Palladium catalysed coupling reactions using guanidinium phosphine complexes on glass beads

Article abstract of DOI:10.1055/s-1999-2907

Supported palladium catalysts generated from guanidinium phosphine 1 and palladium acetate have been shown to exhibit high activity and low leaching of catalyst. The nature of the catalyst allows facile separation and recycling.

Full text of DOI:10.1055/s-1999-2907

LETTER
1645
Palladium Catalysed Coupling Reactions using Guanidinium Phosphine
Complexes on Glass Beads
Mathew P. Leese, Jonathan M. J. Williams*
Department of Chemistry, University of Bath, Claverton Down, Bath, UK BA2 7AY
E-mail: j.m.j.williams@bath.ac.uk
Received 5 August 1999
solution. Removal of solvent after 0.5h stirring gave the
supported catalyst 2 as a free flowing golden yellow pow-
der. Optimal ligand to palladium ratio was dictated by bal-
ancing palladium leaching levels and catalytic activity (a
high ligand to metal ratio suppressed reaction), a 2.8:1
ligand/metal ratio was thus established.
Abstract: Supported palladium catalysts generated from guanidin-
ium phosphine 1 and palladium acetate have been shown to exhibit
high activity and low leaching of catalyst. The nature of the catalyst
allows facile separation and recycling.
Key words: palladium catalysis, glass beads, Heck reaction, Sono-
gashira reaction, guanidinium phosphines
NH2Cl
i) MeOH, 1h
P
Transition metal catalysis continues to be a major research
field, with many powerful and selective transformations
now available to synthetic chemists. Limitations to the
)3
+
Pd(OAc)2
Me2N
ii) CPG-239
1
Supported Pd
Catalyst
use of transition metal catalysts in large scale synthesis lie
in the expense of the catalyst and contamination of the
1
2
product. The development of catalyst systems which com- Scheme 1
bine low leaching levels and facile recycling is thus a sig-
nificant goal.
Preliminary experiments showed that these beads were ef-
fective catalysts for Heck reaction of aryl iodides an alk-
enes. Optimised results from a series of experiments are
given in Table 1.
Supported liquid phase catalysis is an approach which ut-
lilises the hydrophilic surface of a glass bead to adhere a
polar solvent. A polar soluble transition metal complex
can thus reside in this supported layer. This non-covalent
method of support yields a mobile catalyst which acts at
the interface between the supported solvent and the bulk,
apolar solvent in which both substrate and product reside.
On completion of the reaction, the product and catalyst are
separated readily from each other by filtration or decant-
ing.
2
Table 1 Glass bead catalysed Heck Reactions
R1
R1
I
R2
R2
+
1
mol%
3
4
5
Our first attempts to use this approach utilised water solu-
Alkene (1.25 eq.), triethylamine (3 eq.), Toluene, 100 °C
ble complexes of palladium and TPPTS, [P(m-
3
NaO SC H ) ]. Sulfonation of aromatic phosphines can
R1
R2
3
6
4 3
Reaction Isolated Pd leaching
(%)*
however be capricious, with phosphine oxides often
emerging as the major product of reaction, this is an im-
portant practical concern as separation of phosphine and
time (h)
Yield (%)
1
2
3
4
5
6
7
H
H
CO Me
2
16
16
16
48
48
60
60
87
58
65
0.2
t
CO2 Bu
0.1
4
phosphine oxide is non-trivial. The harsh conditions re-
m-Me
CO2Me
0.1
quired for sulfonation may also cause epimerisation of
chiral, non racemic ligands. We thus sought to investigate
the use of alternative water soluble ligands in the reaction,
with mild, late stage introduction of the water soluble
moiety being a prime objective. Guanidinium phosphine
ligands 1 have recently been introduced by Schmidtchen
†
70
H
CH(OH)Me
0.1
†
72
m-Me CH(OH)Me
0.1
H
Ph
55
83
<0.03
0.1
o-Br
CO2Me
*
Palladium level detected in the product as a percentage input into
†
5
the reaction. The isolated product in this case is the tautomeric
and used to form water soluble palladium catalysts. We
ketone.
thus synthesised the ligand in three clean, high yielding
steps from commercially available starting materials and
6
began to explore their use on glass beads.
Attractive features exhibited by these systems are the low
catalyst loading required for successful reaction, with just
We synthesised the supported phase palladium catalyst in
a straightforward manner by precomplexing the phos-
phine 1 and palladium acetate in methanol before adding
1
mol% providing sufficient activity and also the very low
levels of leaching observed.
7
degassed controlled pore glass (CPG-239) to the resultant
Synlett 1999, No. 10, 1645–164 ISSN 0936-5214 © Thieme Stuttgart · New York

1
646
M. P. Leese, J. M. J. Williams
LETTER
The utility of the catalyst beads was further demonstrated zene reacted to give the desired coupling product in 92%
by carrying out recycling experiments. Catalyst beads pre- yield over a period of 60h in the presence of 1 mol% Pd
pared as above (but substituting palladium chloride for beads and 3 mol% CuI, significantly low levels of catalyst
palladium acetate) were subjected to the reaction condi- leaching, in this case (0.1 %), were obtained.10
tions above. At the conclusion of the reaction the bulk sol-
vent was removed by decanting and the reaction vessel
recharged with substrate, solvent and base. As shown in
Table 2 high yields and low leaching were maintained
Finally, an allylic substitution reaction of allyl acetate 6
with dimethylmalonate 7 was achieved using the support-
ed catalyst. The substitution product 8 was formed in rea-
sonable, unoptimised yield and with a low level of
over 4 reaction cycles.
palladium leaching.
For all of the supported liquid phase reactions, we have
Table 2 Recycling Experiments
shown (at least when sulfonated ligands are employed)
that dissociation of the catalyst into the bulk phase takes
place during the course of the reaction, but only to a rela-
CO2Me
I
+
CO2Me
Ph
Ph
1
mol%
11
Alkene (1.25 eq.), triethylamine (3 eq.), Toluene, 100 °C, 16h.
tively small extent. From a practical standpoint, the fact
that after cooling the catalyst is returned to the glass bead,
is adequate for most synthetic purposes.
Isolated
Yield (%)
Pd leaching
Cycle no.
%)*
(
1
2
3
4
85
78
83
81
0.01
0.20
0.05
0.20
OAc
1
mol%
CH(CO2Me)2
6
t
3 eq. [(C H N) P=N Bu
4
8
3
CH2(CO2Me)2
8
Toluene, 16h, 100 °C
*Palladium level detected in the product as a percentage input
into the reaction.
7
56%, 0.3% Pd leaching
Scheme 2
In order to broaden the scope of application of the sup-
ported phase palladium catalyst, Sonogashira couplings of We have demonstrated the use of guanidinium functiona-
8
aryl iodide and acetylenes were attempted. Pleasingly a lised phosphine in supported phase catalyst systems. Of
toluene solution of iodobenzene, phenyl acetylene and di- particular note are the recyclability of catalyst and the low
ethylamine underwent reaction in the presence of our Pd levels of palladium leaching into the product. The guanid-
catalyst and CuI in quantities of 5 and 10 mol% respec- ium moiety affords a mild, late stage introduction of water
tively. The product of reaction, however, proved to be that solubility to amino functionalised ligands a significant
9
of an oxidative dimerisation of the acetylene. A high consideration in derivatisation of non-trivial asymmetric
yielding (96%) dimerisation of phenyl acetylene could ligands, the applications of which we are now pursuing.
thus be effected at room temperature with our catalyst
beads and CuI (1 and 2 mol % respectively) in just 16h
Acknowledgement
with very low Pd leaching levels (<0.01%), this oxidative
process is presumably reliant on the presence of air.
We wish to thank the EPSRC (Clean Synthesis) for support of this
work.
However, successful Sonogashira couplings were realised
by performing the reaction at 60 °C in rigorously degassed
toluene as shown in Table 3. Thus 1-hexyne and iodoben- References and Notes
(
1) W. A. Herrmann, W. A.; Cornils B. Angew. Chem. Int. Ed.
Engl. 1997, 36, 1048.
(
2) For lead references on supported liquid phase catalysis see
(
a) Anson, M. S.; Leese M. P.; Tonks L.; Williams J. M. J. J.
Table 3 Glass bead catalysed Sonogashira couplings
R1 R1
R2
Chem. Soc., Dalton Trans., 1998, 3529 and references therein.
(b) Davis M. E. Chemtech 1992, 498.
R2
I
(
(
(
(
3) Tonks L.; Anson M. S.; Hellgardt K.; Mirza A. R.; Thompson
D. R.; Williams J. M. J.Tetrahedron Lett. 1997, 38, 4319.
4) For a discussion on the use of sulfonated phosphines see see
Cornils B.; Kuntz E. G. J. Organomet. Chem. 1995, 502, 177.
5) Hessler, A; Stelzer, O.; Dibowski, H.; Worm, K.;
Schmidtchen, F. P. J. Org. Chem. 1997, 62, 2362.
+
1
mol%
CuI 3 mol%, alkyne (3 eq.), diethylamine (3 eq.), Toluene, 60 °C
R1
R2
Reaction Isolated Pd leaching
time (h)
Yield (%)
%)*
(
6) Synthesis of the ligand involves Grignard reaction of
commercially available 2-(N,N-bis-trimethylsilyl-
1
2
3
4
H
H
nBu
60
3
87
58
65
70
0.2
0.1
0.1
Ph
Ph
Ph
amino)phenyl magnesium chloride with phosphorus
trichloride. The triaryl phosphine product is then deprotected
at nitrogen in refluxing methanol and transposed to the
hydrochloride salt before a final condensation with dimethyl
cyanamide affords the desired guanadinium phosphine 1.5
H
6
m-Me
14
0.1
*Palladium level detected in the product as a percentage of that input
Synlett 1999, No. 10, 1645–164 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Palladium Catalysed Coupling Reactions using Guanidinium Phosphine Complexes on Glass Beads
1647
(
(
7) CPG beads are commercially available from Cambio, 34 (11) Mirza, A.R.; Anson, M.S.; Hellgardt, K.; Leese, M.P.;
Newnham Road, Cambridge, UK CB3 9ET.
8) Sonogashira, K. In Comprehensive Organometallic Synthesis;
Trost B.M.; Fleming I., Ed.;Pergamon: Oxford, 1991; Vol. 3,
p 521.
Thompson, D.F.; Tonks, L.; Williams, J.M.J. Org. Proc. Res.
Dev. 1998, 2, 325.
(
9) Rossi, R.; Carpita, A.; Bigelli, C. Tetrahedron Lett. 1985, 26,
Article Identifier:
523.
1437-2096,E;1999,0,10,1645,1647,ftx,en;L12499ST.pdf
(
10) The fate of the copper salt has not been established.
Synlett 1999, No. 10, 1645–164 ISSN 0936-5214 © Thieme Stuttgart · New York