Phosphapalladacycle-catalyzed heck reactions for efficient synthesis of trisubstituted olefins: Evidence for palladium(0) intermediates

Article abstract of DOI:10.1002/(SICI)1099-0682(199801)1998:1<29::AID-EJIC29>3.0.CO;2-X

The coupling reaction of 1,1-disubstituted olefins (α-methyl-styrene, n-butyl methacrylate) with various aryl bromides (Heck reaction) has been studied as a new concept to synthesize trisubstituted olefins. Surprisingly, the nature of the base dramatically influences the product distribution. Thus, a systematic investigation on the role of base in Heck reactions of 1,1-disubstituted olefins was performed. Less coordinating bases like NaOAc, NaOBz or Na₂CO₃ yield a statistical distribution of regioisomers with the terminal olefin 10 as the major product. However, by using amines like Bu₃N or diisopropylethylamine (DIPEA) as base internal olefins can be synthesized with high selectivities. With phosphapalladacycle 3 as catalyst precursor, we were able to obtain catalyst turnover numbers up to 1000, while Pd(OAc)₂/2PPh₃ was one order of magnitude less active. Analysis of the reaction profile by kinetic investigations led to the postulation of a reduction and subsequent oxidative addition of the catalyst precursor 3 to form 12 as catalytically active intermediate.

Full text of DOI:10.1002/(SICI)1099-0682(199801)1998:1<29::AID-EJIC29>3.0.CO;2-X

FULL PAPER
Palladium-Catalyzed Reactions for Fine Chemical Synthesis, 4[᭛]
Phosphapalladacycle-Catalyzed Heck Reactions for Efficient Synthesis of
Trisubstituted Olefins: Evidence for Palladium(0) Intermediates
Matthias Beller* and Thomas H. Riermeier
Anorganisch-chemisches Institut der Technische Universität München,
Lichtenbergstraße 4, D-85747 Garching, Germany
Fax: (internat.) ϩ 49(0)89/289-13743
Received July 9, 1997
Keywords: Catalysis / Palladium / CϪC coupling / Metallacycles / Trisubstituted olefins
The coupling reaction of 1,1-disubstituted olefins (α-methyl-
styrene, n-butyl methacrylate) with various aryl bromides
jor product. However, by using amines like Bu3N or diisopro-
pylethylamine (DIPEA) as base internal olefins can be syn-
(Heck reaction) has been studied as a new concept to synthe- thesized with high selectivities. With phosphapalladacycle 3
size trisubstituted olefins. Surprisingly, the nature of the base as catalyst precursor, we were able to obtain catalyst turno-
dramatically influences the product distribution. Thus, a sys- ver numbers up to 1000, while Pd(OAc)2/2PPh3 was one or-
tematic investigation on the role of base in Heck reactions
of 1,1-disubstituted olefins was performed. Less coordinating
bases like NaOAc, NaOBz or Na2CO3 yield a statistical distri-
der of magnitude less active. Analysis of the reaction profile
by kinetic investigations led to the postulation of a reduction
and subsequent oxidative addition of the catalyst precursor
bution of regioisomers with the terminal olefin 10 as the ma- 3 to form 12 as catalytically active intermediate.
see ref.[6]). Recently, we have developed cyclometallated pal-
ladium complexes as a new type of efficient catalysts for
Heck and related reactions.[7][8] Obviously, olefins A or B
should be available by Heck olefination. However, the low
regioselectivity in the formation of the CϭC double bond
using disubstituted as well as aliphatic olefins is still a
major disadvantage for the application of Heck reactions.
The low regioselectivity can be explained either by unselec-
tive β-hydride elimination during the catalytic cycle, and/or
by the tendency of the eliminated HPdX species to undergo
readdition and subsequent β-hydride elimination to differ-
ent hydrogen atoms, thus resulting in isomerization reac-
tions.
Introduction
Trisubstituted, arylated olefins of the type A or B display
interesting pharmacological or physical properties.[2]
Figure 1. Trisubstituted aromatic olefins
As a concept for controlling regioselective olefin forma-
tion Tietze et al. have shown that silyl groups in allylsilanes
act as terminating group in intramolecular Heck reactions
enabling regio- and enantioselective construction of triple-
substituted double bonds.[9] More recently, Ricci et al. re-
ported a stereo- and regiocontrolled synthesis of di- and
trisubstituted olefins starting from vinylsilanes by a Heck
reaction and Pd-catalyzed desilylation reaction sequence.[10]
Despite the high selectivities for specific double-bond iso-
mers using silyl-substituted olefins it is obvious that control
of the double-bond formation in Heck reactions using sim-
ple disubstituted alkenes by other means is a very attrac-
tive goal.
In this respect derivatives of α-methylcinnamic acids dis-
play hypolipidemic activities.[2a] Further, the substructure is
found in active substances such as the antibiotic hygromy-
cin A.[2b] In addition, certain α-methylstilbenes can be used
as nematic liquids[2c] with interesting retinoidal proper-
ties.[2d] Despite of their broad applications the stereodefined
synthesis of trisubstituted alkenes is still a challenge of
modern synthetic organic chemistry. Although a number of
methodologies have been reported in the literature[3] the
problems in synthesizing highly substituted alkenes, e.g. by
Wittig reaction are well documented.[4] Thus, the search for
alternative methodologies is ongoing.[5]
In this respect, the palladium-catalyzed Heck reaction is
a powerful tool for the construction of aryl- and vinyl-sub-
stituted carbonϪcarbon double bonds as it does not inter-
fere with a broad range of functional groups (for reviews
Results and Discussion
The interesting properties of α-methylcinnamic acid de-
rivatives[2a][2b][11] prompted us to study regioselective Heck
reactions of various aryl bromides 1 with n-butyl methacry-
[᭛]
Part 3: Ref.[1]
.
Eur. J. Inorg. Chem. 1998, 29Ϫ35
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ1948/98/0101Ϫ0029 $ 17.50ϩ.50/0
29

M. Beller, T. H. Riermeier
FULL PAPER
late (2) as model reaction.[12] Clearly, after insertion of the about two orders of magnitude higher compared to pre-
olefin into the ArϪPdϪBr complex there are two possibi- vious data of similar reactions.[14]
lities for β-hydride elimination: it can proceed towards the
In order to distinguish between thermodynamic or kine-
methyl or the benzylic methylene group. The former reac- tic control of the double-bond formation the terminal olefin
tion leads to α-benzylacrylates 5 with a terminal double 5 was isolated and treated with NaOAc and Bu3N under
bond, the latter to α-methylcinnamic acid esters 4 with an reaction conditions in the presence of the palladium cata-
internal double bond. Because of the reactivity of terminal lyst. Only a negligable amount of isomerization of 5 to the
olefins 5 the doubly arylated product 6 can subsequently be more stable internal olefin 4 was detected. Thus, we assume
formed. In addition, E and Z isomers of 4 and 6 could be that the formation of the double bond occurs under kine-
formed. Due to the importance of the β-hydride elimination tic control.
for determining the double-bond selectivity we anticipitated
Comparison of the phosphapalladacycle 3 with conven-
that the nature of the base will be of major influence for the tional palladium phosphane complexes which were tested
product distribution. Thus, various bases were examined for as in situ catalysts shows the superiority of 3. Significantly
the reaction of n-butyl methacrylate with 1-bromo-4-fluoro- lower conversions (16% and < 5%) were obtained in the
benzene using the phosphapalladacycle catalyst 3 {trans- presence of tributylamine as base and 0.1 mol% Pd(OAc)2/
di(µ-acetato)bis[o-(di-o-tolylphosphanyl)benzyl]dipal- PPh3 (1:2) or Pd(OAc)2/P(o-tolyl)3 (1:2), respectively.
ladium(II)[13]}. The results are sumarized in Table 1.
In agreement with previous results[6] the electronic nature
of the aryl bromide has dominating effect on the yield of
the reaction (runs 5Ϫ10). Here, electron-withdrawing sub-
stituents enhance the productivity. It is noteworthy that, in
all cases where amine is used as base, the ratio of internal
olefin to terminal and doubly arylated olefin is approxi-
mately 80:20, while with NaOAc as base the ratio is about
40:60. Therefore, we conclude that the electronic nature of
the aryl bromide has just a minor effect on the selectivity.
As a suitable candidate for further evaluation of the in-
fluence of base on the regioselectivity of Heck reactions
with 1,1-disubstituted olefins we considered α-methylstyr-
ene (8) which is available on large scale.[15] Here, the olefin-
ation of 1-bromo-4-chlorobenzene (7) was performed in the
presence of 8 different bases. As shown in Scheme 1 five
different Heck products can in principle arise from the reac-
tion of 7 and 8 (both E and Z isomers are possible).
Table 1. Heck reaction of n-butyl methacrylate with aryl bromides
To suppress formation of doubly arylated Heck products
an excess of α-methylstyrene 8 (5 equiv.) was used. The re-
sults of the model reaction are summarized in Table 2.
Using the phosphapalladacycle trans-di(µ-acetato)bis[o-
(di-o-tolylphosphanyl)benzyl]dipalladium(II) (3) as catalyst
under standard reaction conditions (dimethylacetamide,
NaOAc, 0.1 mol% 3, 135Ϫ140°C, 24 h; run 1) the coupling
products were isolated in excellent yield (94%). Analogous
to the reactions with n-butyl methacrylate the highest selec-
Reaction: 15 mmol of aryl bromide with 22.5 mmol of butyl meth-
acrylate in the presence of 18 mmol of base and 0.1 or 0.01 mol%
of phosphapalladacycle catalyst
3 in 15 ml of DMAc at
[a]
135Ϫ140°C. Ϫ Conversions of the aryl bromides have been de-
termined by GC with internal standard (diethylene glycol dibutyl
ether). Ϫ [b] Selectivities have been determined by GC on the basis
of area percentage.
Indeed, the nature of base significantly influences the re- tivity for the terminal olefin 10 is observed with NaOAc (9/
gioselectivity of the double bond. Sodium acetate favors the 10, 35:65). Only a small amount of the doubly arylated
formation of terminal olefin 5, thus leading to a relatively product 11 is formed (4% yield). Comparison of the phos-
large extent of doubly arylated products 6 (Table 1, run 1). phapalladacycle 3 with a classical catalyst system [0.1 mol%
In contrast, in the presence of 1.2 equiv. of tributylamine Pd(OAc)2/2 PPh3] gave similar results (run 10). The same
the internal olefin 4 is formed as major product (runs 2 and regioselectivity (9/10, 37:63), albeit a lower total yield (66%)
3). Further improvement of selectivity towards the internal was obtained with sodium benzoate as base (run 2). Using
olefin 4 was achieved using the sterically hindered base inorganic salts such as sodium carbonate (run 3) or sodium
N,N-diisopropylethylamin (DIPEA) which gives so far the hydrogen carbonate (run 4) excellent total yields (97% and
best selectivities for this reaction (internal olefin/terminal 95%) were obtained, but the selectivity towards the ter-
products, 82:18, run 4). Interestingly, the productivity of the minal product is slightly lower (9/10, 38:62 and 46:54,
catalyst system is reduced by a factor of 3 in the presence respectively). On the other hand high selectivities up to 95:5
of amines compared to NaOAc (0.01 mol% Pd, runs 1 and towards the internal olefin 9 are observed using tertiary
2) which is explained by enhanced stabilization of PdII in- amines as bases. Again, best results were obtained using the
termediates. While the selectivity in the presence of NaOAc sterically hindered amine N,N-diisopropylethylamine
is only moderate the catalyst productivity (TON ϭ 8300) is (DIPEA, run 9, 65% total yield). A slightly lower regiose-
30
Eur. J. Inorg. Chem. 1998, 29Ϫ35

Palladium-Catalyzed Reactions for Fine Chemical Synthesis, 4
FULL PAPER
Scheme 1. Heck reaction of 1-bromo-4-chlorobenzene with α-methylstyrene
Table 2. Heck reaction of α-methylstyrene with aryl bromides
Scheme 2. Regiochemistry of the β-hydride elimination
Reaction: 15 mmol of 4-bromo-1-chlorobenzene with 75 mmol of
α-methylstyrene in the presence of 18 mmol of base and 0.1 mol%
Next we investigated whether an isomerization of the ter-
minal olefin 10 to the internal olefin 9 in the presence of
amine is possible. Neither heating of the terminal olefin 10
with catalyst 3 and DIPEA, nor heating of the internal ole-
fin 9 with 3 and sodium acetate results in significant iso-
merization (< 10%). In addition, a thermodynamic iso-
merization would not explain the observed amount of the
Z isomer Z-9. Thus, we propose a kinetic control for the
of phosphapalladacycle catalyst
3 in 15 ml of DMAc at
135Ϫ140°C. After 24 h, the reaction mixture was extracted and the
solvent and not converted starting materials were removed in
vacuo. The remaining reaction mixture was analyzed by GC, GC/
[a]
[b]
MS, and 1H NMR. Ϫ
Isolated yields. Ϫ
Selectivities have
[c]
been determined by NMR of the isolated products. Ϫ TON ϭ
mol product/mol catalyst. Ϫ
[d]
1,8-Bis(dimethylamino)naphtha-
lene.
lectivity (9/10, 89:11) is observed using tributylamine as reaction. In order to explain the different regio- and stereo-
base (66% total yield). Other bases or mixtures of bases selectivities (Scheme 2) we propose different mechanism for
gave results inbetween sodium acetate and DIPEA. No re- the proton abstraction (β-hydride elimination) depending
action is observed using chelating amines like TMEDA.
on the base. Unfortunately, the β-hydride elimination and
In the presence of amine as base the phosphapalladacycle reductive elimination of HX from “HϪPdϪX” have been
catalyst 3 was found to be superior in productivity com- scarcely investigated in the literature.[16] Based on simple
pared to classical catalyst precursors. Thus, the use of 0.1 conformation analysis the stereochemical outcome of the
mol% Pd(OAc)2/2 PPh3 led to a significantly lower total products is explained by β-hydride elimination from the
yields of 9 and 10 (14%, run 11). The selectivity towards possible intermediates I1 (leading to 10), I2 (leading to E-9
the internal olefin is similar to the phosphapalladacycle and 10) and I3 (leading to Z-9 and 10) (Scheme 3).
catalyst 3 (9/10, 93:7), but there is a clear difference in the
E/Z selectivity (E/Z ϭ 12).
In the presence of less basic salts like NaOAc, NaOBz
direct elimination of “HϪPdϪX” will occur from inter-
Eur. J. Inorg. Chem. 1998, 29Ϫ35
31

M. Beller, T. H. Riermeier
FULL PAPER
Scheme 3. Conformational analysis of the β-hydride elimination
Figure 2. Concentration vs. reaction time diagrams; top: NaOAc
as base; bottom: DIPEA as base; ᭹ terminal olefin 10; ᭿ Z-9;
᭡ E-9[a]
mediates I1, I2 and I3. Because of the nearly statistical
product ratio (terminal/internal, 3:2) we assume that the
less coordinating bases are not involved in the elimination
step of “HϪPdϪX”. However, the amine-assisted elimi-
nation takes place towards the most acidic protons
(Scheme 2). A direct abstraction of the more acidic benzylic
protons would explain for the low selectivity for the E iso-
mer E-9 because of a similar acidity of both benzylic pro-
tons. Thus, we believe that the base is involved in the
β-hydride elimination by proton abstraction. Support for
the direct abstraction of the proton may also result from
the fact that the amine is blocking coordination sites on the
metal centre, which are esssential for β-hydride elimination.
To obtain further mechanistic information about the sys-
tem concentration vs. time diagrams for the reaction of 1-
bromo-4-chlorobenzene (7) with α-methylstyrene (8) in the
presence of NaOAc and DIPEA were compared (Figure 2).
As shown in Figure 2 the base influences the reaction
profile dramatically. In the presence of NaOAc the reaction
starts immediately leading to the terminal olefin 10 as the
major isomer. If the concentration of 10 is high enough
(after ca. 5 h) the doubly arylated product 11 (not shown
in Figure 2) is produced in small quantities by a consecutive
Heck reaction. No initial catalyst activation is observed.
However, the reaction with DIPEA as base (Figure 2, lower
diagram) shows a sigmoidal concentration vs. time diagram
indicating preformation of the active catalyst species. The
induction period might be explained by a slowly occurring
reduction of the phosphapalladacycle 3[17] to a monophos-
phanepalladium(0) compound. Immediately, this species
would undergo oxidative addition to 1-bromo-4-chloroben-
zene (7) to give the palladium complex 12 (Scheme 4). To
proof this assumption the oxidative addition product 12
was prepared independently from Pd[P(o-tolyl)3]2Cl2 and 7
in a total yield of 66% according to a procedure described
by J. F. Hartwig et al.[18]
[a]
Reaction conditions: 15 mmol of 1-bromo-4-chlorobenzene, 75
mmol of α-methylstyrene, 18 mmol of base and 2 ml of diethylene
glycol dibutyl ether were heated in 15 ml of DMAc at 135°C. At
t ϭ 0 a solution of 0.1 mol% of the phosphapalladacycle 3, dissol-
ved in 1 ml of DMAc, was added. The course of the reaction was
monitored by removing 0.1-ml samples of the reaction mixture.
These were washed with 5% HCl solution, then extracted with 1.5
ml of dichloromethane and analyzed by GC.
Scheme 4. Postulated formation of 12 from the phosphapallada-
cycle 3 (R ϭ o-tolyl)
The concentration vs. time diagrams for the Heck reac-
tion of 1-bromo-4-chlorobenzene (7) with α-methylstyrene
(8) in the presence of 0.1 mol% 12 as catalyst with NaOAc
and DIPEA as base are shown in Figure 3.
No induction period is observed in both cases! In the
presence of NaOAc as base the reaction catalyzed by 12
starts faster compared to the reaction catalyzed by 3 and a
similar product distribution is observed. Also using DIPEA
as base the raction starts immediately. The ratio of products
initially formed is comparable to the ratio of the different
isomers obtained in the phosphapalladacycle-catalyzed re-
action. However, after several hours E-9 is formed preferen- phosphanepalladium(0) species. The nature of the reducing
tially with 12 as catalyst yielding in a higher E/Z ratio. agent is not clear in detail. Coordination of amines stabi-
This findings are rationalized as follows: As shown in lizes the phosphapalladacycle 3 slowing down the formation
Scheme 4 the phosphapalladacycle 3 is reduced to a mono- of a palladium(0) compound. This results in an induction
32
Eur. J. Inorg. Chem. 1998, 29Ϫ35

Palladium-Catalyzed Reactions for Fine Chemical Synthesis, 4
FULL PAPER
Figure 3. Concentration vs. reaction time diagrams; top: NaOAc bromo-4-fluorobenzene behave in a similar manner as de-
as base; bottom: DIPEA as base; ᭹ terminal olefin 10; ᭿ Z-9;
scribed for 1-bromo-4-chlorobenzene. In contrast to Table
1 the coupling reactions with n-butyl methacrylate were
᭡ E-9[a]
performed in the presence of five equivalents of the olefin.
Thus, the amount doubly arylated products could be mini-
mized.
Table 3: Heck reaction of 1,1-disubstituted olefins with various aryl
bromides
Reaction: 15 mmol of aryl halide with 75 mmol of olefin in the
presence of 18 mmol of base and 0.1 mol% of phosphapalladacycle
[a]
catalyst 3 in 15 ml of DMAc at 135Ϫ140°C (24 h). Ϫ
Total
isolated yields. Ϫ [b] Selectivities have been determined by NMR of
the isolated products. Ϫ TON ϭ mol product/mol catalyst. Ϫ
[c]
[d]
Determined by GC analysis.
In conclusion we have demonstrated that the nature of
base significantly influences the regiochemistry of the
double bond in certain types of Heck reactions. Based on
this finding we have developed a new and general valid con-
cept to determine the regiochemistry of Heck reactions of
1,1-disubstituted olefins simply by changing the added base.
Moreover, by using phosphapalladacycle 3 as catalyst pre-
cursor we have been able to improve the catalyst pro-
ductivity for Heck coupling reactions of 1,1-disubstituted
olefins by at least one order of magnitude. Comparison of
the concentration vs. time diagrams for Heck reactions with
the phosphapalladacycle 3 and the oxidative addition prod-
uct 12 as catalysts provides evidence that 3 is a catalyst pre-
cursor with 12 as active species.
[a]
Reaction conditions: 15 mmol of 1-bromo-4-chlorobenzene, 75
mmol of α-methylstyrene, 18 mmol of base and 2 ml of diethylene
glycol dibutyl ether were heated in 15 ml of DMAc at 135°C. At
t ϭ 0 a solution of 0.1 mol% of 12, dissolved in 1 ml of DMAc,
was added. The course of the reaction was monitored by removing
0.1-ml samples of the reaction mixture. These were washed with
5% HCl solution, then extracted with 1.5 ml of dichloromethane
and analyzed by GC.
We gratefully acknowledge support from the Hoechst AG and the
“Bayerischer Forschungsverbund Katalyse” (FORKAT). Prof. W. A.
Herrmann, C.-P. Reisinger, A. Zapf (TU München), Prof. K.
Kühlein, Dr. H. Geißler and Dr. R. W. Fischer (Hoechst AG) are
thanked for many stimulating discussions.
period for the proper catalyst. Furthermore, the rate of oxi-
dative addition is decreased in the presence of amines by
blocking of free coordination sites on the Pd0 species. On
the other hand it is known, that the coordination of NaOAc
to phosphanepalladium(0) complexes increases the rate of
oxidative addition for haloarenes.[19] This explains in gen-
eral the higher catalyst activity in the presence of NaOAc.
Further evidence for the discussed transformation of the
phosphapalladacycle 3 to 12 is obtained by comparing the
E/Z ratio for the different catalysts (3 and 12) using DIPEA
as base. As a matter of fact the E/Z ratio produced by 12
in the first hour is identical to that of the phosphapallada-
cycle 3 during the whole reaction time. Due to side-reac-
tions in the formation of 12 from 3 the productivity of 3
is lower.
Experimental Section
Chemicals were obtained from Aldrich and Fluka. NMR spectra
(1H,13C,31P) were recorded with a Jeol JMX-GX 400 or a Bruker
AM 360 instrument. GC-MS spectra were measued with a Hewlett
Packard gas chromatograph GC 5890 A equipped with a mass-
selective detector MS 5970 B. Elemental analyses were carried out
by the Microanalytical Laboratory at the TU München. Quantitat-
ive analyses were performed with a Hewlett Packard 6890 instru-
ment using a HP-5 capillary column in conjunction with a flame
ionization detector (GC/FID). Ϫ Except for the work-up of reac-
tion mixtures, all operations were carried out under nitrogen. Sol-
vents were carefully dried according to standard procedures.
In order to demonstrate that the determination of the
regiochemistry by the base is a general principle for Heck
reactions of 1,1-disubstituted olefins we studied coupling
reactions of 2 and 8 with various other aryl bromides (Table
1. Heck Reactions with n-Butyl Methacrylate: 15 mmol of aryl
halide, 22.5 mmol of n-butyl methacrylate (3.58 ml), 18 mmol of
3). As can be seen from Table 3 4-bromotoluene and 1- tri-n-butylamine (4.29 ml), 0.1 mg of 2,6-di-tert-butyl-4-methyl-
Eur. J. Inorg. Chem. 1998, 29Ϫ35 33

M. Beller, T. H. Riermeier
FULL PAPER
phenol and 7.0 mg of the phosphapalladacycle 3 were dissolved in The solvent and starting materials were removed in vacuo. The ra-
15 ml of N,N-dimethylacetamide and stirred at 135Ϫ140°C (reac- tio of the regioisomers was calculated by the relative intensities of
1
tion time as indicated). After cooling, the reaction mixture was the typical signals in the H-NMR spectrum (CDCl3, TMS as in-
extracted with diethyl ether and 5% aqueous HCl for three times. ternal standard). Total yields for different bases are given in Tables
The combined organic phases were neutralized with 10% aqueous
NaHCO3 and dried with MgSO4. After removal of the solvent and
starting materials in vacuo the products were isolated as colourless
oils by distillation or by column chromatography.
2 and 3.
(E)-4-Chloro-Ͱ-methylstilbene (E-4): 1H NMR (360 MHz, 20°C,
CDCl3): δ ϭ 7.5Ϫ7.0 (m, 9 H, aromatic protons), 6.76 (s, 1 H, Cϭ
CH), 2.25 (s, 3 H, ϪCH3). Ϫ MS (70 eV); m/z: 228, [Mϩ], 211,
193, 178.
(Z)-4-Chloro-Ͱ-methylstilbene (Z-4): 1H NMR (360 MHz, 20°C,
CDCl3): δ ϭ 7.5Ϫ7.0 (m, 9 H, aromatic protons), 6.40 (s, 1 H, Cϭ
CH), 2.19 (s, 3 H, ϪCH3). Ϫ MS (70 eV); m/z: 228, [Mϩ], 211,
193, 178.
3-(4-Chlorophenyl)-2-phenyl-1-propene (5): 1H NMR (360 MHz,
20°C, CDCl3): δ ϭ 7.5Ϫ7.0 (m, 9 H, aromatic protons), 5.48 (s, 1
H, CϭCH), 5.01 (s, 1 H, CϭCH), 3.79 (s, 2 H, ϪCH2Ar). Ϫ MS
(70 eV); m/z: 228, [Mϩ], 193, 103.
n-Butyl 3-(4-Fluorophenyl)-2-methylpropenoate: Isolated by dis-
tillation (60%, b.p. 95Ϫ100°C/1 mbar). Ϫ 1H NMR (360 MHz,
3
20°C, CDCl3): δ ϭ 7.64 (s, 1 H, CHϭC), 7.36 (dd, 2 H, JHH
ϭ
4
3
3
8.6 Hz, JFH ϭ 5.5 Hz), 7.06 (dd, JHH ϭ 8.6 Hz, JFH ϭ 8.6 Hz,
2 H), 4.21 (t, JHH ϭ 6.6 Hz, 2 H, COOCH2), 2.10 (s, 3 H, ϭ
CϪCH3), 1.71 (tt, JHH ϭ 6.7 Hz, JHH ϭ 6.7 Hz, 2 H, CO-
OCH2CH2) 1.46 (tt, JHH ϭ 6.7 Hz, JHH ϭ 7.3 Hz, 2 H, CO-
OCH2CH2CH2), 0.97 (t, 3JHH ϭ 7.3 Hz, 3 H, CH2CH3). Ϫ MS (70
eV); m/z: 236 [Mϩ], 180, 134, 109.
3
3
3
3
3
n-Butyl 3-(4-Acetylphenyl)-2-methylpropenoate: Isolated by col-
umn chromatography (ethyl acetate/hexane, 1:8, 74%). Ϫ 1H NMR
(360 MHz, 20°C, CDCl3): δ ϭ 7.97 (d, 2 H, 3JHH ϭ 8.3 Hz, 3-Ph-
H, 5-Ph-H), 7.67 (s, 1 H, CHϭC), 7.46 (d, 2 H, 3JHH ϭ 8.3 Hz, 2-
(E)-4-Fluoro-Ͱ-methylstilbene: 1H NMR (360 MHz, 20°C,
CDCl3): δ ϭ 7.5Ϫ7.0 (m, 9 H, aromatic protons), 6.78 (s, 1 H, Cϭ
CH), 2.24 (s, 3 H, ϪCH3). Ϫ MS (70 eV); m/z: 212, [Mϩ], 197,
177, 133.
(Z)-4-Fluoro-Ͱ-methylstilbene: 1H NMR (360 MHz, 20°C,
CDCl3): δ ϭ 7.5Ϫ7.0 (m, 10 H, aromatic and olefinic protons),
2.18 (s, 3 H, ϪCH3). Ϫ MS (70 eV); m/z: 212, [Mϩ], 197, 177, 133.
3-(4-Fluorophenyl)-2-phenyl-1-propene: 1H NMR (360 MHz,
20°C, CDCl3): δ ϭ 7.5Ϫ7.0 (m, 9 H, aromatic protons), 5.48 (s, 1
H, CϭCH), 5.01 (s, 1 H, CϭCH), 3.79 (s, 2 H, ϪCH2Ar). Ϫ MS
(70 eV); m/z: 212, [Mϩ],197, 134, 103.
(E)-4-Methyl-Ͱ-methylstilbene: 1H NMR (360 MHz, 20°C,
CDCl3): δ ϭ 7.5Ϫ6.8 (m, 9 H, aromatic protons), 6.43 (s, 1 H, Cϭ
CH), 2.36 (s, 3 H, Ar-CH3), 2.28 (s, 3 H, -CH3). Ϫ MS (70 eV);
m/z: 208, [Mϩ], 193, 178, 115.
(Z)-Ͱ,4-Dimethylstilbene: 1H NMR (360 MHz, 20°C, CDCl3):
δ ϭ 7.5Ϫ6.8 (m, 10 H, aromatic and olefinic protons), 2.18 (s, 3
H, Ar-CH3), 2.22 (s, 3 H, ϪCH3). Ϫ MS (70 eV); m/z: 208, [Mϩ],
193, 178, 115.
3-(4-Methylphenyl)-2-phenyl-1-propene: 1H NMR (360 MHz,
20°C, CDCl3): δ ϭ 7.5Ϫ6.8 (m, 9 H, aromatic protons), 5.44 (s, 1
H, CϭCH), 4.77 (s, 1 H, CϭCH), 3.74 (s, 2 H, ϪCH2Ar), 2.38
(s, 3 H, Ar-CH3). Ϫ MS (70 eV); m/z: 208, [Mϩ], 193, 178, 130,
115, 103.
3
Ph-H, 6-Ph-H), 4.23 (t, JHH ϭ 6.6 Hz, 2 H, COOCH2), 2.60 (s, 3
3
H, COCH3), 2.12 (s, 3 H, ϭCϪCH3), 1.73 (tt, JHH ϭ 6.6 Hz,
3
3JHH ϭ 6.6 Hz, 2 H, COOCH2CH2) 1.47 (tt, JHH ϭ 6.7 Hz,
3
3JHH ϭ 7.4 Hz, 2 H, COOCH2CH2CH2), 0.98 (t, JHH ϭ 7.4 Hz,
3 H, CH2CH3). Ϫ MS (70 eV); m/z: 260 [Mϩ], 245, 204, 189, 161,
145, 115.
n-Butyl 3-(4-Methoxyphenyl)-2-methylpropenoate: Isolated by
column chromatography (ethyl acetate/hexane, 1:10, 20%). Ϫ 1H
NMR (360 MHz, 20°C, CDCl3): δ ϭ 7.64 (s, 1 H, CHϭC), 7.37
3
3
(d, 2 H, JHH ϭ 8.3 Hz, 2-Ph-H, 6-Ph-H), 6.91 (d, 2 H, JHH
ϭ
3
8.3 Hz, 3-Ph-H, 5-Ph-H), 4.18 (t, JHH ϭ 6.6 Hz, 2 H, COOCH2),
3.82 (s, 3 H, OCH3), 2.13 (s, 3 H, ϭCϪCH3), 1.65 (tt, JHH ϭ 6.6
3
3
3
Hz, JHH ϭ 6.6 Hz, 2 H, COOCH2CH2) 1.44 (tt, JHH ϭ 6.7 Hz,
3JHH ϭ 7.4 Hz, 2 H, COOCH2CH2CH2), 0.97 (t, JHH ϭ 7.4 Hz,
3
3 H, CH2CH3). Ϫ MS (70 eV); m/z: 248 [Mϩ], 192, 146, 115, 91.
n-Butyl 3-(4-Chlorophenyl)-2-methylpropenoate: Isolated by dis-
tillation (55%, b.p. 120Ϫ125°C/1 mbar). Ϫ 1H NMR (360 MHz,
20°C, CDCl3): δ ϭ 7.63 (s, 1 H, CHϭC), 7.38Ϫ7.32 (m, 4 H), 4.23
3
(t, JHH ϭ 6.6 Hz, 2 H, COOCH2), 2.10 (s, 3 H, ϭCϪCH3), 1.74
3
3
(tt, JHH ϭ 6.6 Hz, JHH ϭ 6.6 Hz, 2 H, COOCH2CH2) 1.44 (tt,
3JHH ϭ 6.7 Hz, JHH ϭ 7.4 Hz, 2 H, COOCH2CH2CH2), 0.98 (t,
3
3JHH ϭ 7.4 Hz, 3 H, CH2CH3). Ϫ MS (70 eV); m/z: 252 [Mϩ], 196,
3. Synthesis of {Pd[P(o-tolyl)3](p-ClC6H4)(Br)}2 (12): 1.21 g
(1.54 mmol) of Pd[P(o-tolyl)3]2Cl2 and 0.54g (1.78 mmol) of P(o-
tolyl)3 were suspended in 8 ml of toluene. To this suspension NaOH
(0.14 g) in 8 ml of ethanol was added. The reaction mixture was
heated at 90°C for 5 h. After cooling, the yellow precipitate was
filtered off and washed several times with water, ethanol and diethyl
ether. After drying in vacuo, 1.06 g of crude Pd[P(o-tolyl)3]2 was
obtained. This product was stirred with 1.60 g of 1-bromo-4-chlo-
robenzene in 20 ml of benzene at room temp. overnight. The sus-
pension was filtered and concentrated. After addition of 20 ml of
diethyl ether and cooling to Ϫ30°C 0.60 g (66%) of 12 precipitated.
179, 150, 115, 89.
n-Butyl 3-(4-Nitrophenyl)-2-methylpropenoate: Isolated by col-
umn chromatography (ethyl acetate/hexane, 1:15, 75%). Ϫ 1H
NMR (360 MHz, 20°C, CDCl3): δ ϭ 8.25 (d, 2 H, 3JHH ϭ 8.9 Hz,
3
3-Ph-H, 5-Ph-H), 7.70 (s, 1 H, CHϭC), 7.54 (d, 2 H, JHH ϭ 8.9
Hz, 2-Ph-H, 6-Ph-H), 4.25 (t, 3JHH ϭ 6.6 Hz, 2 H, COOCH2), 2.12
3
3
(s, 3 H, ϭCϪCH3), 1.72 (tt, JHH ϭ 6.6 Hz, JHH ϭ 6.6 Hz, 2
3
3
H, COOCH2CH2) 1.47 (tt, JHH ϭ 6.7 Hz, JHH ϭ 7.4 Hz, 2 H,
COOCH2CH2CH2), 0.98 (t, 3JHH ϭ 7.4 Hz, 3 H, CH2CH3). Ϫ MS
(70 eV); m/z: 263, [Mϩ], 207, 190, 161, 115, 89.
2. Heck Reactions with α-Methylstyrene: 15 mmol of the aryl
Ϫ
1H NMR (400 MHz, 80°C, [D8]toluene): δ ϭ 7.03Ϫ6.45 (br.
bromide, 75 mmol of α-methylstyrene (8.86 g, 2), 18 mmol of the m, 32 H), 2,14 (br. s, 18 H). Ϫ 31P{1H} NMR (162 MHz, 80°C,
base and 7.0 mg of the phosphapalladacycle 3 (0.1 mol% Pd) were [D8]toluene): δ ϭ 26.8. Ϫ IR (KBr): ν˜ ϭ 3054 (m), 2970 (m), 2922
dissolved in 15 ml of N,N-dimethylacetamide and stirred under a
gentle stream of nitrogen at 135Ϫ140°C for 24 h. After cooling,
(m), 2861 (m), 1589 (m), 1566 (w), 1467 (vs), 1449 (s), 1200 (w),
1088 (s), 1004 (vs), 807 (vs), 752 (vs), 717 (s), 680 (m), 561 (m), 534
the reaction mixture was extracted with dichloromethane and 5% (s), 466 (s). Ϫ FAB MS; m/z: 520 [Mϩ Ϫ HBr], 430 [Mϩ Ϫ BrϪto-
aqueous HCl for three times. The combined organic phases were lyl]. Ϫ C54H50Br2Cl2P2Pd2 ·Et2O (1278.6): calcd. C 54.48, H 4.73,
neutralized with 10% aqueous NaHCO3 and dried with MgSO4.
P 4.84; found C 54.66, H 4.44, P 4.88.
34
Eur. J. Inorg. Chem. 1998, 29Ϫ35

Palladium-Catalyzed Reactions for Fine Chemical Synthesis, 4
FULL PAPER
[1]
[7]
[8]
M. Beller, T. H. Riermeier, A. Zapf, Top. Catal., in print.
W. A. Herrmann, C. Broßmer, K. Öfele, C.-P. Reisinger, T. Pri-
ermeier, M. Beller, H. Fischer, Angew. Chem. 1995, 107, 1989;
Angew. Chem. Int. Ed. Engl. 1995, 34, 1844.
[2] [2a]
T. Watanabe, K. Hayashi, S. Yoshimatsu, K. Sakai, J. Med.
[2b]
Chem. 1980, 23, 50. Ϫ
J. G. Buchanan, D. G. Hill, R. H.
[2c]
Wightman, Tetrahedron 1995, 51, 6033. Ϫ
W. R. Young, A.
M. Beller, H. Fischer, W. A. Herrmann, K. Öfele, C. Broßmer,
Angew. Chem. 1995, 107, 1992; Angew. Chem. Int. Ed. Engl.
1995, 34, 1848.
[2d]
Aviram, R. J. Cox, J. Am. Chem. Soc. 1972, 94, 3976. Ϫ
H.
Kagechika, T. Himi, K. Namikawa, E. Kawachi, Y. Hashimoto,
[9]
K. Shudo, J. Med. Chem. 1989, 32, 1098.
R. Schimpf, L. F. Tietze, Angew. Chem. 1994, 106, 1138; Angew.
Chem. Int. Ed. Engl. 1994, 33, 1089.
[3]
[4]
[3a] H. J. Bestmann, O. Vostrowsky, Top. Curr. Chem. 1983, 109,
[3b]
[10]
[11]
85. Ϫ
863.
B. E. Maryanoff, A. B. Reitz, Chem. Rev. 1989, 89,
A. Ricci, E. Blart, M. Comes-Franchini, G. Reginato, P. Zani,
Pure Appl. Chem. 1996, 68, 679.
D. J. Ager, Org. React. 1990, 38, 1.
α-Methyl-substituted cinnamic acid derivatives can easily be
converted into indanones: M. Beller, T. H. Riermeier, S. Klein,
K. Kühlein, unpublished results.
[5] [5a]
S. Denmark, J. Amburgery, J. Am. Chem. Soc. 1993, 115,
[5b]
10387. Ϫ
I. Creton, I. Marek, D. Brasseur, J.-L. Jestin, J.-F.
[5c]
[12]
[13]
Normant, Tetrahedron Lett. 1994, 35, 6837. Ϫ
S. Sharma,
M. Beller, T. H. Riermeier, Tetrahedron Lett. 1996, 37, 6535.
W. A. Herrmann, C. Brossmer, C.-P. Reisinger, T. H. Riermeier,
K. Öfele, M. Beller, Chem. Eur. J. 1997, 3, 1357.
A. C. Oehlschlager, J. Org. Chem. 1989, 54, 5064. Ϫ [5d] I. Flem-
ming, T. W. Newton, J. Chem. Soc., Perkin Trans. 1 1984, 1805.
[5e]
[14] [14a]
Ϫ
B. H. Lipshutz, E. L. Ellsworth, S. H. Dimock, D. C.
B. A. Patel, C. B. Ziegler, N. A. Cortese, J. E. Plevyak, T.
Reuter, Tetrahedron Lett. 1989, 30, 2065.
C. Zebovitz, M. Terpko, R. F. Heck, J. Org. Chem. 1977, 42,
[6] [6a]
[14b]
R. F. Heck, Palladium Reagents in Organic Synthesis, Aca-
demic Press, London, 1985. Ϫ [6b] R. F. Heck in Comprehensive
Organic Synthesis, vol. 4 (Eds.: B. M. Trost, I. Flemming), Per-
3903. Ϫ
H. A. Dieck, R. F. Heck, J. Org. Chem. 1975, 40,
1083. Ϫ [14c] A. Kasahara, T. Izumi, M. Maemura, Bull. Chem.
Soc. Jpn. 1977, 50, 1021.
gamon Press, Oxford, 1991, p 833. Ϫ [6c] W. Cabri, I. Candiani,
H. A. Dieck, R. F. Heck, J. Am. Chem. Soc. 1974, 96, 1133.
V. V. Grushin, Chem. Rev. 1996, 96, 2011.
[15]
[16]
[17]
[6d]
Acc. Chem. Res. 1995, 28, 2. Ϫ
J. Tsuji, Palladium Reagents
and Catalysts Ϫ Innovations in Organic Synthesis, Wiley, Chi-
J. Louie, J. F. Hartwig, Angew. Chem. 1996, 108, 2531; Angew.
Chem. Int. Ed. Engl. 1996, 35, 2359.
[6e]
chester, UK, 1995. Ϫ
W. A. Herrmann in Applied Homo-
[18]
[19]
geneous Catalysis (Eds.: B. Cornils, W. A. Herrmann), VCH,
Weinheim, 1996, p 712. Ϫ
F. Paul, J. Patt, J. F. Hartwig, Organometallics 1995, 14, 3030.
[6f]
´
A. de Meijere, F. Meyer, Angew.
C. Amatore, E. Carre, A. Jutand, M. A. MЈBarki, G. Meyer,
Chem. 1994, 106, 2437; Angew. Chem. Int. Ed. Engl. 1994, 33,
Organometallics 1995, 14, 5605.
2379.
[97166]
Eur. J. Inorg. Chem. 1998, 29Ϫ35
35