Highly active Pd(II) PCP-type catalysts for the Heck reaction

Full text of DOI:10.1021/ja9729692

J. Am. Chem. Soc. 1997, 119, 11687-11688
11687
Communications to the Editor
Highly Active Pd(II) PCP-Type Catalysts for the
Heck Reaction
Manuela Ohff, Andreas Ohff, Milko E. van der Boom, and
David Milstein*
Department of Organic Chemistry
The Weizmann Institute of Science, 76000 RehoVot, Israel
ReceiVed August 25, 1997
The palladium-catalyzed vinylation of aryl halides, also
known as the “Heck reaction” (eq 1) is a very useful synthetic
method for the generation of carbon-carbon bonds1 that has
found many applications.2 We report here a new type of
palladium catalyst for this reaction which shows outstanding
activities and yields. These catalysts are exceedingly thermally
stable and are not sensitive to oxygen. Our evidence suggests
that, contrary to the classical mechanism, in this case Pd(0) may
not be involved as the active species. The issue of a possible
Pd(II)/Pd(IV) cycle in Heck catalysis is currently under debate.3,4
Recently, highly efficient catalysis using cyclopalladated tri-o-
tolylphosphine complexes3,5 and palladium carbene complexes6
has been reported.
We have reported catalysis of the Heck reaction with aryl
chlorides using electron-rich trigonal bisphosphine palladium(0)
complexes.7 These reactions exhibit a remarkable chelate effect
and enable Heck coupling under reducing conditions in absence
of base,7c although at a moderate rate. In recent years, we have
studied complexes of tridentate PCP-type ligands in various
reactions.8 We have now explored the possibility of catalysis
of the Heck reaction with such palladium(II) complexes.
The new complexes 1, 2, and 3 were prepared by treatment
of Pd(TFA)2 (TFA ) OCOCF3) with the corresponding diphos-
phines in THF at 80 °C.9 Complex 3 was characterized by
Figure 1. Perspective view (ORTEP) of complex 3. Bond distances
(Å) and angles (deg, errors last digits in parentheses): Pd(1)-C(1) )
2.046(5); Pd(1)-O(4) ) 2.167(4); Pd(1)-P(2) ) 2.354(2); Pd(1)-
P(3) ) 2.381(2); C(1)-C(11) ) 1.489(7); C(1)-Pd(1)-O(4) )
175.3(2); P(2)-Pd(1)-P(3) ) 152.22(5); Pd(1)-C(1)-(C11) ) 99.5(3).
X-ray analysis (Figure 1).10 Complexes 1-3 show exceptionally
high activity in the catalytic arylation of olefins with aryl iodides
and bromides (Table 1). Because of the stabilizing tridentate
PCP ligand system, the complexes are extraordinarily thermally
stable and no decomposition is observed at temperatures up to
180 °C. These complexes are also not sensitive to oxygen and
moisture, and the reactions can be carried out in air, with no
change in efficiencies or yield. There is no noticeable catalyst
degradation even after heating complexes 1 and 2 for more than
300 h at 140 °C under the reaction conditions. The catalyst
remains highly active after the reaction is complete, and upon
addition of more substrates, catalysis is resumed, leading to the
coupling product in quantitative yield at essentially the same
rate. The only change observed at the end of the catalysis is
substitution of the trifluoroacetate ligand to give the corre-
sponding palladium halide complex, as shown by 31P NMR
spectroscopy.
While very high turnover numbers (TONs) and yields are
observed for all catalysts, complex 2 exhibits higher turnover
rates. Most experiments were performed with complexes 1 and
2.
In a typical experiment a slight excess of the olefin is added
to a solution of the aryl halide in freshly distilled N-methyl-
pyrrolidone (NMP), followed by addition of an equimolar
amount of sodium carbonate. Pure solvent and starting materials
should be used in the highly catalytic reactions. The catalyst
(1) (a) Heck, R. F. Palladium Reagents in Organic Synthesis; Academic
Press: London, 1985. (b) Heck, R. F. In ComprehensiVe Organic Synthesis;
Trost, B. M., Flemming, I., Eds.; Pergamon Press: Oxford and New York,
1991; Vol. 4, pp 833-863.
(7) (a) Ben-David, Y.; Portnoy, M.; Gozin, M.; Milstein, D. Organo-
metallics 1992, 11, 1995-1996. (b) Portnoy, M.; Ben-David, Y.; Rousso,
I.; Milstein, D. Organometallics 1994, 13, 3465-3479. (c) Portnoy, M.;
Ben-David, Y.; Milstein, D. Organometallics 1993, 12, 4734-4735.
(8) For example: (a) Rybtchinski, B.; Vigalok, A.; Ben-David, Y.;
Milstein, D. J. Am. Chem. Soc. 1996, 118, 12406-12415. (b) Van der Boom,
M. E.; Kraatz, H.-B.; Ben-David, Y.; Milstein, D. Chem. Commun. 1996,
2167-2168.
(9) Complexes 1-3, 5, and 6 have been fully characterized by a
combination of 1H, 31P{1H}, and 13C{1H} NMR, FD-MS, elemental analysis,
and X-ray analysis. See the Supporting Information for details.
(10) Crystal data for 3: C29H49O2P2F3Pd‚1.5C6D6, FW ) 772.18, yellow,
prism, 0.3 × 0.3 × 0.3 mm3, triclinic, P1h (no. 2), a ) 11.738(2) Å, b )
12.183(2) Å, c ) 15.259(3) Å, R ) 107.90(3)°, â ) 109.15(3)°, γ )
95.55(3)° from 25 reflections, T ) 110 K, V ) 1913.6(6) Å3, Z ) 2, Dc )
1.340 Mg/m3, µ ) 0.614 mm-1, Mo KR, 7026 independent reflections, Rint
) 0.0119, final R1 ) 0.0645.
(2) Recent reviews: (a) Grushin, V. V.; Alper, H. Chem. ReV. 1994, 94,
1047-1062. (b) de Meijere, A.; Meyer, F. E. Angew. Chem., Int. Ed. Engl.
1994, 33, 2379-2411. (c) Carbri, W.; Candiani, I. Acc. Chem. Res. 1995,
28, 2-7.
(3) (a) Herrmann, W. A.; Brossmer, C.; O¨ fele, K.; Reisinger, C.-P.;
Priermeier, T.; Beller, M.; Fisher, H. Angew. Chem., Int. Ed. Engl. 1995,
34, 1844-1848. (b) Beller, M.; Riermeier, T. H.; Haber, S.; Kleiner, H.-J.;
Herrmann, W. A. Chem. Ber. 1996, 129, 1259-1264.
(4) Louie, J.; Hartwig, J. F. Angew. Chem., Int. Ed. Engl. 1996, 35, 2359-
2361.
(5) (a) Herrmann, W. A.; Brossmer, C.; O¨ fele, K.; Beller, M.; Fisher,
H. J. Mol. Catal. A: Chem. 1995, 103, 133-146. (b) Herrmann, W. A.;
Reisinger, C.-P.; O¨ fele, K.; Brossmer, C.; Beller, M.; Fisher, H. J. Mol.
Catal. A: Chem. 1996, 108, 51-56.
(6) Herrmann, W. A.; Elison, M.; Fischer, J.; Ko¨cher, C.; Artus, G. R.
J. Angew. Chem., Int. Ed. Engl. 1995, 34, 2371-2374.
S0002-7863(97)02969-7 CCC: $14.00 © 1997 American Chemical Society

11688 J. Am. Chem. Soc., Vol. 119, No. 48, 1997
Communications to the Editor
Table 1. Selected Results of the Heck Reaction with Pd(II)
Chelate Complexes
not observed in our catalysis, a route involving a PCP-Pd(II)
hydride such as 4 is unlikely.
Generation of an anionic Pd(0) complex without chelate
opening (e.g., by deprotonation of a complex such as 4) is also
unlikely. Such a mechanism would involve nucleophilic attack
of the anionic complex on iodobenzene to give the Pd-Ph
complex 6.9 However, complex 6, prepared by reaction of 1
with PhLi, does not lead to any coupling products upon heating
with methyl acrylate. Moreover, treating 6 with various aryl
iodides results in the quantitative formation of the corresponding
palladium iodide complex 7 and biaryls (eq 3).12 Since biaryl
arene/
olefin catalyst
(mmol/ (mmol)
time/
temp
R
yield
ArX
mmol) × 10-5 (h)/(°C) TON (%)a
1
2
3
4
5
6
7
8
9
PhI
PhI
COOMe 5/6
50/60
3.5
17.5
8.75
1
1
1
2
2
2
3
1
1
2
2
1
2
2
2
2
60/40
142 900 100
253/140 267 600 94
350/140 520 500 91
20/140 142 900 100
40/140 277 700 97
40/140 528 700 95
40/140 142 900 100
50/60
5/6
5/6
50/60
5/6
3.5
1.75
9.0
3.5
3.5
3.5
3.5
3.5
COOBu 5/6
40/140
5 650
4
5/6
5/6
5/6
88/160 108 000 77
64/140 142 900 100
14/160 142 900 100
10
11
12 PhI
13 PhI
C(O)Me 1.34/1.5 3.5
40/140
11 500 30
Ph
5/6
3.5
3.5
3.5
3.5
60/140 133 000 93b
16/140 142 900 100
63/140 132 900 93
63/140 113 300 79
formation is not observed in our catalysis, such a mechanism
can be excluded. Thus, we believe that the traditional mech-
anism involving a Pd(0) complex may not be involved. A
Pd(II)/Pd(IV) cycle based on oxidative addition of the aryl halide
to 1-3 is a distinct possiblity.
14 p-MeOPhI COOMe 5/6
15 PhBr COOMe 5/6
16 p-OCHPhBr COOMe 5/6
a Determined by GC, based on haloarene and product (PhCHCH-
COOMe). b Product mixture of two isomers (E/Z ) 7/1).
A competitive experiment using p-BrPhI, PhI, p-MePhI,
p-MeOPhI, and methyl acrylate (in molar ratio of 5:5:5:5:1)
and complex 1 as a catalyst results in a linear correlation with
Hammett σ values, yielding F ) 1.39. While this electronic
effect is not surprising, since the oxidative addition of the aryl-I
bond to the metal is expected to be facilitated by electron-
withdrawing substituents on the aromatic ring, the F value is
too low to fit a rate-determining nucleophilic aromatic substitu-
tion.13 A subsequent rate-determining step with different
electronic requirements, such as olefin insertion, may account
for this observation.
The higher efficiency of complex 2 can be attributed to
electronic factors, the metal center in 2 being more electron
rich than in 1. Complex 3 is less efficient than 2 due to the
higher steric bulk of the phosphine ligands.
In summary, the PCP-type palladium(II) complexes show
exceedingly high catalytic activity in the Heck reaction, includ-
ing reactions of nonactivated aryl bromides. The new catalyst
system is very thermally and air stable. Our evidence suggests
that this catalysis may not involve the classical Pd(0) cycle.
Further investigations aimed at clarification of the scope and
mechanism of these novel catalysts are in progress.
is added and the mixture is stirred at 140 °C while the course
of the reaction is followed by NMR spectroscopy and gas
chromatography. Selected results are listed in Table 1.
With iodobenzene and methyl acrylate, essentially complete
conversions are obtained with exceptionally high TONs of up
to 500 000. Lower conversions and yields were realized with
aryl bromides, although the observed TON of 132 900 with 2
in the case of bromobenzene is among the highest reported.
Chlorobenzene was almost inactive. Thus, we believe that this
catalyst is among the most active and stable reported so far.
The new catalysts are of interest also from the mechanistic
point of view. While the mechanism has not been studied in
detail yet, involvement of a Pd(0) complex as the active species
seems unlikely. It was shown that a metalated Pd(II) o-
tolylphosphine complex can be reduced to a Pd(0) complex
under conditions of the Heck catalysis via formation of a
palladium(II) hydride followed by ring opening by C-H
reductive elimination.4 The presence of the two strong chelating
rings renders opening of the Pd-aryl bond during the reaction
unlikely. Nevertheless, to check this possibility, we have studied
the reactivity of the hydride complex 4.11 Significantly, reaction
of 4 with p-iodoanisole leads exclusively and quantitatively to
anisole and to the iodide complex 5 (eq 2),9 which was
Acknowledgment. We thank Dr. L. J. W. Shimon for the X-ray
analysis of complex 3. This work was supported by The Israel Science
Foundation, Jerusalem, Israel. M.O. thanks the Deutsche Forschungge-
meinschaft for a fellowship. A.O. thanks the Minerva Foundation,
Munich, Germany, for a fellowship. D.M. is the holder of the Israel
Matz Professorial Chair of Organic Chemistry.
Supporting Information Available: Spectroscopic characterization
of complexes 1-3, 5, and 6, text describing X-ray crystal structure
analysis, and tables of crystal data and structure refinement, atomic
coordinates, bond lengths and angles, anisotropic displacement param-
eters, and hydrogen coordinates for complex 3 (9 pages). See any current
masthead page for ordering and Internet access instructions.
independently prepared by reaction of the corresponding TFA
complex with LiI. Since hydrodehalogenation to the arene is
JA9729692
(12) H.-B. Kraatz, M. E. van der Boom, Y. Ben-David, and D. Milstein,
unpublished results.
(13) Portnoy, M.; Milstein, D. Organometallics 1993, 12, 1665-1673.
(11) Moulton C. J.; Shaw B. L. J. Chem. Soc., Dalton Trans. 1976,
1020-1024.