Homogeneous catalysts supported on soluble polymers: Biphasic Sonogashira coupling of aryl halides and acetylenes using MeOPEG-bound phosphine - Palladium catalysts for efficient catalyst recycling

Article abstract of DOI:10.1002/chem.200390161

The Sonogashira coupling of various aryl bromides and iodides with different acetylenes was studied under biphasic conditions with soluble, polymer-modified catalysts to allow the efficient recycling of the homogeneous catalyst. For this purpose, several

Full text of DOI:10.1002/chem.200390161

FULL PAPER
Homogeneous Catalysts Supported on Soluble Polymers: Biphasic
Sonogashira Coupling of Aryl Halides and Acetylenes Using MeOPEG-
Bound Phosphine Palladium Catalysts for Efficient Catalyst Recycling
Axel Kˆllhofer and Herbert Plenio*[a]
Abstract: The Sonogashira coupling of
various aryl bromides and iodides with
different acetylenes was studied under
biphasic conditions with soluble, poly-
mer-modified catalysts to allow the effi-
cient recycling of the homogeneous
catalyst. For this purpose, several steri-
cally demanding and electron-rich
phosphines of the type RPPR2 were
synthesised. They are covalently linked
to a monomethyl polyethylene glycol
used in CH3CN/Et3N/n-heptane (5/2/5).
The combined yields of coupling prod-
uct over five reaction cycles are between
80 95%. There is no apparent leaching
of the catalyst into n-heptane, as evi-
linked to the polymer was used to couple
aryl bromides and acetylenes in DMSO
or DMSO/n-heptane at 608C with
0.5 mol% Na2[PdCl4], 1 mol% RPPR2
and 0.33 mol% CuI. The combined yield
of coupling products over five cycles is
always greater than 90%, except for
sterically hindered aryl bromides. The
determination of the turnover frequency
(TOF) of the catalyst indicates only a
small decrease in activity over five
cycles. Leaching of the catalyst into the
product containing n-heptane solution
could not be detected by means of
1H NMR and TXRF; this is indicative
of >99.995% catalyst retention in the
DMSO solvent.
1
denced by H NMR spectroscopy. The
new
catalyst
[(MeCN)2PdCl2]/2(1-
Ad)2PBn can be used for room-temper-
ature coupling of various aryl bromides
and acetylenes in THF with HNiPr2 as a
base.
A
closely related catalyst
ether with
a
mass of 2000 Dalton
Na2[PdCl4]/2RP-CH2C6H4CH2P(1-Ad)2
À
(RP ˆ MeOPEG2000
)
RPPR2:
PR2 ˆ
À
À
CH2C6H4CH2P(1-Ad)2,
C6H4-P-
Keywords: homogeneous catalysis ¥
biphasic catalysis ¥ C C coupling ¥
palladium ¥ polymers
À
(1-Ad)2, C6H4-PPh2. To couple aryl
iodides and acetylenes, the catalyst
[(MeCN)2PdCl2]/2RP-C6H4-PPh2
was
economically competitive as soon as efficient options for
catalyst recovery are at hand. Hydroformylation relies on a
biphasic aqueous/organic solvent system, with the catalyst
[HRh(CO)(TPPTS)3] residing exclusively in the aqueous
phase, whereas the product is located within the organic
solvent.[9] Hence simple phase separation affects the separa-
tion of catalyst and product; losses of precious Rh are
negligible.[10] Consequently, several liquid liquid biphasic
concepts are currently studied for applications in catalysis:[11]
aqueous/organic,[12] fluorous/organic,[13, 14] ionic liquids,[15, 16]
polymer-supported catalysts[17] and amphiphilic catalysts.[18]
Palladium-catalysed C C coupling reactions are powerful
synthetic methods[19] which have been receiving tremendous
interest in recent years, especially since significant progress in
this field has been made, as represented by publications of the
groups of Beller,[20] Buchwald,[21] Fu,[22] Hartwig,[23] Herr-
mann,[24] Reetz[25] and others.[26] Nevertheless, the high price
of Pd renders commercial processes based on Pd less
attractive[27] unless extremely active and/or recyclable cata-
lysts are available.[28] This is all the more important as
heterogeneous Pd catalysts in coupling reactions rarely match
the activity of the homogeneous ones[29, 30] and they also suffer
from deactivation after recycling.[31]
Introduction
The problems associated with the recovery of catalysts after
product formation pose a serious drawback for large-scale
applications of homogeneous catalysis and are the main
reason why heterogeneous catalysis has a share of more than
80% in industrial processes.[1]
Consequently, numerous approaches that deal with the
problem of catalyst recovery, such as thermal or chemical
recovery,[2] immobilisation of catalysts on solid, liquid or
aqueous supports,[3] membrane processes[4] or multiphase
systems (phase-transfer catalysis, thermoregulated phase-
transfer catalysis, liquid liquid biphasic catalysis),[5] have
been studied.[6]
√
The huge commercial success of the Ruhrchemie/Rhone-
Poulenc hydroformylation[7] and the SHOP process
onstrates that homogeneous catalysis with precious metals is
[8]
dem-
[a] Prof. Dr H. Plenio, Dipl.-Chem. A. Kˆllhofer
Institut f¸r Anorganische Chemie, TU Darmstadt
Petersenstrasse 18, 64287 Darmstadt (Germany)
Fax : (‡49)6151-166040
E-mail: plenio@tu-darmstadt.de
1416
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1416 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 6

1416 1425
Applications of the liquid liquid biphasic concept to Pd-
catalysed C C coupling reactions are rare and mainly[32]
limited to the work of the Bergbreiter group who has
demonstrated that poly(N-isopropylacrylamide) linked to
SCS PdCl complexes yields recyclable Heck Mizoroki and
Suzuki Miyaura catalysts.[33 36] However, the activity of these
catalysts with respect to substrates other than aryl iodides is
low, while for these substrates good heterogeneous catalysts
are also available. Consequently, more active catalysts are
needed to access less reactive aryl bromides or chlorides for
biphasic C C coupling reactions.
in the polar solvent and completely insoluble in the non-polar
solvent to prevent any leaching of the catalyst. Fourthly, it also
has to posses one or a few functional groups that can be used
to attach the catalytically active complex efficiently. Since it is
assumed that the catalytically active species in Pd-catalysed
coupling reactions is a [Pd(PR3)1] complex,[52] a high spatial
density of phosphines groups attached to the polymer is not
desirable, as this would favour chelated and thus higher
coordinated and less active complexes.
For application of these restrictions, monomethyl poly-
ethylene glycol (MeOPEG) appears to be a good choice. The
ether groups are rather unreactive, coordination of the ether
oxygen to the catalytically active Pd0 species is unfavourable
Apart from Bergbreiter×s research, the biphasic concept
that employs soluble polymer-supported catalysts has only
been applied to other metal-catalysed reactions, such as
hydrosilylation,[37] hydrogenation[38] (asymmetric)[39] and hy-
droformylation.[40, 41] The less well-developed status of this
field is somewhat surprising as the concept of the use of
soluble polymer-supported catalysts has been suggested as
À
while the OH is suitable for the attachment of the catalyti-
cally active group. The linear polyether is highly soluble in a
number of polar solvents, but completely insoluble in
alkanes.[43] Furthermore, MeOPEG is available commercially
in various molecular weights ranging from 350 to 5000 Dal-
ton.[53]
44]
early as 1976 by Beyer and Schurig.[42 Compared with the
fluorous biphasic concept, the use of soluble polymer-
supported catalysts appears to have a number of advantages:
1) Polymers, such as polyethylene glycol (PEG), are extreme-
ly cheap, quite in contrast to perfluorinated molecules.
2) Standard organic solvents can be applied to biphasic
solvent systems that use PEG-supported catalysts;[45] per-
fluorinated solvents used for fluorous catalysis are not
needed.[46] 3) Polyethylene glycol is environmentally be-
nign,[47] whereas fluorocarbons are persistent and pose certain
risks related to ozone damage and the greenhouse effect.
4) Leaching of polymer-supported catalysts into the product
phase is negligible, whereas this often poses a problem with
fluorous biphasics.
Consequently, we have initiated a program to develop
polymer-tagged palladium phosphine complexes for car-
bon carbon bond-forming reactions, such as Heck, Suzuki
and Sonogashira coupling. The Sonogashira reaction[48] is now
widely applied to the synthesis of natural compounds,
pharmaceuticals, liquid crystalline polymers and optical or
electronic materials.[49, 50, 51] We decided to use monomethyl
polyethylene glycol (MeOPEG) as a cheap and chemically
robust support for the attachment of sterically demanding and
electron-rich phosphines, to allow the efficient recycling of the
catalyst and its multiple use. We report here on our results on
the palladium-catalysed Sonogashira coupling of aryl halides
and acetylenes under biphasic conditions and the efficient
recycling of the catalysts.
A simple polymer-tagged phosphine was synthesised
(Scheme 1) by first reacting 4-iodophenol with Ph2PH under
Pd catalysis.[54] The respective phenol, formed in virtually
Scheme 1. Synthesis of a simple polymer-tagged phosphine.
quantitative yield, is treated with mesylated MeOPEG-OMes
(MeOPEG ˆ CH3(OCH2CH2)nOH, here nmax ˆ 44, 2000 Dal-
ton) to give MeOPEG-O-C6H4PPh2 in 81% yield.[55, 56] We
decided to use the 2000 Dalton polymer because preliminary
1H NMR experiments had shown that this molecular weight is
sufficient to prevent any leaching of the phosphine into n-
heptane.
It should be noted that this phosphine and all other
MeOPEG phosphines synthesised by us display solubility in
water, which should also allow their use in aqueous/organic
solvent systems.[57]
Biphasic Sonogashira reactions of aryl iodides: To validate the
concept of effective catalyst recycling by means of liquid
liquid biphasic catalysis, we applied the standard Sonogashira
coupling reaction to a biphasic system. These conditions
include the classic catalyst[58] consisting of [(PPh3)3PdCl2], CuI
and an amine as the base. Initially, we studied the Sonogashira
Results and Discussion
ꢀ
ꢀ
coupling of 1-hexyne, 1-octyne, PhC CH and Et3SiC CH
with aryl iodides (Scheme 2).
Synthesis of phosphines covalently bonded to MeOPEG: A
soluble polymer used for attachment to a phosphine coordi-
nated to a catalytically active metal should fulfil several basic
requirements. Firstly, the polymer must not have functional
groups along the chain which might react during catalysis.
Secondly, it should not coordinate to the catalytically active
metal. Thirdly, for use in biphasic catalysis, the ideal polymer
also has to be fairly polar (which unfortunately readily
interferes with the first and second conditions) so it is soluble
The coupling reactions of the aryl iodides were performed
in a ternary solvent mixture of CH3CN/Et3N/n-heptane
(5:2:5; vol) as a biphasic system. The catalyst was prepared
in situ from [(CH3CN)2PdCl2] and two equivalents of the
Scheme 2. The Sonogashira coupling reaction of aryl iodides.
Chem. Eur. J. 2003, 9, No. 6
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1417 $ 20.00+.50/0
1417

H. Plenio and A. Kˆllhofer
FULL PAPER
MeOPEG-linked PPh3 (1). Et3N acts not only as a base, but
also as a solubility mediator for this biphasic system, which
helps to produce complete miscibility of the solvents (i.e.
monophasic conditions) above 808C. The biphasic reaction
mixture was heated above the critical mixing temperature[59]
and the reaction was allowed to proceed for the specified
period of time after which it was cooled to room temperature
whereupon phase separation occurred. Each reaction was
repeated five times in the biphasic solvent mixture with a
single batch of catalyst added prior to the first cycle (Fig-
ure 1).
however, can be solved by adding K2CO3 as a second base to
regenerate Et3N by deprotonation of the ammonium salt.
Excess salt (KHCO3, KI) can be removed by filtration after a
number of cycles.
The overall yields for all coupling reactions after a few
hours are higher than 80% or even 90% after chromato-
graphic purification. This compares favourably to results of a
brief report by Bergbreiter et al.;[61] Bannwarth and Markert
recently reported on a fluorous biphasic Sonogashira catalysis
of aryl iodides.[62] However, the amount of isolated product
from the n-heptane layer after each cycle (i.e. phase yield) can
be much lower during the first cycles. This is not caused by a
reduced activity of the catalyst during the first cycles.
Independent tests have shown that the yields of product after
the first cycle are consistently higher than 90%.
Instead, depending on the nature of the coupling product,
its distribution coefficient between CH3CN and n-heptane is
not always favourable. The isolated yields of the products
from the n-heptane layer (phase yields) increase after each
cycle sometimes to values larger than 100% (Table 1), once
the polar CH3CN/NEt3 phase is saturated with the product.
Figure 1. Biphasic catalysis.
‡
The probable explanation is the partial solubility of Et3NH IÀ
in CH3CN/NEt3 leading to an increasing polarity of the lower
phase after each cycle. This will result in a decrease of the
product solubility in the polar acetonitrile layer.
Following each reaction the n-heptane solution containing
most of the product (according to the respective partition
coefficient) was separated, the volatiles were removed in
vacuum to isolate the crude product. The evaporated solvents,
mainly n-heptane and some leached Et3N and CH3CN,[60] can
be 100% recycled by adding the mixture to the CH3CN
solution containing the catalyst prior to the next cycle. For
each consecutive cycle one equivalent of the respective aryl
iodide and the respective acetylene were added to the catalyst
in the acetonitrile solution. The main results of the catalytic
reactions are summarised in Table 1.
Attempted biphasic Sonogashira coupling of aryl bromides
and acetylenes: Encouraged by the positive results with aryl
iodides, the same procedure was tested for the Sonogashira
coupling of aryl bromides, which are commercially more
attractive on account of their lower price, while from a
reactivity point of view they are more challenging substrates.
However, reactions of various aryl bromides with PhCCH in
several biphasic systems (acetonitrile/n-heptane, DMSO/n-
heptane, DMF/n-heptane, propylene carbonate/n-heptane)
gave modest-to-poor yields of the desired coupling product.
Even with extended reaction times of up to 24 h per cycle,
only modest (30 50%) yields with reactive substrates such as
4-bromoacetophenone were isolated. Deactivated substrates,
Initially, perfect miscibility in the ternary solvent system is
possible. However, monophasic conditions are more and
more difficult to achieve after the first cycles as a result of the
formation of HNEt3 IÀ, which dissolves in acetonitrile leading
‡
to an increase in the polarity of this solvent. This problem,
Table 1. Biphasic Sonogashira coupling of aryl iodides and acetylenes.[a]
Entry
Acetylene
Time
[h]
1
[%]
2
[%]
3
[%]
4
[%]
5
[%]
Extraction after
last cycle[%]
Overall
yield[%]
ꢀ
1
2
3
4
5
6
7
8
9
4-Me
4-Me
4-Me
4-MeO
4-MeO
4-MeO
4-Cl
n-C4H9C CH
3
3
1.5
2
2
2
3
3.5
4
89
52
80
82
32
73
78
46
81
93
83
101
89
74
89
9098
76
92
93
89
99
96
74
98
97
92
102
99
109
108
89
94
102
99
117
105
19
64
16
15
96
5
89
90
7
88
91
96
85
83
86
ꢀ
PhC CH
ꢀ
Et3SiC CH
ꢀ
n-C4H9C CH
ꢀ
PhC CH
ꢀ
Et3SiC CH
ꢀ
n-C4H9C CH
99 013 10
96
ꢀ
4-Cl
4-Cl
PhC CH
76
95
109
108
91
95
ꢀ
Et3SiC CH
104
[a] Conditions: solvents CH3CN/Et3N/n-heptane (5:2:5), 1 mol% [(CH3CN)2PdCl2], 2 mol% MeOPEGOC6H4PPh2 1, 4 mol% CuI. The yields given for the
cycles 1 5 are phase yields of the crude products, which are different from the chemical yields of the respective reaction as they are influenced by the
partition coefficient of the product between the two solvents in the biphasic system. Extraction after the last cycle corresponds to the amount of product
extracted completely with n-heptane from CH3CN after the fifth cycle. The overall yields correspond to the amount of isolated products after
chromatographic purification.
1418
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1418 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 6

Biphasic Sonogashira Coupling
1416 1425
such as 4-bromotoluene or 4-bromoanisole, gave less than
10% of the coupling product.
At this point, we decided to investigate in more detail the
reason for this failure. There are three probable causes: a) the
attachment of the MeOPEG polymer to the catalyst, b) the
influence of the polar solvent or c) the standard Sonogashira
conditions may not be suitable for such biphasic solvent
systems and need to be re-optimised.
The basic lesson learned from the recent work on Pd-
catalysed coupling reactions is that low-coordinate PdL1
complexes are the most active catalysts.[63] Consequently, it
is clear both the coordinating solvent CH3CN as well as the
less bulky polymer-linked PPh3 are detrimental to high
catalytic activity.
Scheme 4. Synthesis of phosphonium salts 3 and 4.
under the basic reaction conditions of Sonogashira coupling.
All three phosphines are easily modified to incorporate
additional functional groups which allow their attachment to a
MeOPEG polymer.
Room-temperature Sonogashira coupling: Next, the perform-
ance of the (1-Ad)2-phosphine-based co-catalysts 2 4 was
tested. (Table 2). The only catalyst which gave virtually
quantitative yields for all coupling reactions tested is
[(PhCN)2PdCl2]/(1-Ad)2PBn. It should be stated at this point
that our catalytic system has the same activity as the one
relying on [(PhCN)2PdCl2]/PtBu3 described recently by Buch-
wald, Fu et al.[50] The next step was to attach the top
performers (1-Ad)2PBn (3) and (1-Ad)2PPh (2) to the
MeOPEG polymer and study the coupling reaction under
biphasic conditions.
Our next task, therefore, was to synthesise a sterically
demanding and electron-rich phosphine co-catalyst that could
be readily attached to a MeOPEG polymer.
Synthesis of sterically demanding phosphines: Bulky and
electron-rich phosphines, such as PtBu3, are responsible for
excellent activity of the respective Pd complexes in C
C
coupling reactions.[20 24] The catalyst [(PhCN)2PdCl2]/PtBu3 is
the only one known to allow room-temperature Sonogashira
coupling.[50, 51] Since the phosphine in this catalyst appears to
be unsuitable for the attachment to MeOPEG, we decided to
synthesise various di-(1-adamantyl)phosphines ((1-Ad)2-
phosphines) as Beller et al. showed that (1-Ad)2PBu is an
excellent co-catalyst in Suzuki coupling reactions.[20] (1-
Ad)2PH is readily accessible in quantities >100 g on account
of its facile preparation starting from adamantane and PCl3.[64]
To obtain high activity catalysts for biphasic Sonogashira
coupling, we had to synthesise suitable phosphines that carry
additional functional groups to allow the covalent attachment
to the MeOPEG backbone.
Table 2. Yields of room-temperature Sonogashira reactions of aryl bro-
mides and acetylenes.[a]
R'
PR3
R
4-CH3CO[%]
H[%]
4-CH3O[%]
Ph
Ph
Ph
n-C6H13
4
2
3
3
92
> 95
> 95
> 95
< 10
45
> 95
> 95
< 5
60
> 95
> 95
Prior to the synthesis of MeOPEG-tagged (1-Ad)2-phos-
phines, three low molecular weight analogues 2 4, were
prepared to study their catalytic performance in the mono-
phasic Sonogashira reaction (Scheme 3).
[a] Yields given correspond to isolated compounds after chromatographic
purification. Catalyst: 4 mol% PR3, 2 mol% [(PhCN)2PdCl2], 1.5 mol%
CuI.
Synthesis of sterically demand-
ing and electron-rich phos-
phines covalently linked to
MeOPEG: To synthesise MeO-
PEG-tagged analogues of 2 and
3, phosphines 5 and 6 were
prepared as described for the
Scheme 3. Low molecular weight analogues 2 4 of the MeOPEG-tagged (1-Ad)2-phosphines.
respective unsubstituted rela-
tives 2 and 3 (Scheme 5).
Phosphine 2 was prepared from PhLi/CuI and (1-Ad)2PCl,
the latter being readily accessible in quantitative yield from
(1-Ad)2PH and CCl4. The phosphonium salts 3 and 4 were
obtained by reacting the respective alkyl halides with (1-
Ad)2PH in toluene at 1008C; precipitation from the reaction
mixture gave air-stable, colourless powders in quantitative
yields. It is convenient to store the phosphonium salts rather
than the respective phosphines, since the protonated form is
stable towards oxidation (Scheme 4).
Phosphine 2 and the phosphonium salts 3 and 4 could be
Scheme 5. Sterically demanding and electron-rich phosphines 5 and 6 to be
covalently linked to MeOPEG.
used directly for catalysis as the latter two are deprotonated
Chem. Eur. J. 2003, 9, No. 6
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1419 $ 20.00+.50/0
1419

H. Plenio and A. Kˆllhofer
FULL PAPER
Table 3. Solvent and base screening for Sonogashira coupling.[a, b]
To our surprise, a simple Williamson-type etherification of
MeOPEG2000OH with 6 was not successful. On the other
hand, 7 could be readily synthesised from the reaction of
MeOPEG-OH and 1,4-(CH2Br)2C6H4 with NaH as a base to
produce MeOPEG-O-CH2C6H4CH2Br, which upon reaction
with (1-Ad)2PH generates the respective air-stable phospho-
nium salt in almost quantitative yield. The loading of the (1-
Ad)2P group on the polymer is in excess of 90%, as
determined by 1H NMR spectroscopy (Scheme 6).
MeCN
C6H13 Ph
DMA
C6H13 Ph
DMF
C6H13 Ph
DMSO
C6H13 Ph
PC[c]
C6H13
Ph
NEt3
À
À
À
À
À
À
Æ
À
À
À
À
À
À
Æ
À
À
À
À
À
‡À ‡À
Na2CO3 ‡À À‡
À
À
‡‡ ‡‡
‡‡ ‡‡
Æ
À
Æ
À
À
Æ
À
Æ
À
À
À
K2CO3
Cs2CO3
K3PO4
NaOAc
HNiPr2
Æ
Æ
À
À
À‡
À
À
À
‡‡
À
‡‡
À
‡
À
À
Æ
À
À
À
À
À
À
‡‡ ‡‡
[a] Catalyst: (1-Ad)2PBn (4 mol%), Na2[PdCl4] (2 mol%), CuI (1.5 mol%).
[Pd(OAc)2] was used for reactions with NaOAc. [b] Conversions: À < 40%,
Æ 40 80%, ‡ >80%, ‡ ‡ near quantitative. [c] Propylene carbonate.
The reaction of several aryl bromides (bromoacetophe-
none, bromobenzene, 4-bromochlorobenzene, 4-bromoani-
sole, 2-bromotoluene) and acetylenes (PhCCH, Me3SiCCH,
1-octyne) was tested under the conditions described in
Table 4. The reaction times depend on the substrate and were
optimised for all reactions, by monitoring the progress of the
reaction by TLC. The time needed for apparent completion of
the reaction (TLC test, absence of starting material) in the
first cycle was doubled and used for all consecutive cycles.
When following this procedure, the yield of the reaction is a
useful criterion for the evaluation of catalyst recyclability. The
fact that during each individual cycle from 1 5 the reaction is
quantitative for the coupling of PhCCH and Me3SiCCH with
most aryl bromides indicates that there is no significant
deactivation of the catalyst. The concentration of the catalyst
can be lowered slightly to 0.5 mol% of Na2[PdCl4] with only a
small reduction in the yield (Table 5).
Scheme 6. Synthesis of 7.
Deprotection of phosphine 5 with CsF and reaction of the
respective phenol with mesylated MeOPEG-OH gave the
corresponding polymer-tagged phosphine 8 with >90%
loading of the phosphine.
Optimising the coupling reaction in solvents suitable for
biphasic catalysis: For liquid liquid biphasic catalysis, a
solvent is needed which is immiscible at room temperature
with simple alkanes used as the product phase. Low-polarity
solvents (such as dioxane) are not suitable as a catalyst phase,
moreover, the selection of polar solvents is limited as a typical
polar solvent almost invariably has donor groups which can
block the active palladium complex and thus reduce its
catalytic activity. Consequently, the right choice of solvent is
the key for the success of biphasic catalysis.
For a solvent/base screen of the Sonogashira coupling of
phenylacetylene and 1-octyne with co-catalyst 2, we chose the
deactivated substrate p-bromoanisole with different solvent/
base combinations. This first screen is based on estimated
TLC spot intensity of the reaction mixtures. Furthermore, we
decided to use Na2[PdCl4], which is reasonably soluble in
polar solvents, as a simple palladium source since it is cheaper
than [(PhCN)2PdCl2].
Further lowering of the catalyst concentration (PR3
0.4 mol%, Na2[PdCl4] 0.2 mol%, CuI 0.17 mol%) in the
reaction of 4-bromoanisole and PhCCH led to a significant
increase in times needed to achieve quantitative yields, which
is considered impractical.
Much to our surprise, the catalyst performance in the
coupling reactions involving 1-hexyne and 1-octyne deterio-
rated rapidly after the first cycle. While initially almost
quantitative yields are produced, the coupling yields in the
2nd to the 5th cycle drop drastically. We also noticed that in
contrast to the other coupling procedures, the reaction
mixture invariably turns black during the first cycle, indicating
formation of insoluble Pd black. Information from the
vendors revealed that 1-hexyne and 1-octyne are contami-
nated with up to 3% of alkyl bromides.[65] We reasoned that
this could be a cause for catalyst deactivation. Consequently,
commercially available alkynes were treated with Et2NH to
remove the alkyl bromides and tested again. Unfortunately,
this treatment did not result in an improvement of the
coupling reaction in the second cycle.
More forcing conditions are required (compared to the use
of dioxane or THF as a solvent), for coupling reactions in
polar and coordinating solvents (Table 3). The screening
clearly shows that DMSO gives the highest yield both with
Na2CO3, K2CO3 and HNiPr2 as bases. A more detailed screen
revealed HNiPr2 to be the best base for Sonogashira coupling
reactions in DMSO.
Biphasic Sonogashira coupling of aryl bromides and acety-
lenes by the use of polymeric phosphines: Next, the Pd
complexes of the MeOPEG-linked phosphines 7 and 8 were
tested under biphasic conditions in DMSO/heptane with
HNiPr2 as a base and Na2[PdCl4] as the Pd source. It
immediately became clear that 7 performed much better than
8, which then was not studied in more detail.
The chemical yield of the coupling reaction is easy to
determine, but it may not be a perfect criterion to evaluate the
recyclability of catalysts according to Gladysz.[66] Therefore,
we probed the yield of the catalytic reaction during several
cycles after fixed periods of time to obtain TOF data. The
reactions of PhCCH with 4-bromoacetophenone, bromoben-
1420
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1420 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 6

Biphasic Sonogashira Coupling
1416 1425
Table 4. Biphasic Sonogashira coupling of aryl bromides with acetylenes.[a]
Entry
R
Time
[h]
Run 1
[%]
Run 2
[%]
Run 3
[%]
Run 4
[%]
Run 5
[%]
Extraction after
last cycle[%]
Overall
yield[%]
1
2
3
4
5
6
7
8
4-CH3CO
H
4-Me
4-Cl
4-MeO
2-Me
H
4-Cl
4-Me
3CO
4-MeO
2-Me
4-CH3CO
4-Me
4-MeO
Ph
Ph
Ph
Ph
Ph
Ph
Me3Si
Me3Si
Me3Si
Me3Si
Me3Si
Me3Si
n-C6H13
n-C6H13
n-C6H13
0.5
0.5
1
1
2
3
3
3
3
2
4
5
12
12
12
36
73
90
87
48
94
80
93
94
76
92
83
95
95
81
79
99
99
97
88
100
88
99
107
93
99
92
107
104
102
93
103
103
101
110
99
109
109
103
105
100
104
100
103
100
102
105
90906
25
123
108
104
105
103
104
105
103
104
103
108
72
21
13
16
77
10
25
9
94
94
96
94
94
96
92
93
94
92
93
9
9
25
5
104-CH
11
12
13
14
103
88
36
47
87
8052
77
81
57
6
6
16
6
33
15
52
15
61
31
037
[a] Conditions: catalyst: PR3 (7, 2 mol%), [Na2PdCl4] (1 mol%), CuI (0.7 mol%), DMSO/n-heptane, 608C. Reactions involving 1-octyne and 1-hexyne were
conducted at 758C. The yields given for the cycles 1 5 are phase yields of the crude products (taking account of the partition coefficient of the product).
Extraction after last cycle corresponds to the amount of product extracted with n-heptane from DMSO after the fifth cycle. The overall yields correspond to
the amount of isolated products after chromatographic purification.
Table 5. Biphasic Sonogashira coupling of aryl bromides with acetylenes (0.5 mol% Pd).[a]
Entry
R
Time
[h]
Run 1
[%]
Run 2
[%]
Run 3
[%]
Run 4
[%]
Run 5
[%]
Extraction after
last cycle[%]
Overall
yield[%]
1
2
3
4-CH3CO
H
4-MeO
PhCCH
PhCCH
PhCCH
1
3
4
46
82
63
83
91
87
96
105
97
104
104
103
115
109
110
45
13
56
93
91
90
[a] Conditions: catalyst: PR3 (7, 1 mol%), Na2[PdCl4] (0.5 mol%), CuI (0.35 mol%), DMSO/n-heptane, 608C. See also the legend to Table 4.
zene and 4-bromoanisole were conducted and GC samples
taken, prior to completion of the reaction.[67] The individual
yields (run, TOF) of the coupling reactions were as follows:
4-bromoanisol for 30min 84% (run 1, 336 h À1), 80% (run 2,
320h À1), 73% (run 3, 292 hÀ1), 65% (run 4 260h À1), 63%
(run 5, 252 hÀ1); bromobenzene for 20min: 75% (run 1,
440h À1), 72% (run 2, 432 hÀ1), 65% (run 3, 390h À1), 56%
(run 4, 336 hÀ1), 52% (run 5, 312 hÀ1); 4-bromoacetophenone
for 10min: 96% (run 1, 1150h À1), 93% (run 2, 1120h À1),
85% (run 3, 1020 hÀ1), 76% (run 4, 912 hÀ1), 73% (run 5,
880h À1), 83% (run 6, 1000 hÀ1). In all three test reactions
there is a small but significant decrease in the chemical yield
over the five cycles probed. It is, however, pleasing to note
that the addition of CuI after the 5th cycle reactivates the
catalysts. In this manner, the catalytic coupling of 4-bromoa-
cetophenone rises to 83% yield (TOF 1000 hÀ1) in a 6th cycle
(from 73%, TOF 880h À1 in the 5th run) after having added
the same amount of CuI as used initially. We thus believe the
deactivation of the catalyst to be mainly caused by adventi-
tious oxidation of CuI to CuII.
out the reason for this somewhat lower activity of the
polymeric catalyst in DMSO. To do this, we compared the
performance of Na2[PdCl4]/(1-Ad)2PBn and Na2[PdCl4]/RP-
OCH2C6H4CH2P(1-Ad)2
[(PhCN)2PdCl2]/(1-Ad)2PBn
in
DMSO
and
and
that
of
[(PhCN)2PdCl2]/RP-
OCH2C6H4CH2P(1-Ad)2 in THF. There is no significant
difference between the two catalysts in DMSO, while the
polymeric catalyst is less efficient in THF.
Retention of the catalyst in DMSO: The absence of leaching,
that is the virtually quantitative retention of the catalyst in the
polar solvent solution, is very important for the evaluation of
1
biphasic catalysis. As evidenced by H NMR spectra of the
crude product there seems to be no leaching of MeOPEG-
phosphine from CH3CN or DMSO into the n-heptane layer
during the reactions of the aryl halides. Because of the large
signal associated with the polymer, the detection limit for
phosphine leaching is fairly low and the absence of a signal
shows that retention of the catalyst must be much larger than
99.5%. We also studied catalyst leaching by means of total
reflection X-ray fluorescence (TXRF), which is a very
sensitive technique for metal (here palladium and copper)
detection. The amount of palladium lost into the n-heptane is
extremely small and below the detection limit of the TXRF
method, this means that the retention of the catalyst within
Comparison of the catalysts–polymeric versus low molecular
weight: As can be seen from the Tables above, the polymeric
catalyst derived from phosphine 7 in DMSO solution is
slightly less efficient than the small molecule catalyst with
phosphine (1-Ad)2PBn (3) in THF solvent. We wanted to find
Chem. Eur. J. 2003, 9, No. 6
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1421 $ 20.00+.50/0
1421

H. Plenio and A. Kˆllhofer
FULL PAPER
200 MHz, 31P NMR 81 MHz) spectrometer. 1H NMR were referenced to
residual protonated impurities in the solvent, 13C NMR to the solvent signal
(CDCl3 : d(1H) ˆ 7.24 ppm, d(13C) ˆ 77.0ppm) and 31P NMR spectra were
referenced to PMe3 (38% in benzene d ˆ À62 ppm) as an external
standard. Starting materials were commercially available or prepared
according to literature procedures: (1-Ad)2PH,[64] MetBu2SiO(CH2)4I,[73]
[(PhCN)2PdCl2] and [(CH3CN)2PdCl2].[74]
the DMSO solution must be >99.995%![68] The same applies
to copper leaching.
During each cycle there is, however, a small loss of the polar
solvent DMSO into the n-heptane layer, since a saturated n-
heptane solution contains 1.1% (vol.) of DMSO.[69]
Screening experiments: To evaluate the yield in the first solvent, base
screening of coupling reactions was carried out by estimating the TLC spot
intensities.
Conclusions
4-BrC6H4OSitBuMe2: To 4-bromophenol (3.46 g, 20mmol) in CH 2Cl2
(50mL), were added triethylamine (2.53 g, 25 mmol) and ClSi tBuMe2
(3.02 g, 20 mmol) at room temperature with vigorous stirring. The solution
was stirred for 12 h at room temperature and subsequently added to a
separatory funnel containing water (50mL) and CH 2Cl2 (50mL). The
phases were separated and the organic phase was washed with water (2 Â
50mL). After drying and evaporation of the solvent, the oily residue was
purified by chromatography on silica with cyclohexane. Yield 5.63 g (98%)
with spectroscopic data identical to those previously reported.[75]
Sterically demanding and electron-rich phosphines were
covalently linked to soluble monomethyl polyethylene glycol
polymer RP (RP ˆ MeOPEG2000). The use of a suitable
palladium source yields highly effective palladium phos-
phine catalysts for Sonogashira-type coupling reactions with
aryl iodides and bromides, which can be conducted under
biphasic conditions, to allow the efficient recycling of the
polymeric catalyst.
4-(HO-C6H4)PPh2 (1): This compound was prepared according to a
procedure reported by Stelzer and co-workers with a modified workup.[54]
The crude product, dissolved in cyclohexane/ethyl acetate (4:1), was
filtered over a silica plug to yield 4-(HO-C6H4)PPh2 (1) as a pale yellow
solid in 92% yield (literature value: 67%).
Aryl iodides can be coupled to acetylenes in almost
quantitative yields during several cycles by means of a ternary
biphasic solvent system consisting of MeCN/Et3N/n-heptane
and a simple catalyst derived from MeOPEG-OC6H4PPh2 and
[(MeCN)2PdCl2]. To use less reactive substrates, a new
catalyst [(PhCN)2PdCl2]/(1-Ad)2PBn was synthesised, which
is able to couple aryl bromides with acetylenes in quantitative
yields at room temperature. A closely related catalyst was
Na2[PdCl4]/RP-OCH2C6H4CH2P(1-Ad)2 tested under biphasic
conditions. By the use of DMSO as a polar catalyst phase and
n-heptane as the product phase, aryl bromides can be coupled
with various acetylenes with 0.5 mol% catalyst at 608C in
almost quantitative yields over five catalytic cycles. The
leaching of the catalyst into the n-heptane layer appears to be
negligible because TXRF measurements show the retention
of the catalyst in the DMSO solution to >99.995%. The
stability of the catalyst over several cycles was demonstrated
by the almost quantitative yields of the coupling reaction and
by a nearly constant turnover frequency of the catalyst. The
concept of sterically demanding phosphines as highly active
co-catalysts can be applied to polymer-tagged phosphines.
Comparative experiments have also shown that the MeOPEG
polymer chain has no detrimental effect on catalyst perform-
ance in aryl bromide coupling. The application of the concept
of MeOPEG-supported carbon carbon bond-forming cata-
lysts appears to be quite general as biphasic Heck Mizoroki
and Suzuki coupling reactions are also possible.[70, 71]
(1-Ad)2PCl: (1-Ad)2PH (12.1 g, 40mmol), dissolved in dry and degassed
CCl4 (100 mL), was stirred at 508C for 16 h. After removal of the solvent
and drying in a vacuum, the pure product was obtained as a colourless
powder with spectroscopic data identical to those reported by Schmutzler
et al.[64] Yield: 13.5 g (100%).
(1-Ad)2PPh (2): CuBr (1.44 g, 10mmol) was added at room temperature to
the freshly prepared Grignard reagent, made from bromobenzene (1.46 g
10mmol) and Mg turnings (270mg, 11.1 mmol) in THF (50mL). The
mixture was stirred for 20min to effect a complete transmetallation. After
addition of 2 (3.37 g, 10mmol) the mixture was heated under reflux for
14 h. The solvent was removed, the dark residue was extracted with hot
cyclohexane and filtered to yield a yellow solution containing the product
as the CuI complex. This solution was extracted repeatedly with 10%
aqueous ammonia (50mL) until the aqueous phase remained colourless.
After the organic layer (MgSO4) had been dried and the solvent removed,
the crude product was obtained as a sticky yellow solid which was
recrystallised from ethanol to yield (1-Ad)2PPh (3) as a pale yellow powder.
Yield 2.46 g (65%); 1H NMR (CDCl3): d ˆ 7.65 7.45 (m, 2H; ArH), 7.30
7.20(m, 3H; ArH), 2.10 1.55 (m, 30H; AdamantylH); 13C NMR (CDCl3):
d ˆ 134.5, 130.9, 128.8, 127.3, 42.4, 36.6, 34.8, 28.9; 31P NMR (CDCl3): d ˆ
41.9.
4-(tBuMe2SiO)C6H4P(1-Ad)2 (5): This compound was prepared as descri-
bed for 2 from 4-(tBuMe2SiO)-bromobenzene (2.87 g, 10mmol) instead of
bromobenzene. Yield 3.21 g (63%); 1H NMR (CDCl3): d ˆ 7.60 7.40 (m,
3
2H; ArH), 6.77 (d, 2H; J ˆ 8.1 Hz, ArH), 2.10 1.50 (m, 30H; Adaman-
tylH), 0.96 (s, 9H; SiC(CH3)3), 0.20 (s, 6H; SiCH3); 13C NMR (CDCl3): d ˆ
15 6.6, 125.8, 119.2, 118.9, 41.7, 37.0, 36.2, 28.9, 25.7, 18.3, À4.3; 31P NMR
(CDCl3):d ˆ 39.4.
General procedure for the synthesis of alkyl-di-(1-adamantyl)phosphoni-
um halides: (1-Ad)2PH (1.21 g, 4 mmol) and the respective alkyl halide
(5 mmol) were dissolved in toluene (20mL) and stirred at 90 8C. In all
reactions, precipitation of the product started after a few minutes. After
14 h heating of the reaction mixture ended. The product was filtered
through a fritted funnel and washed with toluene (20mL) and ether (3 Â
20mL). After drying in a vacuum the phosphonium salts were obtained as
colourless powders.
Experimental Section
General: MeOPEG2000OH, aryl halides and acetylenes were used as
received. Solvents were purified by standard procedures.[72] Carbonate and
acetate bases were dried at 808C under vacuum. Reactions were performed
under an atmosphere of argon by means of standard Schlenk techniques.
(1-Ad)2PBn ¥ HBr (3): Prepared from benzyl bromide (0.86 g, 5 mmol).
Yield: 1.75 g (92%); 1H NMR (CDCl3): d ˆ 7.95 (dt, 1J(P,H) ˆ 478, 3J ˆ
6.0Hz, 1H; PH), 7.63 (d, 3J ˆ 7.2 Hz, 2H; ArH), 7.40 7.10(m, 3H; ArH),
3.83 (dd, 2J(P,H) ˆ 13.3, 3J ˆ 6.0Hz, 2H; ArCH 2), 2.40 1.55 ppm (m, 30H;
AdamantylH); 13C NMR (CDCl3): d ˆ 130.7 (d, 2J(P,C) ˆ 7.5 Hz), 130.3 (d,
Column chromatography was performed on silica MN60(63 200 mm), and
thin-layer chromatography (TLC) on Merck plates coated with silica gel60,
F254. Yields of the coupling reactions were determined by 1H NMR
spectroscopy. Gas chromatography was carried out with an Perkin Elmer
Autosystem. NMR spectra were recorded at 293 K with a Bruker AC300
(1H NMR 300 MHz, 13C NMR 75 MHz) or a Bruker 200AC (1H NMR
1
3J(P,C) ˆ 5.8 Hz), 129.6, 128.2, 38.6, 38.2, 35.6, 27.6, 19.6 ppm (d, J(P,C) ˆ
37.2 Hz); 31P NMR (CDCl3): d ˆ 23.6 ppm.
4-BrC6H4CH2P(1-Ad)2 ¥ HBr: Prepared from 4-bromobenzyl bromide
(1.25 g, 5 mmol). Yield: 1.96 g (89%); 1H NMR (CDCl3): d ˆ 8.32 (dt,
1J(P,H) ˆ 480, 3J ˆ 6.1 Hz, 1H; PH), 7.60(d, 3J ˆ 8.2 Hz, 2H; ArH), 7.48 (d,
1422
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1422 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 6

Biphasic Sonogashira Coupling
1416 1425
3J ˆ 8.2 Hz, 2H; ArH), 3.75 (dd, 2J(P,H) ˆ 13.1, 3J ˆ 6.1 Hz, 2H; ArCH2),
2.40 1.55 ppm (m, 30H; AdamantylH); 13C NMR (CDCl3): d ˆ 132.7,
3
ArH), 6.86 (d, J ˆ 78.4 Hz, 2H; ArH), 4.14, (m, 2H; a-CH2), 3.9 3.4 (m,
ꢁ170 180H; PEGH), 3.34 ppm (s, 3H; CH 3OPEG); 31P NMR (CDCl3):
d ˆ 34.4 ppm.
3
2
132.1 (d, J(P,C) ˆ 5.7 Hz), 130.1 (d, J(P,C) ˆ 7.6 Hz), 38.7, 38.3, 35.6, 27.6,
1
19.1 ppm (d, J(P,C) ˆ 37.8 Hz); 31P NMR (CDCl3):d ˆ 23.3 ppm.
4-(MeOPEG2000OCH2)C6H4CH2P(1-Ad)2 ¥ HBr (15): 4-(MeOPeg2000CH2)-
C6H4CH2Br (11, 4.4 g, 2 mmol) and (1-Ad)2PH (1.2 g, 4 mmol) were
dissolved in toluene (50mL) and heated to 90 8C for 14 h whereupon a
small amount of colourless precipitate was formed. The warm reaction
mixture was filtered over Celite, concentrated to ꢁ15 mL in a vacuum and
added to diethyl ether (60mL) with vigorous stirring. The precipitate was
collected by filtration, washed with diethyl ether and dried in a vacuum.
4-BrCH2C6H4CH2P(1-Ad)2 ¥ HBr (6): Prepared from 1,4-(CH2Br)2C6H4
(1.00 g, 3.79 mmol). Yield: 1.69 g (79%); 1H NMR (CDCl3): d ˆ 8.14 (dt,
1J(P,H) ˆ 478, 3J(H,H) ˆ 5.8 Hz, 1H; PH), 7.6 (d, 3J ˆ 7.8 Hz, 2H; ArH),
7.34 (d, 3J ˆ 7.8 Hz, 2H; ArH), 4.40(s, 2H; CH 2Br), 3.78 (dd, 2J(P,H) ˆ 13.2,
3J(H,H) ˆ 5.8 Hz, 2H), 2.4 1.6 ppm (m, 30H; 1-AdH); 31P NMR (CDCl3):
d ˆ 23.1 ppm.
tBuMe2SiO(CH2)4P(1-Ad)2 ¥ HI (4): Prepared from tBuMe2SiO(CH2)4I
(1.57 g, 5 mmol). Yield: 2.32 g (94%); 1H NMR (CDCl3): d ˆ 7.39 (dt,
1J(P,H) ˆ 464, 3J ˆ 3.6 Hz, 1H; PH), 3.59 (t, 3J ˆ 5.6 Hz, 2H; OCH2), 2.40
1.50(m, 36H; AdamantylH and C H2CH2CH2P), 0.77 (s, 9H; SiC(CH3)3),
À0.05 ppm (s, 6H; SiCH3); 13C NMR (CDCl3): d ˆ 60.8, 38.7, 38.0, 36.1,
32.9, 27.5, 25.8, 23.3, 18.3, 12.0, À5.7 ppm; 31P NMR (CDCl3): d ˆ 22.7 ppm.
1
3
Yield 4.6 g (92%); H NMR (CDCl3): d ˆ 7.55 (d, J ˆ 7.6 Hz, 2H; ArH),
3
7.32 (d, J ˆ 7.6 Hz, 2H; ArH), 4.50(s, 2H; ArCH O), 3.8 3.4 (brs, 170
2
180H; PEG), 3.33 (s, 3H; H3COPEG), 2.4 1.6 ppm (m, 30H; AdH); 31P
NMR (CDCl3): d ˆ 23.7 ppm; 1H NMR spectroscopy showed 80 85%
coverage of the polymer with the phosphine.
General procedure for the biphasic Sonogashira coupling of aryl iodides
and acetylenes: To a thoroughly deoxygenated mixture of n-heptane
(10mL), acetonitrile (10mL) and Et 3N (4 mL), were added 1.5 mmol
of the respective aryl iodide, 1.7 mmol of the respective acetylene,
MeOPEG2000OTs (9) and MeOPEG2000OMes (10): A solution of MeOPE-
G
2000OH (20g, 10mmol), triethylamine (2.02 g 20mmol) and the respec-
tive sulfonyl chloride (TsCl, 2.86 g 15 mmol or MesCl 1.72 g 15 mmol) in
dry CH2Cl2 (200 mL) were stirred for 10 h at room temperature. The
reaction mixture was added to a separating funnel containing CH2Cl2
(800 mL) and water (100 mL). The organic layer was washed with water
(2 Â 100 mL), dried over MgSO4 and evaporated to yield a sticky, pale
yellow solid. The solid was stirred with diethyl ether (200 mL) for 2 h,
filtered and washed again with diethyl ether to remove traces of excess
amine and sulfonyl chloride. Drying in a vacuum yielded the desired
products as nearly colourless solids.
MeOPEG2000O(C6H4)PPh2
(70mg,
1 mol%),
(MeOPEG 2000O-
(C6H4)PPh2)2PdCl2 (140mg, 2 mol%) (alternatively a mixture of two
equivalents of the phosphine and [(MeCN)2PdCl2] can be used), CuI
(12 mg, 4 mol%) and K2CO3 (210mg, 1.5 mmol). The mixture was heated
under reflux until the starting materials were consumed (TLC). After
cooling to room temperature the upper layer was separated by means of a
cannula. The solvent was evaporated to yield the crude product. The
reaction vessel was recharged with degassed n-heptane (10mL), triethyl-
amine (2 mL), K2CO3 (210mg, 1.5 mmol) and the two substrates and the
next catalytic cycle was started. After the last cycle, the lower phase was
extracted with n-heptane (3 Â 10mL) to isolate the remaining product. The
crude products from the different runs were combined and purified by
column chromatography on silica (n-heptane) to yield the respective pure
compounds.
MeOPEG2000OTs: Yield 19.6 g (91%): 1H NMR (CDCl3): d ˆ 4.35 4.28
(m, 2H; SO2OCH2), 3.80 3.45 (m, ꢁ170 180H; PEGH), 3.32 (s, 3H;
CH3OPEG), 2.35 ppm (s, 3H; CH3Ar).
1
MeOPEG2000OMes: Yield 17.9 g (86%): H NMR (CDCl3): d ˆ 4.35 4.28
(m, 2H; SO2OCH2), 3.80 3.45 (m, ꢁ170 180H; PEGH), 3.32 (s, 3H;
CH3OPEG), 3.03 ppm (s, 3H; CH3SO2); 1H NMR spectroscopy showed
almost quantitative coverage of the polymer with OTs or OMes groups.
General procedure for the room-temperature Sonogashira coupling of aryl
bromides and acetylenes: HNiPr2 (0.3 mL, 2.1 mmol), [(PhCN)2PdCl2]
(11.5 mg, 2 mol%), CuI (3.8 mg, 1.3 mol%) and the respective phosphine
(4 mol%) were added to THF (2 mL). The mixture was sonicated until all
compounds had dissolved (5 10min). The respective aryl bromide and
acetylene were added and the mixture was stirred at room temperature.
Precipitation of HNiPr2 ¥ HBr occurred after a few minutes. After 12 h, the
reaction mixture was filtered over Celite, evaporated to dryness and
purified by chromatography on silica.
4-(MeOPEG2000OCH2)C6H4CH2Br (12): To a solution of MeOPEG2000OH
(10g, 5 mmol) in THF (50mL) was added NaH (50wt% suspension in
mineral oil, 960mg, 20mmol) and the mixture was stirred for 30min at
room temperature. Solid 1,4-dibromomethylbenzene (10g, 37.8 mmol) was
added and the resulting yellow solution was stirred at room temperature for
another 3 h. After filtration over Celite, the solution was concentrated to
20mL in a vacuum. Diethyl ether (100mL) was added which yielded a pale
yellow precipitate that was collected by filtration. To remove impurities
(mainly excess 1,4-dibromomethylbenzene), the precipitate was suspended
in ether (100 mL), stirred for 1 h and filtered again. This procedure was
repeated 3 4 times until virtually all impurities had been removed (TLC).
After drying in a vacuum, the desired product was obtained as a pale yellow
powder. Yield 9.7 g (89%). 1H NMR spectroscopy showed almost
quantitative etherification of the terminal OH group (>90%).
General procedure for the biphasic Sonogashira coupling of aryl bromides
and acetylenes: To thoroughly deoxygenated DMSO (5 mL) were added
1.5 mmol of the respective halide, 1.8 mmol of the acetylene, 4-(MeOPE-
G2000OCH2)C6H4CH2P(1-Ad)2 ¥ HBr (75 mg, 2 mol%), Na2[PdCl4] (4.5 mg,
1 mol%), CuI (2 mg, 0.7 mol%) and HNiPr2 (0.5 mL, 3.5 mmol). The
mixture was heated to 608C until the starting materials were consumed
(TLC). After the mixture had been allowed to cool to room temperature, n-
heptane (15 mL) was added and the mixture was stirred for 5 min. The
upper layer was removed by means of a cannula and evaporated to yield the
crude product. The reaction vessel was recharged with HNiPr2 (0.5 mL,
3.5 mmol) and the two substrates and another reaction cycle was started.
After the last cycle, the DMSO was extracted with n-heptane (3 Â 10mL)
to isolate the remaining product. The crude products from the different
runs were combined and purified by column chromatography on silica (n-
heptane or cyclohexane/ethyl acetate 4:1) to yield the respective pure
compounds.
4-(MeOPEG2000O)C6H4PPh2 (13): MeOPEG2000OMes (10.4 g 5 mmol) 9,
4-(HO-C6H4)PPh2 (1, 2.09 g 7.5 mmol) and K2CO3 (2.8 g 20.2 mmol) in
CH3CN (100 mL) were heated under reflux for 14 h. The mixture was
cooled to room temperature and filtered over Celite. The solvent was
removed in a vacuum and the brownish residue was dissolved in warm
ethanol (25 mL). The product precipitated upon addition of diethyl ether
(100 mL) and cooling to 48C. This precipitate was collected by filtration,
suspended in diethyl ether (100 mL), stirred for 2 h at room temperature
and filtered again to obtain the pure product as a pale yellow powder, which
1
was dried in a vacuum. Yield: 10.1 g (85%); H NMR (CDCl3): d ˆ 7.54
7.25 (brm, 12H; ArH), 6.93 (d, 3J ˆ 7.9 Hz, 2H; ArH), 4.14, (m, 2H; a-
CH2), 3.9 3.4 (m, ꢁ170 180H; PEGH), 3.34 ppm (s, 3H; CH OPEG);
3
31P NMR (CDCl3): d ˆ À6.9 ppm. 1H NMR spectroscopy showed 90 95%
General procedure for the TOF experiments over five cycles: To a mixture
of Na2[PdCl4] (2.2 mg, 0.5 mol%), RP-OCH2C6H4CH2P(1-Ad)2 (7, 45 mg,
1 mol%) and CuI (1.0mg, 0.4 mol%) in DMSO (3 mL) and HN iPr2
(0.5 mL, 3.6 mmol) held at 608C, were added the respective aryl bromide
(1.5 mmol) and PhCCH (220 mL, 2 mmol). The reaction was stirred for
30min (in the case of 4-bromoacetophenone for 10min) and a 50 mL
sample was taken. This was added to a 0.043m solution of diethylene glycol
dibutyl ether in acetone (1 mL) and examined by GC. After taking the GC
sample, the reaction was reheated and run to completion, then the whole
procedure was repeated.
coverage with the phosphine moiety.
[4-(MeOPEG2000O)C6H4PPh2]2PdCl2 (14): 4-(MeOPEG2000O)C6H4PPh2
(12, 2.76 g, 1 mmol) and [(CH3CN)2PdCl2] (130mg 0.5 mmol) were stirred
in CH2Cl2 (10mL) for 15 min to yield a yellow solution. The solvent was
removed in a vacuum and the sticky residue was treated with diethyl ether
(25 mL) until a fine yellow powder remained. The diethyl ether was
removed by careful decantation and the product was dried in a vacuum.
Yield: 2.80g (99%); 31P NMR spectroscopy showed quantitative con-
version; 1H NMR (CDCl3): d ˆ 7.69 7.52 (m, 6H; ArH), 7.40 7.22 (m, 6H;
Chem. Eur. J. 2003, 9, No. 6
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1423 $ 20.00+.50/0
1423

H. Plenio and A. Kˆllhofer
FULL PAPER
Tetrahedron 2001, 47, 7449; d) I. P. Beletskaya, A. V. Cheprakov,
Chem. Rev. 2000, 100, 3009; e) R. B. Bedford, C. S. J. Cazin, Chem.
Commun. 2001, 1540; f) J. A. Loch, M. A. Albrecht, E. Peris, J. Mata,
Acknowledgement
Support of this work by the DFG through grant Pl-178/8 and the TU
Darmstadt is acknowledged. We thank Dipl.-Ing. J. Saroukh and Prof. Dr.
H. Ortner for the TXRF determinations of Pd and Cu leaching.
J. W. Faller, R. H. Crabtree, Organomet. 2002, 21, 70
0 ; g) S. R.
Chemler, D. Trauner, S. J. Danishefsky, Angew. Chem. 2001, 113,
4676; Angew. Chem. Int. Ed. 2001, 113, 4544; h) G. Y. Li, J. Org. Chem.
2002, 67, 3643; i) K. H. Shaugnessy, R. S. Booth, Org. Lett. 2001, 3,
2757; j) L. J. Goo˚en, J. Paetzold, Angew. Chem. 2002, 114, 1285;
Angew. Chem. Int. Ed. 2002, 41, 1237.
[1] Applied Homogeneous Catalysis with Organometallic Compounds,
2nd edition (Eds.: B. Cornils, W. A. Herrmann), Wiley-VCH, 2002.
[2] Proceedings of the European Workshop Spent Catalysts: Recycling,
Regeneration and Reuse, DECHEMA 2002.
[3] a) N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102, 3345; b) C. A.
McNamara, M. J. Dixon, M. Bradley, Chem. Rev. 2002, 102, 3275;
c) M. E. Davis in Aqueous-Phase Organometallic Chemistry (Eds.: B.
Cornils, W. A. Herrmann), Wiley-VCH, 1998, Chapt. 4.7, p. 241.
[4] H. P. Dijkstra, G. P. M. van Klink, G. van Koten, Acc. Chem. Res. 2002,
35, 798.
[5] a) D. E. Bergbreiter, Chem. Rev. 2002, 102, 3345; b) T. J. Dickerson,
N. N. Reed, K. D. Janda, 2002, 102, 3325; c) A. Behr, C. F‰ngwisch,
Chem. Eng. Technol. 2002, 25, 143.
[6] Thematic Issue: Recoverable Catalysts and Reagents, (Ed.: J. A.
Gladysz), Chem. Rev. 2002, 102, 3215 3892.
[27] Beller and Zapf[20a] calculated that 1 mol% Pd catalyst for
200 Dalton product adds 112 Euros to the cost!
a
[28] R. Akiyama, S. Kobayashi, Angew. Chem. 2001, 113, 3577; Angew.
Chem. Int. Ed. 2001, 40, 3469, and references therein.
[29] S. W. Kim, M. Kim, W, H, Lee, T. Hyeon, J. Am. Chem. Soc. 2002, 124,
7642.
[30] R. G. Heidenreich, K. Kˆhler, J. G. E. Krauter, J. Pietsch, Synlett.
2002, 1118.
[31] K. Kˆhler, R. G. Heidenreich, J. G. E. Krauter, J. Pietsch, Chem. Eur.
J. 2002, 8, 622.
[32] a) J. Moineau, G. Pozzi, S. Quici, D. Sinou, Tetrahedron Lett. 1999, 40,
7683; b) C. Rocaboy, J. A. Gladysz, Org. Lett. 2002, 4, 1993; c) T.
Mizugaki, M. Murata, M. Ooe, K. Ebitani, K. Kaneda, Chem.
Commun. 2002, 52.
[33] D. E. Bergbreiter, P. L Osburn, A. Wilson, E. M. Sink, J. Am. Chem.
Soc. 2000, 122, 9058.
[34] D. E. Bergbreiter, P. L. Osburn, J. D. Frels, J. Am. Chem. Soc. 2001,
123, 11105.
[35] D. E. Bergbreiter, Y.-S. Liu, P. L. Osburn, J. Am. Chem. Soc. 1998, 120,
4250.
[36] D. E. Bergbreiter, P. L. Osburn, Y.-S. Liu, J. Am. Chem. Soc. 1999, 121,
9531.
[37] A. Behr, N. Toslu, Chem. Eng. Technol. 2000, 23, 122.
[38] J. A. Loch, C. Borgmann, R. H. Crabtree, J. Mol. Catal. A: Chem.
2001, 170, 75.
[39] P. Guerreiro, V. Ratovelomanana-Vidal, J.-P. Genet, P. Dellis,
Tetrahedron Lett. 2001, 42, 3423.
[7] a) B. Cornils, Org. Proc. Res.& Dev. 1998, 2, 121; b) C. W. Kohlpaint-
ner, R. W. Fischer, B. Cornils, Appl. Cat. A 2001, 221, 219.
[8] D. Vogt in Aqueous-Phase Organometallic Chemistry (Eds.: B.
Cornils, W. A. Herrmann), Wiley-VCH, 1998, Chapt. 7.1, p. 539.
[9] TPPTS ˆ triphenylphosphine trisulfonate
[10] Reported losses are in the range of 10À9 g per kg of aldehyde.[7]
[11] Z. Lu, E. Lindner, H. A. Mayer, Chem. Rev. 2002, 102, 3543.
[12] Aqueous-Phase Organometallic Chemistry (Eds.: B. Cornils, W. A.
Herrmann), Wiley-VCH, 1998.
[13] R. Fish, Chem. Eur. J. 1999, 5, 1677.
[14] a) S. Schneider, W. Bannwarth, Angew. Chem. 2000, 112, 4293; Angew.
Chem. Int. Ed. 2000, 39, 4142; b) J. J. J. Juliette, D. Rutherford, I. T.
¬
Horvath, J. A. Gladysz, J. Am. Chem. Soc. 1999, 121, 2696; c) E.
[40] A. N. Ajjou, H. Alper, J. Am. Chem. Soc. 1998, 120, 1466.
[41] Z. Jin, X. Zheng, B. Fell, J. Mol. Cat. 1997, 116, 55.
[42] E. Bayer, V. Schurig, Chemtech 1976, 212.
[43] M. Reggelin, Nachr. Chem. Tech. Lab. 1997, 45, 1002, Nachr. Chem.
Tech. Lab. 1997, 45, 1196.
de Wolf, E. A. Speets, B.-J. Deelman, G. van Koten, Organometallics
2001, 20, 3686.
[15] P. Wasserscheid, W. Keim, Angew. Chem. 2000, 112, 3926; Angew.
Chem. Int. Ed. 2000, 39, 3772.
[16] R. Sheldon, Chem. Commun. 2001, 2399.
[44] P. H. Toy, K. Janda, Acc. Chem. Res. 2000, 33, 546.
[45] For example: acetonitrile/n-heptane or DMF/n-heptane or DMSO/n-
heptane.
[46] Recently, thermomorphic catalysts have been reported that utilise
octane solvent for fluorous catalysis: M. Wende, R. Meier, J. A.
Gladysz, J. Am. Chem. Soc. 2001, 123, 11490.
[47] In fact, it is an additive in many supermarket products.
[48] I. B. Champbell in Organocopper Reagents, IRL Press, Oxford, 1994,
p. 217.
[17] P. L. Osburn, D. E. Bergbreiter, Prog. Polym. Sci. 2001, 26, 2015.
[18] M. S. Goedheijt, B. E. Hanson, J. N. H. Reeck, P. C. Kamer,
P. W. N. M. van Leeuwen, J. Am. Chem. Soc. 2000, 122, 1650.
[19] Metal-catalysed Cross-Coupling Reactions (Eds.: F. Diederich, P. J.
Stang) Wiley-VCH, Weinheim 1998.
[20] a) A. Zapf, M. Beller, Chem. Eur. J. 2001, 7, 2908; b) A. Zapf, A.
Ehrentraut, M. Beller, Angew. Chem. 2000, 112, 4315; Angew. Chem.
Int. Ed. 2000, 39, 4153; c) M. Beller, W. M‰gerlein, A. F. Indolese, C.
Fischer, Synth. 2001, 7, 1099.
¬
[49] a) M. Erdelyi, A. Gogoll, J. Org. Chem. 2001, 66, 4165, and references
[21] a) M. H. Ali, S. L. Buchwald, J. Org. Chem. 2001, 66, 2560; b) A.
Aranyos, D. W. Old, A. Kiyomori, J. P. Wolfe, J. P. Sadighi, S. L.
Buchwald, J. Am. Chem. Soc. 1999, 121, 4369; c) J. P. Wolfe, R. P.
Singer, B. H. Yang, S. L. Buchwald, J. Am. Chem. Soc. 1999, 121, 9550.
[22] a) A. F. Littke, G. C. Fu, J. Am. Chem. Soc. 2001, 123, 6989; b) C. Dai,
G. C. Fu, J. Am. Chem. Soc. 2001, 123, 2719; c) M. R. Netherton, C.
Dai, K. Neusch¸tz, G. C. Fu, J. Am. Chem. Soc. 2001, 123, 10099.
[23] a) K. H. Shaughnessy, P. Kim, J. F. Hartwig, J. Am. Chem. Soc. 1999,
121, 2123; b) J. P. Stambuli, S. R. Stauffer, K. H. Shaugnessy, J. F.
Hartwig, J. Am. Chem. Soc. 2001, 123, 2677; c) O. Lˆber, M.
Kawatsura, J. F. Hartwig, J. Am. Chem. Soc. 2001, 123, 4366;
d) A. H. Roy, J. F. Hartwig, J. Am. Chem. Soc. 2001, 123, 1232.
[24] a) W. A. Herrmann, V. P. W. Bˆhm, C.-P. Resinger, J. Organomet.
Chem. 1999, 576, 23; b) C. W. K. Gstˆttmayr, V. P. W. Bˆhm, E.
Herdtweck, M. Grosche, W. A. Herrmann, Angew. Chem. 2002, 114,
1421; Angew. Chem. Int. Ed. 2002, 41, 1363; c) V. P. W. Bˆhm, W. A.
Herrmann, Chem. Eur. J. 2001, 7, 4191.
therein; b) M. R. Eberhard, Z. Wang, C. M. Jensen, Chem. Commun.
2002, 818; c) L. Anastasia, E. Negishi, Org. Lett. 2001, 3, 3111; d) C.
Yang, S. P. Nolan, Organomet. 2002, 21, 1020; e) H. Plenio, J.
Hermann, A. Sehring, Chem. Eur. J. 2000, 6, 1820; f) M. Alami, B.
Crousse, F. Ferri, J. Organomet. Chem. 2001, 624, 114; g) R. A. Batey,
M. Shen, A. J. Lough, Org. Lett. 2002, 4, 1411.
[50] T. Hundertmark, A. F. Littke, S. L. Buchwald, G. C. Fu, Org. Lett.
2000, 2, 1729.
[51] V. P. W. Bˆhm, W. A. Herrmann, Eur. J. Org. Chem. 2000, 3679.
[52] J. P. Stambuli, M. B¸hl, J. F. Hartwig, J. Am. Chem. Soc. 2002, 124,
9346.
[53] Fluka offers MeOPEG in five different molecular weights: 350, 550,
750, 2000 and 5000 Dalton.
[54] O. Herd, A. He˚ler, M. Hingst, M. Tepper, O. Stelzer, J. Organomet.
Chem. 1996, 522, 69.
[55] The closely related compound with MeOPEG-3400 was synthesised
by a multistep procedure: F. Sieber, P. Wentworth, J. D. Toker, A. D.
Wentworth, W. A. Metz, N. N. Reed, K. D. Janda, J. Org. Chem. 1999,
64, 5188 with TBDMS protection of the phenol hydroxyl group.
[56] S. Bogdanovic, H.-J. Kleiner, S. Haber, H. Bahrmann, C.-D. Frohning,
Celanese Chem. Europe, US 6160180.
[25] M. T. Reetz, E. Westermann, Angew. Chem. 2000, 39, 170; Angew.
Chem. Int. Ed. 2000, 39, 165.
[26] a) G. A. Grasa, A. C. Hillier, S. P. Nolan, Org. Lett. 2001, 3, 1077;
b) M. R. Buchmeiser, T. Schareina, R. Kempe, K. Wurst, J. Organo-
met. Chem. 2001, 634, 39; c) N. J. Whitcombe, K. K. Hii, S. E. Gibson,
1424
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1424 $ 20.00+.50/0
Chem. Eur. J. 2003, 9, No. 6

Biphasic Sonogashira Coupling
1416 1425
[57] J. P. Genet, M. Savignac, J. Organomet. Chem. 1999, 576, 305.
[58] K. Sonogashira, Y. Tohda, N. Hagahira, Tetrahedron Lett. 1975, 50,
4467.
[59] W. A. Francis, Critical Solution Temperatures, ACS, Washington 1961.
[60] Leaching of CH3CN and Et3N was detected by 1H NMR spectroscopy
of the upper layer.
[68] The absence of matrix effects on the sensitivity of palladium
determination by TXRF was excluded by verifying successfully the
palladium content of a diluted sample of the DMSO catalyst solution.
[69] D. Martin, H. G. Hauthal, Dimethylsulfoxid, Akademie Verlag, Berlin
1971.
[70] H. Plenio, M. Preu˚, unpublished results.
[61] D. E. Bergbreiter, Y.-S. Liu, Tetrahedron Lett. 1997, 38, 7843.
[62] C. Markert, W. Bannwarth, Helv. Chim. Acta 2002, 85, 1877.
[63] J. P. Stambuli, S. R. Stauffer, K. H. Shaughnessy, J. F. Hartwig, J. Am.
Chem. Soc. 2001, 123, 2677.
[64] J. R. Goerlich, R. Schmutzler, Phos., Sulf. and Sil. 1995, 102, 211.
[65] Several vendors do (while others do not) state that 1-octyne and
1-hexyne contain up to 3% of 1-bromohexane and 1-bromobutane,
respectively.
[71] H. Plenio, M. an der Heiden, unpublished results.
[72] D. D. Perrin, W. L. Armarengo, Purification of Laboratory Chemicals,
Pergamon Press, Oxford 1989.
[73] J. E. Nystrˆm, T. D. McCanna, P. Helquist, R. Amoroux, Synthesis
1988, 9, 56.
[74] J. R. Doyle, P. E. Slade, H. B. Jonassen, Inorg. Synth. 1960, 6, 216.
[75] O. Z. Marzena; V. Wendel, M. Fehlinger, B. Ziemer, W. Abraham,
Eur. J. Org. Chem. 2001, 1549.
[66] J. A. Gladysz, Pure Appl. Chem. 2001, 73, 1319.
[67] Conditions are identical to those of the coupling reactions described in
Table 5.
Received: October 2, 2002 [F4473]
Chem. Eur. J. 2003, 9, No. 6
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
0947-6539/03/0906-1425 $ 20.00+.50/0
1425