A survey of reaction conditions for palladium-catalyzed processes

Article abstract of DOI:10.1016/S0022-328X(98)00384-2

Various reaction conditions and catalysts including cyclometallated Pd^II-complexes and palladium on activated carbon have been tested for Pd-catalyzed reactions that are assumed to proceed via Pd^IV-intermediates or via ligand exchange reactions between Pd^II-intermediates.

Full text of DOI:10.1016/S0022-328X(98)00384-2

Journal of Organometallic Chemistry 555 (1998) 141–144
Preliminary communication
A survey of reaction conditions for palladium-catalyzed processes
Gerald Dyker a,*, Andreas Kellner b
a Institut fu¨r Synthesechemie, Fachbereich 6, Gerhard-Mercator-Uni6ersita¨t-GH Duisburg, Lotharstraße 1, D-47048 Duisburg, Germany
b Institut fu¨r Organische Chemie, Technische Uni6ersita¨t Braunschweig, Hagenring 30, D-38106 Braunschweig, Germany
Received 26 September 1997; received in revised form 20 November 1997
Abstract
Various reaction conditions and catalysts including cyclometallated PdII-complexes and palladium on activated carbon have
been tested for Pd-catalyzed reactions that are assumed to proceed via PdIV-intermediates or via ligand exchange reactions
between PdII-intermediates. © 1998 Elsevier Science S.A. All rights reserved.
Keywords: Heck reaction; Palladium catalysis; Ullmann reaction
Recently, some palladium-catalyzed reactions have
been developed that are assumed to involve PdIV-inter-
mediates [1,2]. Originally these processes were found to
proceed under Jeffery-conditions [3] (a catalytic amount
of palladium acetate, K2CO3, tetra-n-butyl ammonium
bromide [4], DMF), without any neutral ligand except
for the solvent DMF. Now we have tested several
reaction conditions and palladium catalysts in order to
study their influence on the chemoselectivity of these
reactions. Palladium on carbon was applied as an ex-
ample for a heterogenous catalyst. Another focus of
interest was the influence of phosphane ligands. Partic-
ularly interesting was the question, how the cyclometal-
lated complex 1 introduced as a catalyst by Herrmann
et al. [5] would perform: In 1983 Spencer reported that
tris-ortho-tolyl phosphane is a superior ligand for the
Heck reaction, achieving turnover numbers up to
130 000 [6]. Heck and coworkers [7] found that under
the typical reaction conditions this ligand leads to the
formation of cylometallated complexes of type 1 (with
bromide and iodide as bridging ligands), but stated that
these PdII-complexes are not the active catalyst. In
contrast, in a series of publications Herrmann et al. [5]
reported that the complex 1 is more active than the in
situ formed system, achieving turnover numbers up to
1 000 000 and allowing the Heck reaction to be carried
out with notoriously sluggish aryl chlorides. The ex-
traordinary activity of the dimer 1 or of the corre-
sponding mono-nuclear monomer is explained by its
increased stability against P–C bond scission. Accord-
ing to Hartwig [8] the PdII-complex 1 is reduced to
Pd0-species under various reaction conditions and in
the case of the Stille reaction there is evidence that a
Pd0-species is the active catalyst. However, in the case
of the Heck reaction Herrmann and coworkers could
not detect such a Pd0-species by NMR experiments.
This surprising result leads to the conclusion that either
an extremely active Pd0-catalyst in very low concentra-
tion is present or different mechanisms such as PdII–
PdIV-cycles have to be taken into consideration even for
the Heck reaction.
* Corresponding author. Fax: +49 203 3794192.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(98)00384-2

142
G. Dyker, A. Kellner / Journal of Organometallic Chemistry 555 (1998) 141–144
The Pd-catalyzed domino process condensating three
Pd-aggregates and in all other experiments including
entries 10 and 11 a typical black solution had been
formed, this type of reaction might actually take place
at the surface of palladium clusters [10], a hypothesis
that is currently under investigation. In comparison, the
heterogeneous reaction with palladium on carbon as
catalyst is indeed successful (entry 9).
equivalents of ortho-iodoanisol 2 to give the substituted
dibenzopyrane 3 [9] is best explained by assuming PdIV-
intermediates, especially for the CH-activation at the
methoxy group. Presumably the catalytic active palla-
dium has to adopt oxidation states ranging from Pd0 to
PdIV and therefore an apparent PdII-catalyst such as 1
Also for Pd-catalyzed Ullmann reactions [11] such as
the reductive coupling of 6 and for annelation reactions
of iodobenzene (8) with tolane (9) [12] PdIV-intermedi-
ates and alternatively ligand exchange reactions [13]
between PdII-species have been suggested to explain the
formation of the biaryl bonds. For both processes the
cyclometallated complex 1 proved to be suitable (Table
2, entries 14 and 15; Table 3, entries 22 and 23).
Considering the difficulties to develop a mechanistic
rationale for these reactions starting from an apparent
PdII-catalyst such as 1 or its monomer it is more likely
that a catalytic active Pd0-species is formed in situ
should not be suitable. Therefore we anticipated that 1
should either fail to catalyze the reaction or at least
should result in a special product distribution. How-
ever, as it turned out 1 behaves similar to palladium
acetate combined with two equivalents of triphenyl
phospane (Table 1, entries 7 and 10): this result sug-
gests that the PdII-complex 1 is not the catalytic active
species in this case. Additional information derived
from Table 1: tetraalkyl ammonium halides are crucial
for achieving high yields of 3 (entry 2); water disturbs
the domino process and at the same time favours the
formation of the hydrolysis product 4 [9] (entries 3 and
4). An excess of either phenanthroline or triphenyl
phospane completely inhibits the domino process (en-
tries 6 and 8). Since a higher concentration of these
stabilizing ligands obviously prevents the formation of
under the reaction conditions [14]. Pd-clusters or other
Pd-aggregates are ruled out for entries 22 and 23 since

G. Dyker, A. Kellner / Journal of Organometallic Chemistry 555 (1998) 141–144
143
Table 1
CH-activation at the methoxy group of 2; a: 2 mmol of 2, 5 mol-% Pd(OAc)2 (or 15 mol-% Pd on carbon or 2 mol-% 1), 2 mmol n-Bu4NBr, 8
mmol K2CO3, 10 ml dimethyl formamide (DMF) or dimethyl acetamide (DMA), 3 d, 100°C; isolated yields are given
Entry
Pd-cat
Variations
Solvent
Yield (%)
3
4
5
1
2
3
4
5
6
7
8
9
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd on C
1
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMA
90
47
25
1
64
–
57
–
61
57
72
–
3
32
15
7
–
7
–
–
Without n-Bu4NBr
+5 vol-% H2O
+10 vol-% H2O
+5 mol-% phenanthroline
+10 mol-% phenanthroline
+10 mol-% PPh3
16
17
–
12
–
–
–
27
9
–
+50 mol-% PPh3
10
28
17
10
11
1
Table 2
Acknowledgements
Pd-catalyzed Ullmann coupling reaction of 6; b: 2 mmol of 6, 5
mol-% Pd(OAc)2 (or 5 mol-% Pd on carbon or 2 mol-% 1), 2 mmol
n-Bu4NBr, 8 mmol K2CO3, 10 ml DMF or DMA, 4-15 h, 135°C;
isolated yields are given
Financial support of the Fonds der Chemischen In-
dustrie and a donation of palladium acetate by the
Degussa AG is gratefully acknowledged.
Entry Pd-cat
Variations (h)
Solvent
Yield [%]: 7
12
13
14
15
Pd(OAc)2 15 h
DMF
DMF
DMF
DMA
80
60
67
47
Pd on C
15 h
15 h
4 h
References
1
1
[1] (a) G. Dyker, Angew. Chem. 104 (1992) 1079–1081; Angew.
Chem. Int. Ed. Engl. 31 (1992) 1023. (b) G. Dyker, J. Org.
Chem. 58 (1993) 6426. (c) G. Dyker, Chem. Ber. 127 (1994) 739.
(d) G. Dyker, Angew. Chem. 106 (1994) 117; Angew. Chem. Int.
Ed. Engl. 33 (1994) 103.
[2] (a) A.J. Canty, Acc. Chem. Res. 25 (1992) 83. (b) A.J. Canty,
Platinum Metals Rev. 37 (1993) 2.
[3] T. Jeffery, Tetrahedron Lett. 26 (1985) 2667.
[4] (a) O. Reiser, M. Weber, A. de Meijere, Angew. Chem. 101
(1989) 1071; Angew. Chem. Int. Ed. Engl. 28 (1989) 1037. (b) K.
Albrecht, O. Reiser, M. Weber, A. de Meijere, Tetrahedron 50
(1994) 383. (c) O. Reiser, B. Ko¨nig, K. Meerholz, J. Heinze, T.
Wellauer, F. Gerson, R. Frim, M. Rabinovitz, A. de Meijere, J.
Am. Chem. Soc. 115 (1993) 3511.
the typical darkening of the reaction mixture did not
appear in these two cases. Most interestingly the change
of the solvent from DMF to DMA had a significant
influence on the product distribution. Whereas with
DMF the 1:2-product 12 is slightly favoured, DMA
leads mainly to the 2:1-product 10. Clearly the choice
of the solvent for palladium catalyzed processes de-
serves more attention, because its influence on the
selectivity can surpass that of additional phosphane or
other ligands.
Table 3
Annelation reaction of 8 with 9; b: 4 mmol of 8, 1 mmol of 9, 5 mol-% of Pd(OAc)2 and Pd(PPh3)4 (or 15 mol-% Pd on carbon or 2 mol-% 1),
2 mmol n-Bu4NBr, 8 mmol K2CO3, 10 ml DMF or DMA, 3 d, 100°C; d: triphenylethene was found as a byproduct in this case with 16% yield
Entry
Pd-cat
Variations
Solvent
Yield [%]
10
11
12
21
4
35
4
B3
10
42
5
16
17
18d
19
20
21
22
23
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(PPh3)4
Pd on C
1
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMA
77
51
14
86
61
37
35
67
4
2
2
without n-Bu4NBr
+20 vol-% H2O
+10 mol-% PPh3
3
33
12
12
25
1

144
G. Dyker, A. Kellner / Journal of Organometallic Chemistry 555 (1998) 141–144
8
[5] (a) W.A. Herrmann, C. Broßmer, K. Ofele, C.-P. Reisinger, M.
Beller, H. Fischer, Angew. Chem. 107 (1995) 1989; Angew. Chem.
Int. Ed. Engl. 34 (1995) 1844. (b) M. Beller, H. Fischer, W.A.
[10] (a) M.T. Reetz, G. Lohmer, J. Chem. Soc. Chem. Commun. (1996)
1921. (b) H. Bo¨nnemann, W. Brijoux, Catalytically active metal
powders and colloids, in: A. Fu¨rstner (Ed.), Active Metals, VCH,
Weinheim, 1996, pp. S351. (c) M. Beller, H. Fischer, K. Ku¨hlein,
C.-P. Reisinger, W.A. Herrmann, J. Organomet. Chem. 520 (1996)
257.
[11] G. Dyker, J. Org. Chem. 58 (1993) 234.
[12] (a) G. Dyker, A. Kellner, Tetrahedron Lett. 35 (1994) 7633. (b)
G. Dyker, Chem. Ber./Recueil 130 (1997) 1567.
[13] (a) F. Ozawa, T. Hidaka, T. Yamamoto, A. Yamamoto, J.
Organomet. Chem. 330 (1987) 253. (b) F. Ozawa, M. Fujimori,
T. Yamamoto, A. Yamamoto, Organometallics 5 (1986) 2144.
[14] This result does not allow a comparison to the performance of 1
in the Heck reaction of bromo- and chloroarenes under different
reaction conditions.
8
Herrmann, K. Ofele, C. Broßmer, Angew. Chem. 107 (1995) 1992;
Angew. Chem. Int. Ed. Engl. 34 (1995) 1846. (c) W.A. Herrmann,
C. Broßmer, K. Ofele, M. Beller, H. Fischer, J. Organomet. Chem.
491 (1995) C1. (d) W.A. Herrmann, C. Brossmer, C.-P. Reisinger,
8
8
T.H. Riermeier, K. Ofele, M. Beller, Chem. Eur. J. 3 (1997) 1357.
[6] H.-U. Blaser, A. Spencer, J. Organomet. Chem. 258 (1983) 101.
[7] T. Mitsudo, W. Fischetti, R.F. Heck, J. Org. Chem. 49 (1984)
1640.
[8] (a) J. Louie, J.F. Hartwig, Angew. Chem. 108 (1996) 2531; Angew.
Chem. Int. Ed. Engl. 35 (1996) 2359; (b) J.F. Hartwig, Synlett.
(1997) 329.
[9] 4 is a hydrolysis product and 5 a fragmentation product of an
intermediary Ar-O-CH2-Pd-I complex.
.