoaryl substitution of 1 might proceed as follows (Scheme 9).Coordination of a TFE molecule to Pd(0) would take place to generate an η 2 -TFE species (A).Then, the combination of Pd(0) and trialkylphosphines with a strong σ-donor ability would enable the oxidative addition of a C-F bond to Pd(0) with no additives, generating a trifluorovinylpalladium(II) fluoride intermediate (B).The transmetalation of B with arylboronates [97], would give C, followed by reductive elimination, which would afford 5 along with a regeneration of the Pd(0) species and boronefluorides.Another possible mechanism involving concerted bimolecular elimination via a five-membered transient intermediate could afford 5 [98].It should be emphasized that no extraneous base is required in this reaction, although extraneous base is generally requisite for the Suzuki-Miyaura cross-coupling reaction to promote a transmetalation step with organoboron reagents.Scheme 9. A plausible reaction mechanism.

Pd(0)-Catalyzed Cross-Coupling Reaction of Perfluoroarenes with Diarylzinc Reagents
Next, we developed the coupling reaction of perfluoroarenes including hexafluorobenzene with diarylzinc compounds, since our methodology suggested the possibility of a cleavage of the unreactive C-F bond of C 6 F 6 via the cooperation of Pd(0) and LiI.First, we simply applied the reaction conditions of the coupling reaction of TFE with Ar 2 Zn to the coupling reaction of C 6 F 6 (12) with Ar 2 Zn (Table 2).In the presence of 5 mol % of Pd 2 (dba) 3 , 20 mol % of PPh 3 , and 2.5 equiv of LiI at 60 °C in THF, the reaction of 12 with 4a, prepared in situ by reacting ZnCl 2 with 2 equiv of PhMgBr, gave a trace amount of pentafluorophenyl benzene (13a), and the 12 remained intact (entry 1).To promote the oxidative addition of 12 to palladium, Pd(PCy 3 ) 2 was examined as a catalyst precursor for the coupling reaction, and 13a was obtained in 70% yield (entry 2).When isolated 4a (purchased from Strem) was employed in the coupling reaction, 13a was obtained in 63% yield (entry 3).Because a catalytic reaction using pentafluoroiodobenzene, C 6 F 5 I, as a substrate gave only a trace amount of 13a (<5% yield, 70% S. M. recovered), we ruled out the possibility that C 6 F 5 I could be generated as a result of a nucleophilic attack reaction of iodide anion on 12.In the absence of a palladium catalyst, no coupling product was observed (entry 4).An increase in the amount of LiI improved the yield of 13a to 75% (entry 5), whereas in the absence of LiI, 13a was obtained in 5% yield even after a prolonged reaction time (entry 6).This result contrasted with that from the reaction of 1 with 4a generated in situ from PhMgBr and ZnCl 2 (Table 1, entry 1) and indicated that the addition of LiI was crucial for the occurrence of the coupling reaction.In the presence of PCy 3 , Pd(OAc) 2 was also effective as a catalyst for the coupling reaction (entry 7).A mixture of Pd 2 (dba) 3 and PCy 3 (5 and 20 mol % each) showed catalytic activity to give 13a in 77% yield, while much more catalyst loading (10 mol % Pd(0)) was required for smooth progress in the coupling reaction (entry 8).Employing either DCPE (1,2-dicyclohexylphosphinoethane) or DCPB (1,4-dicyclohexylphosphinobutane) obviously retarded the desired coupling reaction (entries 9 and 10).

Entry
To gain deeper insight into the reaction mechanism, stoichiometric reactions of 12 with Pd(0) complexes were carried out.In a previous report by Grushin, the reaction of 12 with Pd(PCy 3 ) 2 in THF at 70 °C for 24 h proceeded very slowly to yield a perfluorophenylpalladium(II) fluoride, trans-(PCy 3 ) 2 Pd(F)(C 6 F 5 ), in 3% yield [93].In the presence of LiI, on the other hand, the oxidative addition took place much faster to give a perfluorophenylpalladium(II) iodide, trans-(PCy 3 ) 2 Pd(I)(C 6 F 5 ) (24), which indicated that an important role of lithium iodide was to accelerate the oxidative addition (Scheme 11a).Although the catalytic reaction of 12 with 4a occurred in the presence of Pd 2 (dba) 3 and 4 equiv of PCy 3 to give 13a in 77% yield (Table 2, entry 7), an oxidative addition did not occur in the presence of DBA in the stoichiometric reaction at 60 °C (Scheme 11b).This result might have been due to an inhibition of the coordination of C 6 F 6 to palladium by DBA under the stoichiometric reaction conditions and could indicate why Pd(PCy 3 ) 2 is a more efficient catalyst than the combination of Pd 2 (dba) 3 and PCy 3 .In stark contrast, even in the presence of LiI, the oxidative addition of C 6 F 6 to Pd(PPh 3 ) 4 did not occur, which is consistent with the observation that no reaction occurred in the presence of PPh 3 (Table 2, entry 1).The ORTEP diagram of 24 definitely shows that the palladium in 24 adopts a square-planar coordination geometry and is coordinated with two PCy 3 ligands in a trans manner (Figure 2).A similar coordination geometry was observed in structurally well-defined Pd(II) complexes, such as trans-(PPh  To confirm whether 24 is an intermediate in the Pd(0)-catalyzed cross-coupling reaction of 12 with diarylzinc or not, a stoichiometric reaction of 24 with 4a was also carried out.As a result, only a yield of 5% of 13a was obtained from a stoichiometric reaction conducted at 60 °C for 7 h in the presence of an excess amount of LiI, whereas 13a was obtained under the catalytic reaction conditions mentioned above (60 °C for 6 h; Scheme 10).This result strongly indicates that 24 is an unlikely reaction intermediate due to the steric hindrance around the palladium center caused by the two bulkier PCy 3 ligands.Thus, we assumed that the oxidative addition of C 6 F 6 to Pd(PCy 3 ) 2 in the presence of LiI might involve a dissociation process of the PCy 3 ligand to give a transient species, (PCy 3 )Pd(C 6 F 5 )(I).The resultant 3-coordinated transient intermediate would undergo re-coordination of a PCy 3 ligand in the absence of 4a to yield the thermodynamically favored, and unreactive 24.On the other hand, in the presence of 4a, transmetalation between the transient iodopalladium(II) species and 4a took place smoothly to give the coupling product 13a.These assumptions are consistent with the results from Hartwig's kinetic studies wherein the oxidative addition of chlorobenzene to Pd(PCy 3 ) 2 , giving trans-(PCy 3 ) 2 Pd(Ph)(Cl), involved the dissociation process of a PCy 3 ligand at the initial stage of the reaction [102].Unfortunately, any attempt to prepare the transient intermediate failed due to its coordinative unsaturation, and therefore, cis-(C 6 F 5 )Pd(I)(py)(PCy 3 ) (25), in which pyridine acts as a labile ligand to generate a tentative 3-coordinate (PCy 3 )Pd(C 6 F 5 )(I) species, was prepared as an alternative catalytic precursor.
Scheme 12 summarizes the synthetic route to the preparation of 25.The reaction of (C 6 F 5 ) 2 Pd(Py) 2 (26) with PdCl 2 in acetone [103,104] followed by treatment with PCy 3 in pyridine resulted in the formation of cis-(C 6 F 5 )(Cl)Pd(Py)(PCy 3 ) (27) in 32% yield.Substitution of an iodide for the chloride ligand in 27 was accomplished by treating an acetone solution of 27 with an excess amount of NaI, giving the desired palladium(II) iodide 25 in 55% yield.The novel pentafluorophenyl palladium(II) iodide 25 was characterized on the basis of NMR spectroscopy and elemental analysis, as well as X-ray diffraction analysis.Both pyridine and PCy 3 ligands in 25 were situated in a mutual cis position with a square-planar Pd(II) geometry, as shown by X-ray diffraction (Figure 3).We then evaluated the reactivity of 25 toward 4a in the presence or absence of lithium iodide.In the presence of LiI (1.5 equivalent), 25 reacted smoothly with 4a, which was carefully purified by sublimation prior to use, in THF at room temperature to afford 13a in 63% yield as a sole product (Scheme 13).The addition of PCy 3 to this reaction mixture affected neither the yield nor the selectivity of the reaction product.On the other hand, in the absence of LiI, the reaction of 25 with 4a under the same reaction conditions afforded a pentafluorophenylzinc species, C 6 F 5 ZnX (X = I or C 6 F 5 ), as a major product (54%), and 13a was concomitantly obtained as a minor product in 27% yield.These observations suggest the following: (a) a transient (PCy 3 )Pd(C 6 F 5 )(I) species, generated via dissociation of the labile pyridine ligand of 25, could be crucial for the smooth occurrence of transmetalation between the palladium(II) species and 4a, and (b) the existence of LiI is essential for selective transmetalation to generate 13a.Based on those results, a plausible reaction mechanism was proposed, as shown in Scheme 14.In the presence of LiI, the oxidative addition of the C-F bond in 12 to Pd(0) would occur by dissociation of a PCy 3 ligand, forming a (C 6 F 5 )(I)Pd(PCy 3 ) intermediate (A).We speculate that the LiI-promoted C-F bond activation of 12 on Pd(PCy 3 ) 2 would take place to give the cis-oxidative addition product, and the rapid dissociation of a PCy 3 ligand then might occur due to the steric hindrance of the two bulkier PCy 3 ligands.On the basis of theoretical and experimental studies, Radius et al. assumed that a related C-F bond activation of 12 on a Ni(NHC) 2 fragment, yielding trans-[Ni(NHC) 2 (C 6 F 5 )(F)], would proceed via the corresponding cis-oxidative addition product [61].Transmetalation between A and 4 in the presence of LiI would take place to give a biarylpalladium(II) intermediate (B).This transmetalation step would progress in preference to the re-coordination of a PCy 3 ligand, giving an unreactive trans-(PCy 3 ) 2 Pd(C 6 F 5 )(I) (A').The role of LiI in this reaction step might be rationalized by the formation of a reactive zincate such as Li[ArZnXI] (X = Ar or I) that would enable the efficient formation of B [41,68].Then, reductive elimination from B followed by the re-coordination of a PCy 3 ligand would yield the coupling product along with a regeneration of the Pd(0) species.Another possible route for the coupling reaction might involve the formation of a dimer intermediate [(PCy 3 )Pd(C 6 F 5 )(µ-I)] 2 .Hor argued for the possibility that both catalytic pathways, via mononuclear cis/trans geometric isomers and a dinuclear iodide-bridged intermediate, might contribute to the Pd(0)-catalyzed coupling reaction of pentafluorophenyl iodide with phenylboronic acid [101].Scheme 14.A plausible reaction mechanism.

Conclusions
In this review, we reported recent results for a palladium-catalyzed cross-coupling reaction of perfluoroorganic compounds with organometallic reagents.We have developed the first palladium-catalyzed monoarylation of TFE by employing in situ-prepared diarylzinc reagents to yield (α,β,β-trifluoro)styrene derivatives in excellent yield and with high selectivity.C-F bond activation of TFE was achieved by the synergetic effects of the palladium(0) species and LiI to generate the trifluorovinyl palladium(II) intermediate.
We also demonstrated the Pd(0)/PR 3 -catalyzed cross-coupling reaction of TFE with arylboronate.This reaction required neither an extraneous base to enhance the reactivity of organoboronates nor a Lewis acid additive to promote the oxidative addition of a C-F bond.The key palladium(II) fluoride intermediate that showed a unique reactivity toward organoboron compounds was isolated.These results may open new avenues for the development of a base-free Suzuki-Miyaura coupling reaction including the in situ generation of a metal fluoride intermediate via C-F bond activation.Furthermore, our development of palladium-catalyzed selective transformations of TFE via C-F bond activation greatly increases the potential of 1 as a useful starting material for a variety of organofluorine compounds.
In addition, we developed a Pd(0)/PCy 3 -catalyzed cross-coupling reaction using C 6 F 6 with a variety of diarylzinc compounds to give the corresponding pentafluorobiaryls in good to excellent yields.Stoichiometric reactions employing model complexes, trans-(PCy 3 ) 2 Pd(I)(C 6 F 5 ) and cis-(C 6 F 5 )Pd(I)(py)(PCy 3 ), with diphenylzinc in the presence of lithium iodide revealed both the catalytic reaction mechanism and the role of lithium iodide in this catalytic reaction.The key intermediate in this catalytic cycle was a transient 3-coordinated, monophosphine-ligated species, (PCy 3 )Pd(C 6 F 5 )(I), which was generated by the oxidative addition of the C-F bond of C 6 F 6 to Pd(PCy 3 ) 2 followed by the dissociation of a PCy 3 ligand.In this catalytic reaction, lithium iodide accelerated the oxidative addition step and generated a reactive zincate such as Li[ArZnXI] (X = Ar or I) that enabled an efficient transmetalation with the key intermediate.We have also described how this catalytic reaction could be applied to other monocyclic perfluorinated compounds, such as octafluorotoluene and pentafluoropyridine, as well as to polycyclic perfluorinated compounds, such as perfluoronaphthalene and perfluorobiphenyl, to give the corresponding coupling products.

Figure 1 .
Figure 1.Molecular structure of 3 with thermal ellipsoids at the 30% probability level.H atoms are omitted for clarity.

Figure 2 .
Figure 2. Molecular structure of 24 with thermal ellipsoids at the 30% probability level.H atoms are omitted for clarity.

Figure 3 .
Figure 3. Molecular structure of 25 with thermal ellipsoids at the 30% probability level.H atoms are omitted for clarity.

Table 2 .
Optimization of the catalytic reaction of C 6 F 6