Palladium on carbon encapsulated in POEPOP1500: A resin-supported catalyst for hydrogenation reactions

Article abstract of DOI:10.1021/ol016812x

(formula presented) A new a versatile catalyst for hydrogenation reactions wherein palladium on arbon is encapsulated in POEPOP₁₅₀₀-resin is described. This polymer-supported catalyst has been successfully used in solution phase hydrogenation of a double and a triple bond as well as hydrogenolysis of a benzyl-protecting group. While the activity of the new catalyst is marginally lower than standard 10% Pd/C, it has the advantage of being reused several times without significant loss of reactivity.

Full text of DOI:10.1021/ol016812x

ORGANIC
LETTERS
2
002
Vol. 4, No. 1
7-30
Palladium on Carbon Encapsulated in
POEPOP : A Resin-Supported
2
1500
Catalyst for Hydrogenation Reactions
†
‡
Anita M. Jansson, Morten Grøtli, Koen M. Halkes, and Morten Meldal*
Department of Chemistry, Carlsberg Laboratory, Gamle Carlsberg Vej 10,
DK-2500 Valby, Copenhagen, Denmark
mpm@crc.dk
Received September 24, 2001
ABSTRACT
A new a versatile catalyst for hydrogenation reactions wherein palladium on carbon is encapsulated in POEPOP1500-resin is described. This
polymer-supported catalyst has been successfully used in solution phase hydrogenation of a double and a triple bond as well as hydrogenolysis
of a benzyl-protecting group. While the activity of the new catalyst is marginally lower than standard 10% Pd/C, it has the advantage of being
reused several times without significant loss of reactivity.
The use of polymer-supported catalysts in organic chemistry
has continued to be an area of rapid development. The
published by Zecca et al.2b However, broad usage of these
polymer-supported catalysts was limited as a result of their
poor solvent compatibility. Recently, Niu et al. reported size-
selective hydrogenation of olefins by dendrimer-encapsulated
1
intrinsic advantage of these catalysts in synthetic chemistry
is that they can be removed by filtration upon completion
of the reaction. Furthermore, the polymer-supported catalysts
are easily recycled, thereby reducing the costs. Using a
polymer-supported catalyst also makes monitoring of a
reaction by simple analytical techniques more straightfor-
ward.
2
c
palladium nanoparticles.
We report herein a new and versatile catalyst for hydro-
genation reactions, where the palladium on carbon is
encapsulated in POEPOP1500 resin, a polyoxyethylene-
3
polypropoxylene polymer containing only ether bonds. The
Solution-phase hydrogenation using palladium as catalyst
is widely used, and there are only a few examples reported
in the literature where the catalyst is polymer-supported. Jo
et al. have reported gel-like and macroporous polymer
supports for palladium(II) complexes,2a and the use of ion-
exchange resins as support for a palladium catalyst was
POEPOP resin was chosen as solid support because of its
excellent swelling properties in solvents ranging from water
to dichloromethane. The POEPOP1500-Pd/C resin was
synthesized by introduction of 10% Pd/C directly in the
mixture of reactants used for anionic suspension polymeri-
4
zation. A mixture of the monomer, Pd/C, tBuOK (acting as
base) and surfactant was added to silicon oil at 120 °C with
stirring. The resulting beads were fractionated according to
†
Current address: Biotechnology Centre of Oslo, University of Oslo,
P. O.Box 1125 Blindern, N-0317 Oslo, Norway.
‡
Current address: Bijvoet Center for Biomolecular Research, Utrecht
University, Padualaan 8, Utrecht, The Netherlands.
(2) (a) Jo, Y.-D.; Ahn, J.-H.; Ihm, S.-K. Polym. Int. 1997, 44, 49. (b)
Zecca, M.; Fi sˇ era, R.; Palma, G.; Lora, S.; Hronec, M.; Kr a´ lik, M. Chem.
Eur. J. 2000, 6, 1980. (c) Niu, Y.; Yeung, L. K.; Crooks, R. M. J. Am.
Chem. Soc. 2001, 123, 6840.
(3) (a) Renil, M.; Meldal M. Tetrahedron Lett. 1996, 37, 6185. (b) Grøtli,
M.; Gotfredsen, C. H.; Rademann, J.; Buchardt, J.; Clark, A. J.; Duus, J.
Ø; Meldal, M. J. Comb. Chem. 2000, 2, 108. (c) Grøtli, M.; Rademann, J.;
Groth, T.; Lubell, W. D.; Miranda, L. P.; Meldal. M. J. Comb. Chem. 2001,
3, 28.
(
1) Reviews: (a) Shuttleworth, S. J.; Allin, S. M.; Sharma, P. K. Synthesis
1
997, 1217. (b) Shuttleworth, S. J.; Allin, S. M.; Wilson, R. D.; Nasturica,
D. Synthesis 2000, 1035. (c) Drewry, D. H.; Coe, D. M.; Poon, S. Med.
Res. ReV. 1999, 19, 97. (d) de Miguel, Y. R. J. Chem. Soc., Perkin Trans.
1
2000, 4213. (e) Lindner, E.; Schneller, T.; Auer, F.; Mayer, H. A. Angew.
Chem., Int. Ed. 1999, 38, 2154. (f) Pittman, C. U., Jr. Polym. News 1998,
2
3, 416. (g) Choplin, A.; Quignard, F. Coord. Chem. ReV. 1998, 178-180,
1
679.
1
0.1021/ol016812x CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/11/2001

4
particle sizes by sieving. To investigate whether the presence
of palladium particles had any influence on the polymeri-
zation process, the swelling of the POEPOP1500-Pd/C resin
was measured in tetrahydrofuran (THF), MeCN, dimethyl-
formamide (DMF), water, and dichloromethane (DCM) and
compared to conventional POEPOP1500 resin3b (Figure 1).
0.25 mmol/g, which was significant lower than for conven-
3
b
tional POEPOP1500 resin (0.4-0.6 mmol/g). This indicates
that the Pd/C particles prevent full access to the functional
sites on the resin, probably through coordination of PEG
chains to the Pd atoms. However, the amount of functional
groups available for attachment was not important for this
study since hydrogenation of the substrates takes place on
the surface of the palladium.
5
With this resin in hand, its usefulness as catalyst in
solution-phase hydrogenation was investigated. Three model
compounds were chosen (Scheme 1): a double bond (allyl
Scheme 1
Figure 1. Swelling properties of POEPOP1500 resin with encap-
sulated Pd/C (grey) and conventional POEPOP1500 resin (white) in
different solvents determined by the syringe technique.5
The POEPOP1500-Pd/C resin exhibited slightly higher swell-
ing (5-10%) than that of conventional resin, indicating that
the palladium particles interfere slightly with the polymer-
ization process, resulting in a lower cross-linking and a
marginally higher swelling. The superb swelling of the
POEPOP1500-Pd/C resin in solvents of ranging polarity is
advantageous and broadens the scope of solvents that can
be used for hydrogenations.
The loading (quantity of hydroxyl groups available for
substrate attachment) for the POEPOP1500-Pd/C resin was
(4) Preparation of POEPOP1500-Pd/C Resin. The PEG1500 mac-
romonomer (1.5 equiv functionalization) and the polymeric surfactant were
synthesized as previously described in ref 3c. A suspension of PEG1500
macromonomer (2.0 g, 1.25 mmol) in dichloromethane (4 mL) was mixed
with a solution of polymeric surfactant (0.2 g) in dichloromethane (2 mL),
and the solvent was evaporated in vacuo. A solution of potassium tert-
butoxide (0.10 g, 0.89 mmol) in tert-butanol (1.0 mL) was mixed with the
PEG1500,macromonomer at 40 °C, and 10% palladium on carbon (0.4 g)
added. After 5 min of rapid stirring the mixture was added to silicon oil
group) of a fully protected adenosyl derivative (entry A), a
triple bond of N-propyne-substituted phthalimide (entry B),
R
and a benzyl protecting group on the side chain of an N -
Boc-protected tyrosine derivative (entry C).
Hydrogenation was performed in parallel in Teflon tubes
in a Parr apparatus at room temperature, under pressure or
(175 mL) at 120 °C, in a 250 mL polymerization flask fitted with an
overhead stirrer (300 rpm.). The reaction was stirred at 120 °C for 12 h.
The silicon oil was decanted, and the polymer particles were suspended in
dichloromethane and placed on a 50 µm filter to remove excess of palladium
on carbon and the smallest polymer particles. The beads were washed with
dichloromethane, dimethyl formamide, methanol, and water (4 × 5 mL for
each solvent); treated with 4 M NaOH (16 mL) for 2 h at 60 °C; washed
extensively with water, methanol, dimethyl formamide, and dichloromethane
in atmospheric pressure, overnight (5 µmol substrate, 6 mg
POEPOP1500-Pd/C resin, 0.4 mL solvent, without stirring).6
The substrates were dissolved in appropriate solvents (MeCN
in entries A and B, and MeOH/water, 10:1 in entry C), the
(
4 × 5 mL for each solvent); dried under high vacuum for 24 h; and sieved.
(6) General Procedure for Hydrogenation Reactions. The substrate
(5 µmol) was dissolved in the appropriate solvent (acetonitrile, entry A
and B, or methanol/water 10:1, entry C, 0.4 mL), and the solution was
added to the resin-Pd/C (6 mg, 500 < x < 1000 µm) in a Teflon tube.
After 30 min of swelling, the Teflon tubes were placed in a Parr apparatus
and after flushing with argon and hydrogen, hydrogenation was started under
pressure (40 or 20 bar) or under atmospheric pressure in room temperature,
overnight, without stirring. The resin was filtered off and washed with
acetonitrile (entries A and B) or 50% aqueous methanol (entry C), 2 × 0.2
mL. The combined solutions were analyzed with HPLC and ES-MS (entry
A).
Yield: 1.69 g (with approximately 20% Pd/C by weight), 77%; 500 < x <
1
1
000 µm, 0.12 g; 300 < x < 500 µm, 0.33 g; 212 < x < 300 µm, 1.09 g;
06 < x < 212 µm, 0.15 g. The loading of the resins was determined by
esterification with Fmoc-Gly activated by MSNT in the presence of
N-methyl imidazole in CH2Cl2 and subsequent measurement of the UV
absorbency of the adduct of dibenzofulvene and piperidine formed on
treatment of a weighed polymer sample with 20% piperidine in DMF.
Loading: 0.25 mmol/g. Swelling was determined by the syringe technique,
described in ref 5 (Figure 1).
(5) Auzanneau, F.-I.; Meldal, M.; Bock, K. J. Pept. Sci. 1995, 1, 31.
28
Org. Lett., Vol. 4, No. 1, 2002

POEPOP1500-Pd/C resin was added to the solution, and after
swelling (30 min) hydrogenation was started.
The potential of the POEPOP1500-Pd/C resin to be
recycled was tested by reusing the resin in hydrogenation
reactions of the three substrates (5 µmol substrate, 6 mg
Solvent was selected on the basis of the solubility of the
substrates. The workup involved a simple filtration and
6
resin-Pd/C, 40 bar, overnight). The resin was washed with
7
washing of the resin, followed by HPLC analysis and ES-
50% aqueous methanol (3 × 0.4 mL), water (3 × 0.4 mL),
and methanol (3 × 0.4 mL) before it was used in the next
cycle. During the whole recycling process analysis of the
crude products by HPLC (215 nm) showed complete
conversion (>99%) of the starting material to reduced
product with all three substrates. The total conversion of
compound 5 to product 6 was additionally confirmed by ES-
MS (see above) after each cycle (no signal from the starting
material was detected). After five cycles the catalyst remained
active and gave complete conversion with all the substrates
(Figure 2). The loss of palladium from the resin was found
MS.8
To optimize hydrogenation conditions, the substrates were
reduced under three different hydrogen pressures (40 bar,
6
2
0 bar, and atmospheric pressure) overnight. Hydrogenation
at atmospheric pressure resulted in complete reduction of
the double bond of compound 1. The triple bond of
compound 3, however, needed higher hydrogen pressure (20
bar) for completion (2% unconverted starting material
remained when the reaction was performed at atmospheric
pressure). Cleavage of the benzyl group of compound 5 was
a much slower reaction and required a hydrogen pressure of
40 bar overnight for completion. To investigate if the
POEPOP1500-Pd/C resin particle size had any influence on
the outcome of the hydrogenation, three different particle
sizes (212 < x < 300 µm, 300 < x < 500 µm, and 500 <
x < 1000 µm) were used in hydrogenation of compound 5
(40 bar, overnight). Complete cleavage of the benzyl group
with all three particle sizes resulted. Since no significant
influence of particle size was observed during hydrogenation,
the largest particles, 500 < x <1000 µm, were used in all
following experiments because of ease of handling.
For complete structural determination of the hydrogenation
products, the reactions were run on larger scale (0.05 mmol)
under pressure (40 bar) following the procedure in ref 6.
On the basis of HPLC traces the reactions proceeded cleanly
7
with total conversion of the starting materials, and the crude
1
9
products were analyzed by H NMR. The chemical shifts
for the appropriate groups of products 2 and 4 are inserted
in Scheme 1. Hydrogenation of compound 1 provided the
saturated product 2. A methyl group at δ 0.89 and a saturated
methylene group at δ 1.59 had replaced the allylic signals
at δ 5.19 and δ 5.88. Proton NMR also showed that the
TIPDS (1,1,3,3-tetraisopropyldisiloxane-1,3-diyl) and the
pivaloyl protecting groups were stable under these conditions.
The mass of product 2 was confirmed by ES-MS, which
+
showed a peak at m/z 636.3 for [M + H] (M ) 635.9; no
peak from the starting material was observed, M ) 633.9).
Proton NMR of the fully saturated triple bond in the
substituted phthalimide compound 4 also clearly showed a
methyl group at δ 0.88 and a saturated methylene group at
δ 1.64. Hydrogenolysis of the benzyl protecting group of
compound 5 resulted in compound 6 with no loss of the Boc
group. Proton NMR of the product 6 showed no signals for
the benzyl aromatic protons (δ 7.28-7.44), nor for the singlet
corresponding to the benzylic protons (δ 5.05).
Figure 2. HPLC chromatograms (215 nm) of crude products from
entries A, B, and C (Scheme 1) after five cycles of the POE-
7
POP1500-Pd/C resin. (Retention times for the starting materials
are marked with arrows.)
(7) Analytical HPLC was performed on a Zorbax C18 column with buffers
A (0.1% TFA in water) and B (0.1% TFA in MeCN/H2O 9:1); linear
gradient from 100% A to 100% B in 25 min; flow rate 1 mL/min;
wavelength 215 nm. Retention times: entry A, 23.7 min (1), 24.5 min (2);
entry B, 10.5 min (3), 12.6 min (4); entry C, 16.6 min (5), 10, 2 min (6).
to be in average 0.1% per cycle10 and is in agreement with
the retained catalytic activity.
(8) Electrospray spectra were aquired in positive mode on a VG Quattro
The efficiency of the POEPOP1500-Pd/C resin compared
triple quadropole mass spectrometer (Micromass, Ldt.).
1
11
(
9) H NMR spectra (250 MHz) were recorded on a Bruker DRX 250
to an equivalent quantity of conventional 10% Pd/C was
instrument in CDCl3 (compounds 1-4), and in MeOH-d4 (compounds 5
and 6). Chemical shifts were calibrated to internal solvent signals (7.19
ppm for CDCl3 and 3.30 ppm for MeOH-d4)
investigated (3 mg of POEPOP1500-Pd/C resin with ap-
proximately 20% Pd/C by weight is equivalent to 0.6 mg of
Org. Lett., Vol. 4, No. 1, 2002
29

also was confirmed by ES-MS. Reduction of the triple bond
of compound 3 was slightly slower with POEPOP1500-Pd/C
resin (2% unconverted starting material) than with conven-
tional 10% Pd/C (complete conversion) at the highest
substrate concentration. Cleavage of the benzyl protecting
group of compound 5 was much slower under these condi-
tions. The use of polymer-supported catalyst required higher
hydrogen pressure or longer reaction time for total conversion
of the substrate, even at the lowest concentration. At 10 µmol
concentration none of the two catalysts resulted in complete
reaction and the polymer-supported catalyst gave approxi-
mately 20% lower conversion of this substrate compared to
that of conventional catalyst. In summary, in hydrogenation
reactions of compounds 1, 3, and 5, the polymer-supported
catalyst is efficient but slightly less effective compared to
conventional carbon-supported palladium.
In conclusion, we have developed a stable, readily
recyclable polymer-supported catalyst for solution-phase
hydrogenation. The POEPOP1500-Pd/C resin is easily pre-
pared by an encapsulating procedure, is very convenient to
handle, and is compatible with a large range of solvents.
While the activity of the new polymer-supported resin is
marginally lower than standard 10% Pd/C, it has the
advantage of being reusable for several times without
significant loss of reactivity.
Table 1. Conversion of Starting Material to Reduced Product
a
in Hydrogenation Reactions Using POEPOP1500-Pd/C
resin-Pd/C) and 10% Pd/Cb
(
substrate
(µmol)
catalyst
1c
3c
5d
2
5
0
0
0
.5
resin-PD/C
Pd/C
resin-PD/C
Pd/C
resin-PD/C
Pd/C
resin-PD/C
Pd/C
t.c.e
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
t.c.
98%
t.c.
95%
t.c.
90%
t.c.
75%
91%
n.d.f
n.d.
n.d.
n.d.
1
2
4
resin-PD/C
PD/C
a
Reactions were carried out at 20 bar, rt, overnight (ref 6). b Percentage
c
converted starting material determined by HPLC, 215 nm. In MeCN (0.4
mL). In MeOH/H2O 10:1 (0.4 mL). t.c. ) total conversion (>99%).
n.d. ) not detected.
d
e
f
1
0% palladium on carbon; 20 bar, overnight) (Table 1). A
6
series of experiments were performed in which the quantity
of substrate was doubled for each experiment (2.5, 5, 10,
20, and 40 µmol), and the crude products were analyzed by
HPLC. The double bond of compound 1 was reduced to
completion at all concentrations with both catalysts, which
Acknowledgment. This work was carried out at the
SPOCC center and was supported by the Danish National
Research Foundation.
(
10) . Palladium analysis of filtered hydrogenation solutions was
performed by FI-ICP-MS (flow-injection inductively coupled plasma-MS,
ELAN-5000, PE Sciex) with Rh as internal standard (detection limit 0.05
µg Pd /L). Sample preparation: After each cycle of hydrogenation (with
substrate 3) half of the volume (0.4 mL), after filtration and washing (ref
Supporting Information Available: 1H NMR data for
compounds 1-6. This material is available free of charge
via the Internet at http://pubs.acs.org.
6
), was concentrated, and the residue was dissolved in 1 M HCl (10 mL)
for Pd analysis. The Pd content in samples from cycle 1-5 was measured
to be 12, 8, 3, 3, and 3 µg Pd /L, respectively. This corresponds to an
average loss of 0.1% Pd per hydrogenation cycle. (Pd content of blank
solution with no catalyst added was 0.07 µg Pd /L; all samples were analyzed
three times, with RSD < 3%.)
OL016812X
(11) Palladium on activated carbon (10% Pd) was purchased from
Aldrich.
30
Org. Lett., Vol. 4, No. 1, 2002