Carbonylation reactions of iodoarenes with PAMAM dendrimer-palladium catalysts immobilized on silica

Article abstract of DOI:10.1021/jo020271n

Palladium complexes immobilized onto generation 0-3 PAMAM dendrimers supported on silica were used as catalysts for the carbonylation of iodobenzene in methanol to form methyl benzoate. High yields were obtained and the catalyst was recycled 4-5 times without significant loss of activity. The carbonylation reaction was found to be applicable to a variety of iodoarenes regardless of the nature of the substituent.

Full text of DOI:10.1021/jo020271n

Ca r bon yla tion Rea ction s of Iod oa r en es w ith P AMAM
Den d r im er -P a lla d iu m Ca ta lysts Im m obilized on Silica
Shlomo Antebi, Prabhat Arya,† Leo E. Manzer,‡ and Howard Alper*
Centre for Catalysis Research and Innovation, Department of Chemistry, University of Ottawa,
10 Marie Curie, Ottawa, Ontario K1N 6N5, Canada, Steacie Institute for Molecular Sciences, National
Research Council of Canada, 100 Sussex Dr., Ottawa, Ontario K1A 0R6, Canada, and DuPont Central
Research & Development, Experimental Station, Wilmington, Delaware 19880-0262
halper@uottawa.ca
Received April 17, 2002
Palladium complexes immobilized onto generation 0-3 PAMAM dendrimers supported on silica
were used as catalysts for the carbonylation of iodobenzene in methanol to form methyl benzoate.
High yields were obtained and the catalyst was recycled 4-5 times without significant loss of
activity. The carbonylation reaction was found to be applicable to a variety of iodoarenes regardless
of the nature of the substituent.
In tr od u ction
Different strategies have been successfully pursued to
overcome the low reactivity of chloroarenes toward oxida-
tive addition. Milstein et al.6 described the use of Pd
complexes containing electron-rich and bulky chelating
diphosphines. Also, Osborn et al.7 and Alper and Grushin8
demonstrated that Pd complexes with phosphines pos-
sessing both high basicity (pKa > 6.5) and substantial
steric hindrance are useful catalysts (e.g. (Cy3P)2PdCl2)
for the carbonylation of chlorobenzene under mild condi-
tions. Other methods of functionalizing chloroaromatics
include photochemical activation9 or the use of tri-
carbonyl(chloroarene)chromium complexes which react
in a stoichiometric fashion.10
Heterogeneous catalysts have advantages of heat
stability and ease of separation from the product when
compared with homogeneous catalysts. Application of
solid catalysts for the carbonylation of haloarenes has
been examined by the use of NiCl2 on pumice11 or on
SiO2.12 A study on the use of solid catalysts for the
carbonylation of haloarenes concluded that Pd supported
on charcoal was the most active system.13
The carbonylation of haloarenes has been extensively
investigated in the liquid phase.1 Most of the work
reported in this area has involved bromo- and iodoarenes
as starting materials, and metal complexes as catalysts
which are capable of oxidative addition to a C-X bond
(e.g. Pd or Rh). Phosphine-modified palladium complexes
such as dichlorobis(triphenylphosphine)palladium(II) or
bromoarylbis(triphenylphosphine)palladium(II) are cata-
lytically active for the carbonylation of iodobenzene or
bromobenzene to produce benzoic acid derivatives.2 The
carbonylation of haloarenes with Pd catalysts and phase
transfer systems, (e.g., poly(ethylene glycol) and water-
soluble catalysts3) gave satisfactory yields of benzoic
acids. The carbonylation reaction can proceed in a dif-
ferent manner if the intermediate is intercepted by
another species. An example is the novel approach for
the one-pot preparation of R-amino amides by Pd-
catalyzed double carbohydroamination of haloarenes.4
The preparation of polyesters and polyamides by pal-
ladium-catalyzed alkoxycarbonylation of haloarenes5 is
also noteworthy.
Transition metal catalysis based on functionalized
dendrimers has become an interesting and promising way
of using a heterogeneous catalytic system. From a
catalytic point of view the ideal catalyst should be highly
active and selective under mild conditions, stable and
easy to separate from the product by using a simple
* Address correspondence to this author at the University of Ottawa.
Phone: 613-562-5189. Fax: 613-562-5871.
† National Research Council of Canada.
‡ DuPont Central Research & Development, Experimental Station.
(1) (a) Beller, M.; Cornils, B.; Frohning, C. D.; Kohlpaintner, C. W.
J . Mol. Catal. 1995, 104, 17. (b) Beller, M. In Applied Homogeneous
Catalysis with Organometallic Compounds; Cornils, B., Herrmann, W.
A., Eds.; VCH: New York, 1996; Vol. 1.
(6) Ben David, Y.; Portnoy, M.; Milstein, D. J . Am. Chem Soc. 1989,
111, 8742.
(7) Huser, M.; Youinou, M. T.; Osborn, J . A. Angew. Chem., Int. Ed.
1989, 28, 1427.
(8) (a) Grushin, V. V.; Alper, H. J . Chem. Soc., Chem. Commun.
1992, 611. (b) Grushin, V. V.; Alper H. Chem. Rev. 1994, 94, 1047.
(9) Kashimura, T.; Kudo, K.; Mori, S.; Sugita, N. Chem. Lett. 1986,
851.
(2) (a) Stille, J . K.; Wong, P. K. J . Org. Chem. 1975, 40, 532. (b).
Ozawa, F.; Kawasaki, N.; Okamoto, H.; Yamamoto, T.; Yamamoto, A.
Organometallics 1987, 6, 1640. (c). Schoenberg, A.; Heck, R. F. J . Org.
Chem. 1974, 39, 3318. (d) Hidai, M.; Hikita, T.; Wada, Y.; Fujikura,
Y.; Uchida, Y. Bull. Chem. Soc. J pn. 1975, 48, 2075.
(3) (a) Huang, X.; Wu, J . L. Chem. Ind. 1990, 548. (b) Monteil, F.;
Kalck, P. J . Organomet. Chem. 1994, 482, 45.
(4) Lin, Y. S.; Alper, H. Angew. Chem., Int. Ed. 2001, 40, 779 and
references therein.
(5) (a) Yoneyama, M.; Kakimoto, M.; Imai, Y. Macromolecules 1989,
22, 2593. (b). Perry, R. J .; Turner, S. R.; Blevins, R. W. Macromolecules
1993, 26, 1509. (c) Kubota, Y.; Takeuchi, K.; Hanaoka, T.; Sugi, Y.
Bull. Chem. Soc. J pn. 1994, 67, 563.
(10) Mutin, R.; Lucas, C.; Thivolle-Cazat, J .; Dufaud, V.; Dany, F.;
Basset, J . M. J . Chem. Soc., Chem. Commun. 1988, 896.
(11) Dieterle H.; Eschenbasch, W. Arch. Pharm. 1927, 265, 187.
(12) Palit, S. K.; Tripathy, N. J . Appl. Chem. 1969, 19, 301.
(13) Maeda, K.; Yagita, H.; Omata, K.; Fujioto, K. J . Mol. Catal.
1992, 71, 347.
10.1021/jo020271n CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/28/2002
J . Org. Chem. 2002, 67, 6623-6631
6623

Antebi et al.
F IGURE 1. Structures of generation 0-4 of PAMAM-SiO2 dendrimers.
process such as filtration.14 Dendrimer metal catalysts
can, in principle, fill the gap between homogeneous and
heterogeneous systems, due to their monodisperse struc-
ture and multiarm topology that give the peripheral
catalytic sites the ability to act as soluble supported
catalysts.
drimer catalyst was used for the coupling of bromoben-
zene and styrene to form stilbene. The Heck reaction of
bromoarenes and styrene was also investigated by using
a Pd catalyst in which the metal was complexed to
phosphines of a PAMAM dendrimer supported on silica
gel.16b
PAMAM (polyaminoamido) dendrimers supported on
silica, with their globular shape, make them more suit-
able for recycling than other soluble polymer-supported
catalysts. They can be easily removed from the reaction
mixture by membrane or nanofiltration techniques be-
cause of their large size compared to the products.
Several groups prepared systems that were functional-
ized with phosphine ligands at the periphery of the
dendrimers. Reetz et al.15 used commercially available
DAB-dendr(NH2)2 dendrimer as a source of dendrimer-
N(CH2PPh2)2 groups. The latter function as bidentate
ligands when complexed with Pd(tmeda)Me2. The den-
The hydroformylation of styrene was realized by using
a bis-diphenylphosphine terminated PAMAM dendrimer
supported on SiO2, complexed to Rh[(CO)2Cl]2.16a,c A solid
phase and biomimetic approach to obtain catalytic den-
drimer ligands attached to polystyrene beads was em-
ployed by Alper and co-workers.16d The rhodium com-
plexes of these dendrimers were excellent catalysts for
the hydroformylation of several olefins.
We now describe the carbonylation of haloarenes using
catalytic quantities of Pd-PPh2-PAMAM dendrimer on
(16) (a) Bourque, S. C.; Maltais, F.; Xiao, W. J .; Tardif, O.; Alper,
H.; Arya, P.; Manzer, L. E. J . Am. Chem. Soc. 1999, 121, 3035 and
references therein. (b) Alper, H.; Arya, P.; Bourque, S. C.; J efferson,
G. R.; Manzer, L. E. J . Can. Chem. 2000, 78, 920. (c) Arya, P.; Panda,
G.; Rao, N. V.; Alper, H.; Bourque, S. C.; Manzer, L. E. J . Am. Chem.
Soc. 2001, 123, 2889. (d) Arya, P.; Rao, N. V.; Singkhonrat, J .; Alper,
H.; Bourque S. C.; Manzer, L. E. J . Org. Chem. 2000, 65, 1881.
(14) Oosterom, G. E.; Reek, J . N. H.; Kamer, P. C. J .; van Leeuven,
P. W. N. M. Angew. Chem., Int. Ed. 2001, 40, 1828.
(15) Reetz, M. T.; Lohmer, G.; Schwickardi, R. Angew. Chem., Int.
Ed. 1997, 36, 1526.
6624 J . Org. Chem., Vol. 67, No. 19, 2002

Carbonylation Reactions of Iodoarenes
F IGURE 2. Proposed structures of generation 0-2 of PdCl2-PPh2-PAMAM-SiO2 dendrimers.
silica. It was of interest to compare these results with
other heterogeneous methods for the carbonylation reac-
tion.
with either Pd(RCN)2Cl2 (R ) Me, Ph) or Pd(tmeda)Me2
according to literature procedures.16c,17 The complexation
was usually carried out in toluene, after mixing the
dendrimer and the palladium compound for 1 h to
overnight, at room temperature. The Pd dendrimer
complexes generally showed three types of chemical 31P
shifts (Table 1): (a) at 9.0-9.4 ppm for the PdMe2
complexes (comparing well with the previously published
value of 9.7 ppm);16c (b) at 12.2-13.8 ppm for the PdCl2
complexes; and (c) and at 28.0-28.5 for the PPh3 or dba
complexes.
Resu lts a n d Discu ssion
P r ep a r a tion of P AMAM-SiO2 Den d r im er s. Com-
mercial aminopropylsilica gel with aminopropyl groups
(0.9 mmol/g ( 0.1) was used to prepare generations 0-3
of polyaminoamido (PAMAM) dendrimers (Figure 1),
using a previously described method.16a
P r ep a r a tion of P h osp h on a ted P AMAM-SiO2 Den -
d r im er s. The dendrimers on silica were phosphonated
with PPh2H to provide a coordination site for palladium.
Double phosphinomethylation of each terminal amine
was carried out with diphenyphosphinomethanol pre-
pared in situ from paraformaldehyde and diphenylphos-
phine. The detailed preparation was described before16a
and based on the procedure by Reetz.15 The resulting
phosphonylated dendrimers (Figure 2) were character-
ized by 31P solid-state NMR. A chemical shift of -27 ppm
for the various generations compared well with the
published value of -28 ppm for the homogeneous polyami-
nophosphonated dendrimer.15,16a When dicyclohexylphos-
phine was used for the phosphinomethylation reaction,
the chemical shift of the free phosphine groups on the
dendrimer was observed at -28.3 ppm.16a
When the reaction time was reduced to 1.0-2.5 h
(entries 1, 13) with Pd(PhCN)2Cl2 as the Pd reactant,
chemical shifts at 13.2 and 13.8 ppm were observed in
the 31P NMR spectrum (in accord with literature for non-
oxidized, complexed PPh2 groups). When the ratio be-
tween the phosphinated dendrimer and the palladium
compound was 4:1 (entry 10 in Table 1), after a reaction
time of 2.5 h, the chemical shift in the 31P NMR was at
8.6 ppm (with some oxide formed). In the preparation of
the G3 dendrimer, using Pd(PhCN)2Cl2 and a reaction
time of 1.5 h, there was little complexation as indicated
by the 31P solid-state NMR spectrum. The major signal
was at -28 ppm, accompanied by one at 13.8 ppm. These
observations are in accordance with the low (0.8%) Pd
loading, which is 6.5% of the expected theoretical value
(ICP, Table 1, entry 12). When the reaction was repeated
Com p lexa tion of th e P h osp h in o-P AMAM Den -
d r im er Su p p or ted on SiO2 w ith P d . The phospho-
nated dendrimers (i.e. with PPh2) were complexed to Pd
(17) Mizugaki, T.; Ooe, M.; Ebitani, K.; Kaneda, K. J . Mol. Catal.
1999, 145, 329.
J . Org. Chem, Vol. 67, No. 19, 2002 6625

Antebi et al.
TABLE 1. Solid -Sta te 31P NMR a n d ICP Resu lts in th e P r ep a r a tion of P a lla d iu m Ca ta lysts on P AMAM Su p p or ted on
SiO2
conditions for preparation
(rt; argon)
31P-solid
NMR δ
(ppm)c
theoretical
amounts
% Pd
catalyst
no.
complexing agent
A
ICP results
% Pd
yield
% Pd
dendrimer catalysta
Bb
solvent
time, h
1
2
3
4
5
6
7
8
9
10
11
12
13
G1-C2-PPh2-PdCl2
G1-C2-PCycl2-PdCl2
G1-C2-PPh2-PdCl2
G0-C2-PPh2-PdMe2
G1-C2-PPh2-PdMe2
G1-C2-PPh2-Pd(PPh3)2
“G1-C2-PPh2-Pd(dba)2”
G1-C2-PPh2-PdCl2
G1-C2-PPh2-PdCl2
G2-C2-PPh2-PdCl2
G2-C2-PPh2-PdCl2
G3-C2-PPh2-PdCl2
G3-C2-PPh2-PdCl2
Pd(PhCN)2Cl2
Pd(PhCN)2Cl2
Pd(MeCN)2Cl2
Pd(tmeda)Me2
Pd(tmeda)Me2
Pd(PPh3)4
Pd(dba)3‚CHCl3
Pd(MeCN)2Cl2
Pd(MeCN)2Cl2
Pd(PhCN)2Cl2
Pd(PhCN)2Cl2
Pd(PhCN)2Cl2
Pd(PhCN)2Cl2
2:1
2:1
2:1
2:1
2:1
2:1
2.5:1
2:1
4:1
4:1
toluene
toluene
toluene
benzene
toluene
toluene
toluene
benzene
toluene
toluene
toluene
toluene
toluene
1
1
13.2
13.7
12.2
9.0
9.4
28
28.5
12.4
12.8
8.6
8.9
13.8
13.8
5.6
6.7
8.8
5.5
2.0
1.1
6.5
8.3
5.0
2.7
4.0
0.8
3.1
8.7
8.7
8.7
6.5
8.9
6.3
9.1
8.7
4.9
65
77
100
85
22
17
19
19
19
19
19
19
19
2.5
2.5
1.5
19
71
96
102
46
38.5
6.5
26.5
5.2
2:1
2:1
2:1
10.4
11.7
11.7
a
Gn: n designates the generation number. Cm: m designates the number of carbons in the diamino linker, i.e., m - 2 in NH2(CH2)2NH2.
b
PPh2, Pcycl2 (cyclohexyl) are the phosphine groups to which the Pd is complexed. B is the molar ratio between the no. of equivalents of
PPh2 in PPh2-PAMAM-dendrimer on SiO2 to the no. of moles of Pd (A) used. c Signals for oxidized phosphorus were sometimes observed
in the 31P spectra. Mixture of recycled catalysts from entries 8 and 10 from the carbonylation reactions of PhI.
d
TABLE 2. Com p a r ison of Th eor etica l P er cen t P d
Loa d in g in P d Cl2-P P h 2-P AMAM-SiO2 in Gen er a tion s 0, 1,
2, 3, a n d 4
content. A Perkin-Elmer Optima 3000 Radial View ICP-
AES (Inductively Coupled Plasma-Atomic Emission Spec-
trometer) was used for ICP analysis. Measurements were
performed at 340.458 nm with Pd standard in 1 M HCl,
using external calibration.
% growth of Pd loading
theoretical %
Pd loading
Gn/G0a
Gn/G1b
G(n+1)/Gnc
The ICP results for Pd shed some light on the Pd
loading in the various dendritic catalysts, prepared under
different conditions (Table 1). When the complexation
process was effected overnight with Pd(RCN)2Cl2 (R )
Me, Ph) as the Pd source, Pd immobilization was near
quantitative (entries 3, 4, 8, and 9). Exceptions occurred
with either Pd(tmeda)Me2 or Pd(PPh3)4 (entries 5 and 6).
G0
G1
G2
G3
G4
6.3
8.6
10.4
11.7
12.4
136
165
185
196
136
121
112
106
121
136
144
a
The ratio indicates the increase of % Pd loading for each
generation with reference to the 0 generation of PdCl2-PPh2 APS
b
(aminopropylsilica). The ratio indicates the increase of % Pd
loading with reference to the first generation of the PdCl2-PPh2-
PAMAM on SiO2. c The ratio indicates the % Pd loading between
two subsequent generations of Gn-PPh2-PdCl2.
Use of short reaction times (between 1 and 2.5 h) did
not give full complexation of the metal to the dendrimer.
However, these conditions allow the isolation of the
complexed Pd-PPh2 with little if any oxidation at the
phosphorus.
overnight at rt, there was a slight increase of the signal
at 13.8 ppm in the 31P NMR spectrum (assigned to the
coordinated phosphine-Pd bond). In this case, 3.1% Pd
loading occurred, which is 26% of the expected theoretical
value (as determined by ICP Table 1, entry 13).
Ca ta lytic Ca r bon yla tion of Iod oben zen e w ith
P d Cl2 Com p lexed to P P h 2-P AMAM Den d r im er s
Su p p or ted on Silica . The carbonylation of iodoarenes
compounds 1 to the corresponding methyl esters, 2, was
studied initially with iodobenzene (R ) H) as the model
reactant (Scheme 1).
The structure of the dendrimer complex plays a role
in the catalytic system of the PAMAM dendrimers. As
one proceeds to higher generations of dendrimer, the
percentage of Pd loading does not increase by the same
factor compared to the growth of the number of equiva-
lents of the amine and phosphine. Thus when going to a
higher generation the number of equivalents is multiplied
by a factor of 2 for the amine equivalents, and by a factor
of 4 for the phosphine equivalents (Figures 1 and 2).
However, the calculated percentage of the Pd loading
increases insignificantly (only 6%) between generation
3 and generation 4 (for a PdCl2 dendrimer complex; Table
2). In terms of actual loading, while 100% Pd loading was
observed for generations 0 and 1, the loading was 26-
40% (Table 1) for generations 2 and 3.
Based on these findings, the fourth generation of the
dendrimer and its Pd complex were not prepared.
ICP An a lysis of th e Com p lexes. Samples of the
catalyst complexes were digested in aqua-regia (HCl-
HNO3; 3:1) and boiled for 30 min. The samples were
diluted to 25 mL with H2O and were examined for Pd
1
The reactions were monitored by H NMR (tetrachlo-
roethane (TCE) used as reference) and by GC (area %)
analyses. Isolation of the reaction products by liquid
chromatography showed good correlation between the
two methods. This allowed us to monitor the reactions
using GC (area %).
The Pd-dendrimer catalyst (entry 8, Table 2)-G1-PPh2-
PdCl2 (2:1) was used for the first series of carbonylation
reactions of iodobenzene. The reaction in MeOH was
completed overnight, when a pressure of 600 psi of CO
was applied and in the presence of an excess of Et3N as
the HI acceptor (Table 3, entry 1). This result was
comparable with the reaction where PdCl2(MeCN)2 was
used alone (entry 3) or with the precursor of catalyst 8
(entry 2). When K2CO3 was used as the base (entries 4
and 5), the reaction was complete only at the higher
temperature of 105 °C. The time required for the comple-
tion of the reaction was 6 h even at lower CO pressure
(100 psi; entries 6-8).
6626 J . Org. Chem., Vol. 67, No. 19, 2002

Carbonylation Reactions of Iodoarenes
SCHEME 1. Ca r bon yla tion of Iod oa r en es to Th eir Meth yl Ester s
TABLE 3. Scr een in g Exp er im en ts w ith G1-P P h 2-P d Cl2 (ca ta lyst en tr y 8, Ta ble 1)a
base
yieldb
PhCOOMe,
dendrimer-
PdCl2,
mg (G1) 2:1
dendrimer-
CH2PPh2,
mg
Et3N,
mmol
temp,
°C
CO,
psi
time,
h
PdCl2(MeCN)2,
mg
Ph-I,
%
K2CO3
%
1
2
3
4
5
6
7
8
21.5
21.5
21.5
95
105
105
60
105
105
105
105
600
600
600
300
300
100
100
100
22
23
23
22.5
22.5
2
25
17
73
87
100
42
94
72
7.4
7.4
25
2.2
2.2
25
25
25
25
25
29
4.3
4.3
4.3
22
12
5
6
80
90
a
Conditions: PhI (2 mmol); MeOH (6 mL); catalyst (entry 8, Table 1; based on 2:1 G1-PPh2:Pd); b 1H NMR yields.
TABLE 4. Scr een in g Exp er im en ts w ith G1-P P h 2-P d Cl2 (ca ta lyst en tr y 9, Ta ble 1)a ,c,d
time,
h
temp, °C
CO, psi
dendrimer-PdCl2, mg (G1-PPh2) (4:1)
% Ph-I
% PhCOOMeb
remarks
1
2
3
4
5
6
7
8
9
105
105
100
85
100
85
50
50
50
50
50
600
d
16
16
19
19
23
23
23
23
8
25
25
25
25
25
25
28
50
72
95
100
38
36
100
6
92
n.d.
n.d.
n.d.
38
70
105
105
105
95
64
toluene:MeOH 10:1d
Pd/C; 70 mg
no catalyst
94
a
Conditions: PhI (2 mmol); MeOH (6 mL); Et3N (4.3 mmol); catalyst (from entry 10, Table 1; based on 4:1 G1-PPh2:Pd). b 1H NMR
yields. c Use of n-propanol instead of MeOH. By volume.
d
One other catalyst system (system 10, Table 1), having
more vacant PPh2 sites (4:1), gave similar results to those
obtained with the catalyst having fully occupied PPh2 site
(Table 3), and the results are presented in Table 4. The
data compared well even with a theoretically lower
loading of Pd, and with lower CO pressures. Carbonyla-
tion was complete after 8 h when CO was bubbled into
refluxing n-propanol (Table 4; entry 9). The carbonylation
reaction did not proceed without any catalyst (entry 8).
The reaction depends on the temperature. Low conver-
sion of iodobenzene occurred at 60 °C, while temperatures
higher than 85 °C afforded good conversions and yields.
Two molar equivalents of Et3N or K2CO3 was sufficient
to attain complete conversion of iodobenzene to the ester.
Although carbonylation was accomplished under atmo-
spheric pressure (CO bubbling), for ease of handling,
subsequent experiments were carried out in an autoclave
with use of 100 psi of CO.
effect by substituents on the phenyl ring (Table 5). It was
also found that there was no difference between electron-
withdrawing substituents (nitro, ester) and electron-
donating groups (methoxy, methyl).
To determine the difference in the activities of iodoare-
nes with an electron-donating and with an electron-
withdrawing group, reactions were effected for 5 instead
of 19 h. With the electron-withdrawing p-CF3 substituent,
a 73% yield (entry 6; GC, area %) of the ester was
obtained, whereas for the p-OMe analogue, the product
yield was 90% (entry 13). With o-MeO-C6H4I, only 63%
of the ester was obtained, probably because of steric
interactions (entry 5).
When bromobenzene was used as the starting material
the reaction did not take place at 105 °C, nor at higher
temperatures (e.g. 150 °C, KI), The addition of dppp [1,3-
bis(diphenyphospino)propane] or dppb (1,4-bis(diphe-
nylphosphino)butane) to the reaction mixture did not
result in the activation of the C-Br bond toward the
carbonylation reaction. Nonetheless, this lack of activity
can be used to advantage for the preparation of deriva-
tives of methylbenzoate esters, by employing bromoiodo-
arenes as starting materials.
Gen er a lity of th e P d -P P h 2-Den d r itic System for
th e Ca r bom eth oxyla tion of Iod oa r en es. The scope of
the catalytic system for the carbomethoxylation of io-
doarenes was examined by overnight reactions with use
of 1.5 mmol of ArI, 0.4 mL of Et3N, 100 psi of CO, and 5
mL of MeOH, at 100 °C. All reactions went to completion
under these conditions, and there was no significant
In flu en ce of th e Na tu r e of th e P h osp h in e on th e
Ca ta lytic P r ocess. Two PdCl2-phosphinated dendrimers
J . Org. Chem, Vol. 67, No. 19, 2002 6627

Antebi et al.
TABLE 5. Ca r bon yla tion of Iod oa r en es (R-P h I) w ith G1-P R12-P d Cl2 (R1 ) P h or Cycl)a
catalyst;
recovered
catalyst /mg
spectral datai
IR vis; MS m/e
%
ArI
%
R
R1)/mg
ArCOOMeb
1
2
H
H
2; cyclf/17.7
3; Phg/16.1
3; Phg/16.9
3; Phg/16.1
3; Phg/17.1
3; Phg/16.8
3; Phg/22.7
1; Phh/16.6
1; Phh/17.7
1; Phh/16.9
1; Phh/16.5
1; Phh/16.4
1; Phh/18
14.4
11.3
13.3
11.3
10
1724 cm-1
136 m/e
100
1724 cm-1
136 m/e
100
3
4-NO2
Me
1719 cm-1
181 m/e
81c
4
1721 cm-1
150 m/e
100
5
2-OMe
4-CF3
4-I
1727 cm-1
166 m/e
100 (63k)
100 (73k)
100
6
10.2
e
1730 cm-1
204 m/e
7
1723 cm-1
194 m/e
8
3-OMe
4′-I-Phd
4-OH
11.2
e
1722 cm-1
166 m/e
14
86
9
1722 cm-1
270 m/e
100
10
11
12
13
8.9
10.4
7.5
12
1711 cm-1
152 m/e
100
4-Br
1719 cm-1
214 m/e
2
98
CH3CO-
4-MeO-
1679, 1722 cm-1
178 m/e
100
10
90k
a
ArI (1.5 mmol); Et3N (0.4 mL), MeOH (5 mL); CO (100 psi); 100 °C; overnight (∼19 h), unless stated otherwise. b 1H NMR yields and
GC (area %). c 19% methyl p-aminobenzoate; as a result of the reduction of the NO2 function. 4,4′-Diiodobiphenyl (0.75 mmol), Et3N (0.4
d
mL); MeOH (5 mL). e The product was insoluble in MeOH. f R1 ) cyclohexyl (catalyst 2 from entry 2, Table 1). R1) phenyl (catalyst 3
g
from entry 3 Table 1). R1) phenyl (catalyst 3 from entry 4 Table 1). i 1H NMR spectra were in accordance with the structure of the
h
k
compounds. Reaction time 5 h.
TABLE 6. Recyclin g of G1-P R21-P d Cl2-(R1 ) P h or Cycl)
there was actually no significant difference in behavior
of the two catalytic systems. Both catalysts gave complete
in th e Ca r bon yla tion of Iod oben zen ea
b
conversion of iodobenzene to its ester and we were
pleased that both catalysts are capable of being recycled
at least 4-5 times. The reduction in the yield in the last
cycles can be attributed to the decreasing amount of
recycled catalyst.
Com p a r ison of Activities of P d -P h osp h in a ted
Den d r im er s of Differ en t Gen er a tion s, a n d fr om
Differ en t P d Sou r ces. The activity of different Pd-
dendritic complexes was compared for the methoxycar-
bonylation of iodobenzenes. To examine relative catalytic
activities, the reaction time was decreased to 5 h (Table
7).
A. G1-PCycl2-PdCl2
catalyst/ recovered
entry
mg
catalyst/mg recycle no. % PhI % PhCOOM
1
2
3
4
5
16.8b
17.7b
25c;
12.9
14.4
21.4
16.2
13.2
8.3
0
0
1
2
3
4
5
100
100
95
98
71
5
2
29
27
24
21.4d
16.2d
13.2d
7.2d
6
73
76
7e
f
B. G1-PPh2-PdCl2
catalyst/ recovered
entry
mg
atalyst /mg recycle no. % ArI % ArCOOM
(a) Comparison of G0 and G1 complexed with Pd-
(tmeda)Me2 shows that G0 gave a higher yield of the ester
(89%) than G1 (74% yield; Table 7; entries , and 2).
However, it must be noted that in G1 the actual Pd
loading in catalyst 5 was only 3.2 µmol of Pd compared
to 8.8 µmol in catalyst 4.
(b) G1 complexed to PdCl2 gave a 91% yield of the ester
after 5 h (entry 3). The catalyst was isolated as a black
solid. However, when the reaction was repeated with
dppb (entry 4), a yellow catalyst was isolated and 100%
of the ester was formed. Recycling this yellow catalyst
without dppb (entry 5) gave 60% of the ester. It was thus
suspected that the recycled catalyst did not contain Pd
because dppb extracted the Pd into the solution. Never-
theless, recycling (entry 6) the catalyst again (entry 5)
in the presence of dppb increased the yield to 83%.
(c) There was no significant change in the activity
within the same generation for dendrimer catalysts
prepared with different complexation ratios between the
phosphinated dendrimer and the Pd compound, i.e., 2:1
1
2
3
4
5
6
7
16.9f
16.1f
16.7f
25g
9.4
11.3
10.4
22.8
16.8
15.4
0
0
0
1
2
3
4
100
100
100
92
77
71
8
23
29
24
22h
16.8h
15.4h
76
a
All reactions were run for 19 h with 1.5 mmol of PhI, 5 mL of
b
MeOH, 0.4 mL of Et3N, and 100 psi of CO at 100 °C. Catalyst
no. 2, entry 2 in Table 1. c Isolated catalysts from previous cycle
(entries 1 + 2). Isolated catalysts from previous cycle. e 1 mmol
d
of PhI was used. f Catalyst no. 3, entry 3 in Table 1. Isolated
g
d
catalysts from previous cycle (entries 1 + 2 + 3). Isolated
catalysts from previous cycle.
on silica, 2 [based on P(cyclohexyl)2] and 3 (based on
PPh2), were compared for their activities for the carbon-
ylation of iodobenzene (Table 6). Despite the more basic
character, and larger steric bulk, of the dicyclohexylphos-
phine group as compared with that of diphenylphosphine,
6628 J . Org. Chem., Vol. 67, No. 19, 2002

Carbonylation Reactions of Iodoarenes
TABLE 7. Yield s of Meth yl Ben zoa te fr om Iod oben zen e
w ith Differ en t Den d r itic P a lla d iu m Ca ta lystsa
was 71%, entry 20), while the yield of ester was 36% for
a G1 catalyst obtained from Pd(PPh3)4 (17% of the
theoretical Pd, entry 21).
In conclusion, when dendrimers of higher generations
(2 and 3) are used, a lower yield is observed, probably
because of lower Pd loading obtained in the higher
generations.
(g) The TONs obtained are usually around 100 (Table
8) and were comparable to other nondendrimeric Pd
catalysts (120 for Pd/C and 99 for Pd(PhCN)2Cl2 (entries
17 and 18)).
time,
h
%
%
cat.
type
PhCOOMe PhI
1
2
4
5
G0-C2-PPh2-PdMe2
G1-C2-PPh2-PdMe2
5
5
89
74
11
26
3
4
5
8
8
G1-C2-PPh2-PdCl2 no dppb
G1-C2-PPh2-PdCl2 +dppb
1st recycled, 23 mg of 4,
no dppb
2nd recycle 18 mg of 5
+ 20 mg dppb
5
5
5
91
100
60
9
40
17
6
5
83
A TON of 176 (and 89% yield) resulted for the dendritic
G0 complexed to “PdMe2” (entry 12), with nearly identical
data obtained for a G1 catalyst (entry 2) when the ratio
between the phosphonated dendrimer and the Pd com-
pound was 4:1 (leaving vacant PPh2 coordination sites).
Catalyst 9 (entry 2) gave 91% yield versus 88% yield for
catalyst 8 (entry 1), despite the fact that the Pd loading
was less (50%). A similar case is observed for G2 catalysts
where 67% of the ester was obtained for catalyst no. 10
with a TON of 236 (entry 7) compared to 73% yield and
a TON of 179 (entry 8).
A notable TON of 358 was obtained for catalyst no. 5
with a good yield of 74% (entry 13) despite the very low
Pd loading (0.2 mol % Pd, where the ratio substrate/Pd
was 500:1). Another catalyst, no. 6 (entry 14), seemed to
be of comparable activity to catalyst no. 5. It showed a
TON of 318 (with 0.1 mol % of Pd, or substrate/Pd
1000:1). However, repeating a second reaction with 3.6
times the amount of catalyst increased the yield from 36%
to 60%, whereas the TON dropped to 145.
7
8
9
1
3
2
2
G1-C2-PPh2-PdCl2
G1-C2-PPh2-PdCl2
G1-C2-PCy2-PdCl2
G1-C2-PCy2-PdCl2
5
19
5
86
100
87
14
13
10
19
100
11
12
8
9
G1-C2-PPh2-PdCl2 (2:1)
G1-C2-PPh2-PdCl2 (4:1)
5
5
88
91
12
9
13
14
10 G2-C2-PPh2-PdCl2 (4:1)
11 G2-C2-PPh2-PdCl2 (2:1)
5
5
5
67
73
92
93
33
27
8
15b 11 G2-C2-PPh2-PdCl2 (2:1)
16
10 G2-C2-PPh2-PdCl2 (4:1)
19
7
17
18
19
12 G3-C2-PPh2-PdCl2 (2:1)
12 G3-C2-PPh2-PdCl2 (2:1)
G3-C2-PPh2-PdCl2 (2:1)c
19
5
5
89
53
78
11
47
22
20
21
7
6
“G1-C2-PPh2-Pd-dba2”
“G1-C2-PPh2-Pd-(PPh3)2”
5
5
76
36
24
64
a
Conditions: PhI (1.5 mmol), MeOH (5 mL), Et3N (0.4 mL),
CO (100 psi); 105 °C and 17 mg of the dendritic catalyst; product
yields were determined by GC. Pretreatment of catalyst (no. 11
from entry 11 in Table 1) in MeOH; Et3N; 105 °C; 100 psi of CO
without PhI, which formed
b
a black catalyst, then PhI was
introduced and the reaction mixture was heated again for 5 h.
c Retaining 3.1% Pd loading.
The highest TON values are obtained for the carbon-
ylation reaction when a G3 catalyst was used. Thus for
a substrate-to-Pd ratio of 300 a TON of 234 is obtained.
An interesting result is reported in entry 10, where the
substrate-to-Pd ratio was as high as 1226, with a TON
value of 652. These high values can be attributed to the
low loading, but still they compare well with other
examples (entries 2, 12, and 13) for lower generation
catalysts.
and 4:1. For the G1 catalyst 88% and 91% yield of the
ester were observed for 2:1 and 4:1 complexes, respec-
tively (entries 11 and 12). Quite similar results were
observed for the G2 catalyst, where 73% and 67% yields
of the ester were obtained for 2:1 and 4:1 complexes,
respectively (entries 14 and 13). It is clear though that
the activity of the catalyst decreases when going to the
higher generation 2 compared to generation 1.
(d) The yield of ester increases by pretreatment of the
catalyst before the carbonylation reaction. Thus, the ester
yield was 92% when PhI was added to a mixture of all
the components heated beforehand (entry 15) compared
to 73% yield obtained without the pretreatment (entry
14).
It is probable that under the reducing reaction condi-
tions, the PdCl2, immobilized on the dendrimer, is
reduced to Pd(0), which is immobilized on the periphery
of the dendrimer, serving as the activated species.
(e) With the G3 dendrimer catalyst, the results were
surprising due to the low Pd loading (catalyst 15; 0.8%
Pd by ICP analysis, Table 1). The reaction (Table 7, entry
18) gave 53% of the ester after 5 h. However, 89% of
methyl benzoate was obtained when the reaction was
effected overnight (entry 17). When the Pd loading on
the third generation Pd catalyst was increased to 3.1%
(catalyst 13, Table 1), the yield increased to 78% after 5
h (entry 19).
There are several other observations made which merit
attention:
(a) The catalyst changes in color from brown-maroon
to black by the end of the reaction. When isolated from
the reaction mixture, ca. 30% of the original weight was
lost. This black catalyst (from Table 1, entry 8) retained
6.4% Pd by ICP analysis, allowing for the recycling of
the dendritic Pd catalyst as shown above. It was found
that pretreatment of the catalyst under the reaction
conditions, without the substrate (PhI), gave a higher
product yield when PhI was introduced (Table 7, entry
15). It is reasonable to assume that, under CO pressure,
reducing conditions exist, and the Pd(II) immobilized onto
the phosphinated dendritic sites was transformed to
Pd(0), which was stabilized by the dendrimer. Kaneda
et al.20 used dendrimer-bound PdCl2 complexes prepared
(18) (a) Onoda, T. Chemtech 1993, 23, 34. (b) Sato, K.; Miyazawa,
C.; Wada, K.; Onoda, T. Nippon Kagaku Kaishi 1994, 681; Chem. Abstr.
1994, 121, 136524.
(f) In comparison with the G1-PdCl2 catalyst, lower
yields were observed for G1 catalysts obtained with other
Pd complexes (entries 20 and 21), e.g. a catalyst prepared
from Pd-dba2‚CHCl3 gave 76% ester (the loading of Pd
(19) (a) Abu-Gnim, C.; Amer, I. J . Chem Soc., Chem. Commun. 1994,
115. (b) Basoli, C.; Botteghi, C.; Cabras, M. A.; Chelucci, G.; Marchetti,
M. J . Organomet. Chem 1995, C20, 488.
(20) Mizugaki, T.; Murata, M.; Ooe, M.; Ebitani, K.; Kaneda, K. J .
Chem. Soc., Chem. Commun. 2002, 52.
J . Org. Chem, Vol. 67, No. 19, 2002 6629

Antebi et al.
TABLE 8. TON a n d TON/h for Differ en t P a lla d iu m Ca ta lysts Su p p or ted on P AMAM Den d r im er s on SiO2 for th e
Ca r bon yla tion of P h Ia
reaction
no.b
ratio
µmol of Pd/
actual µmol catalyst
% yield
of ester TON TON/hr
subst/Pd mol % Pd 25 mg of catalyst
of Pd
no.
type of catalyst
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
11
12
3
96
194
116
91
1
20.4
11.7
19
16.2
8
8
9
8
8
1
G1-C2-PPh2-PdCl2 (2:1)
G1-C2-PPh2-PdCl2 (4:1)
G1-C2-PPh2-PdCl2
G1-C2-PPh2-PdCl2 + dppb
G1-C2-PPh2-PdCl2
G1-C2-PCycl2-PdCl2
G2-C2-PPh2-PdCl2 (4:1)
G2-C2-PPh2-PdCl2 (2:1)
G2-C2-PPh2-PdCl2 (2:1)
G3-C2-PPh2-PdCl2 (2:1)
G3-C2-PPh2-PdCl2 (2:1)
G0-C2-PPh2-PdMe2
G1-C2-PPh2-PdMe2
G1-C2-PPh2-Pd(PPh3)2
G1-C2-PPh2-Pd(PPh3)2
“G1-C2-PPh2-Pd(dba)2”
10% Pd/C
88
91
91
100
86
87
67
73
92
53
78
89
74
36
60
76
78
75
87
176
105
91
18
35
21
18
28
24
47
36
44
130
47
35
72
64
29
29
24
20
0.5
0.9
1.1
0.6
0.75
0.3
0.4
0.4
0.08
0.33
0.6
0.2
0.1
0.4
0.7
0.60
0.8
13.2
16.4
9.3
11.3
4.3
6.4
6.3
1.25
5
8.8
3.2
1.7
6.2
10.7
10
4c
7
19
166
133
333
245
240
1226
300
167
500
1000
250
143
156
132
13.2
15.8
6.3
9.3
9.3
1.8
7.2
13
4.7
2.5
2.5
15.2
138
119
236
179
220
652
234
176
358
318
145
144
121
99
9
2
13
14
15d
18
19
1
10
11
11
12
13
4
5
6
6
7
2
21
20
11.5
Pd(PhCN)2Cl2
a
General conditions: PhI (1.5 mmol), MeOH (5 mL), Et3N (0.4 mL), and the catalyst with 100 psi of CO at 105 °C over 5 h. Catalyst
was a palladium complex of a phosphinated PAMAM supported on SiO2. Entry from Table 7. c 20.5 mg of dppb (excess) added 3:1 over
Pd; the recycled isolated catalyst is yellow. The same catalyst 11 (entry 11 in Table 1) was used. However, the reaction was carried out
b
d
in two stages: first, the catalyst was heated for 5 h under the same conditions with MeOH and Et3N, but without PhI. The catalyst
turned black. In stage 2, PhI was introduced and the reaction was run under the same conditions for 5 h.
TABLE 9. Eva lu a tion of P a lla d iu m Lea ch in g fr om Differ en t Den d r im er Ca ta lysts in th e Ca r bon yla tion of
Iod oben zen ea
% Pd
reaction
time, h
% yield
of ester
% Pd
leaching
reactionb
catalyst
catalyst no.c
loadingd
ratiod
1
2
3
4
5
7
8
11
14
12
G1-PPh2-PdCl2
G1-PPh2-PdCl2
G2-PPh2-PdCl2
G2-PPh2-PdCl2
G1-PPh2-PdCl2
1
3
10
10
11
65
100
46
46
38
2:1
2:1
4:1
4:1
2:1
5
19
5
19
5
86
100
67
100
73
10.2
10.4
6.3
7.2
7.5
a
b
Based on ICP analysis. The numbering of the reactions (entries) as they appear in Table 7. c The number of the catalysts as they
d
appear in Table 1. Ratio of the phosphinic dendrimer to the Pd agent;
by the reaction of diphenylphosphino-methylated den-
drimers with PdCl2(PhCN)2, after their reduction with
hydrazine to Pd(0) complexes immobilized at the periph-
ery of the dendrimers. These Pd(0) dendrimers showed
high catalytic activity for the stereoselective allylic
substitution of allylic acetates with amines. Although not
practical, the addition of dppb to the reaction mixture
stabilizes the Pd(0) formed under the reducing conditions,
and the catalyst is usually isolated as a yellow solid
(Table 7, entry 4).
(b) The Pd complexes immobilized on the phosphonated
PAMAM dendrimers on SiO2 often gave a signal in the
31P solid-state NMR at +27 ppm, which is assigned to
the oxidized phosphorus in the complex dendrimer (Table
1). Despite the presence of some phosphine oxide, the
reactions went to completion in all cases.
Several authors have described triphenylphosphine
oxide, long considered as being inactive in oxo synthesis,
to be an efficient ligand despite its weak coordination to
metal centers. In 1987 Mitsubishi Kasei launched a
30 000 t/year plant for the production of isononyl alcohol
by reductive hydroformylation of octenes.18 The basic
catalyst was a rhodium-triphenylphosphine oxide (TPPO).
The beneficial effect of phosphine oxides in P-N biden-
tate ligands for the Rh-catalyzed hydroformylation of
styrene was reported by several groups.19a,b The ligand
Ph2PCH2NMe2, used with [RhCl(cod)]2 (cod ) 1,5-cyclo-
octadiene), gave 59% olefin conversion, with aldehydes
formed in a 96:4 i/n ratio, whereas with the corresponding
oxide Ph2P(O)CH2NMe2 conversions of 100% (80% after
1 h) and selectivities of 91:9 were observed under
identical conditions.
(c) Residual solids obtained from five separate reactions
(Table 9) after removal of the solvent and after the
nanofiltration of the catalyst were digested with aqua-
regia and analyzed by ICP. Leaching of palladium ranged
from 6 to 10%. The extent of leaching for the first
generation of the Pd-dendrimer catalyst was higher (ca.
10% entries 1 and 2) than that observed for the second
generation (6-7%; entries 3-5). The loading level of the
Pd and the time for the reaction (5 or 19 h) did not play
a role in the leaching process. The same pattern was
noted for the second-generation Pd dendrimer catalysts,
where the leaching level of Pd was 6.3-7.5% regardless
of the time of reaction, Pd loading level, and the ratio
between the phosphine-substituted dendrimer and the
palladium compound (either 2:1 or 4:1). It is possible that
a longer reaction time had a small effect on the leaching,
where 7.2% leaching was obtained after 19 h of reaction
compared to 6.3% after 5 h. The presence of Pd metal
in the solution is a result of partial dissolution of the
solid silica based catalyst in methanol. The higher the
generation of the dendrimer the less the catalyst dis-
solved.
6630 J . Org. Chem., Vol. 67, No. 19, 2002

Carbonylation Reactions of Iodoarenes
TABLE 10. Ca r bon yla tion of Iod oben zen e w ith Differ en t P d Sou r cesa
µmol of
recovered
catalyst/mg
recycle
no.
%
PhI
%
reaction
catalyst/mg
Pd black/4.7
G1-C2-Pcycl2-PdCl2 + Pd metal
recycle from 2/16.9
10% Pd/C (dry) (50% H2O); 70 mg
APSb + Pd black/17.4 + 5.7
SiO2 + Pd black/16.6 + 5.5
recycle from 6/18 mg
Pd used
PhCOOMe
1
2
3
4
5
6
7
44.2
47
0
0
1
0
0
0
1
76
35
18
24
65
82
100
47
43
16.9; gray
13.4; gray
32.9
53.3
51.7
18.6
18.1
15
53
57
22
78
a
b
Conditions: PhI (1.5 mmol); Et3N (0,4 mL), MeOH (5 mL), at 105 °C with 100 psi of CO, 19 h. APS is aminopropylsilica (0.9 mmol/g
amino groups).
TABLE 11. Ca r bon yla tion of Iod oben zen e w ith
Differ en t Heter ogen eou s P d Su p p or tsa
(d) The yields obtained for the carbonylation of iodo-
benzene decrease when going to higher generations.
Thus, 91% yield was obtained for G1 (and G0), 73% for
G2, and 53% for G3 (after 5 h and a 2:1 ratio of the
phosphinated dendrimer and the Pd reactant; Table 7).
This trend could be attributed to the extent of im-
mobilization of the Pd compound on higher generations
of the dendrimer.
% selectivity
of PhCOOMe
% conversion
reaction
catalyst
of PhI
1
2
3
4
5
6
7
10%Pd/C
5%Pd/C
5%Pd/C
5%Pd/Al2O3
5%Pd/Al2O3
5%Pd/BaSO4
5%Pd/CaCO3
93.8
97.0
98.6
97.0
98.8
97.0
98.5
98.9
99.0
99.5
98.9
98.3
99.7
99.5
Miscella n eou s Issu es. Different nondendrimeric Pd
sources were examined for the carbonylation of PhI under
100 psi of CO at 105 °C with Et3N (Table 10). When Pd
metal was used (entry 1) the yield of the ester was only
24%. Interestingly, when the same amount of Pd metal
was mixed with G1-C2-P(cyclohexyl)2 the yield of the
ester rose to 65%. Recycling the isolated gray catalyst
from the reaction, under the same conditions, afforded
an even higher yield of product (82%). This can be
explained by the formation of finely divided Pd particles
stabilized on the phosphonated dendritic surface. The
mixing of Pd metal with the starting aminopropysilica
(APS; used for the building of the dendrimer) or with SiO2
resulted in similar yields (47% and 43% yield respec-
tively). These results are lower than the 65% yield
obtained with the phosphonated dendrimer (entry 2),
which suggests that the phosphines act as better coor-
dination sites for the fine precipitating Pd particles.
When 10% Pd/C was used the yield was quantitative,
which could explain the role of a high surface factor in
the promotion of the carbonylation with Pd(0). This is
further confirmed with the recycling of the isolated
catalyst mixture of Pd/SiO2 (entry 6) and the increase of
the yield to 78% (entry 7).
In another set of reactions Pd on different supports
such as 10% Pd/C, 5% Pd/C, 5% Pd/Al2O3, 5%Pd/BaSO4,
and 5% Pd/CaCO3 was examined for the carbonylation
of PhI. When the reaction was carried out in MeOH at
105 °C, at 600 psi of CO with Et3N acting as the base, it
was complete in 1 h in high selectivity (Table 11). In
comparison, Yagita et al.13 used water in the reaction,
claiming it enhanced the activity for the Pd/C, for the
carbonylation of bromobenzene. Thus, PhI was carbony-
lated in MeOH/H2O at 160 °C and 154 psi of CO, giving
a mixture of PhCOOMe/PhCOOH in a 77:20 ratio.
Con clu sion s. The dendrimer-bound Pd complexes of
PAMAM dendrimers supported on SiO2 show high cata-
lytic activity for the carbonylation of iodoarenes to the
methyl esters. The reaction can be carried out in MeOH
a
b
Conditions: 10% PhI/10% TEA. /MeOH, 105 °C, 600 psi of
b
CO, 1 h. TEA ) triethylamine.
as the solvent, at low pressure of CO and at 100 °C, using
Et3N as a base. The catalysts are active even with an
oxidized phosphorus in the coordination site. The cata-
lysts could be recycled 4-5 times under the same reaction
conditions. The scope of the carbonylation reaction with
Pd-bound dendrimers was good, as the reaction proceeded
with electron-withdrawing and -donating groups.
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r e for th e Ca r bon yla -
tion of Iod oa r en es. To a 50- mL stainless steel autoclave,
equipped with a glass liner and a magnetic stirring bar, were
charged ArI (1.5 mmol), MeOH (5 mL), Et3N (0.4 mL), and 17
mg of the Pd-dendritic catalyst.16a,c The autoclave was flushed
several times with CO, and then CO was introduced until a
pressure of 100 psi was obtained. The reaction mixture was
heated in an oil bath to 105 °C for a period of 5 h (or overnight,
ca. 19 h). After cooling, the reaction mixture was filtered
through a 0.45 µm membrane and washed with dichoromethane.
The ratio of ester to starting material was determined by GC
analysis.
For quantitative 1H NMR experiments, the solvent was
removed carefully under vacuum at 40 °C. The ratio between
the methyl group in the ester and the two hydrogens in 1,2-
tetraachloroethane (δ 5.9 ppm), used as the reference, gave a
reliable determination of the yield of the formed ester.
The isolation of several products by liquid chromatography
demonstrated good correlation between the GC results and the
1H NMR experiments (i.e. the isolated yield was 4-6% lower
than determined by GC). Consequently, it was decided to
monitor the reaction with GC (area %).
Ack n ow led gm en t . We are grateful to DuPont
Canada and to the NSERC/NRC Research Partnership
Program for support of this research.
J O020271N
J . Org. Chem, Vol. 67, No. 19, 2002 6631