Consequences of Affinity in Heterogeneous Catalytic Reactions: Highly Chemoselective Hydrogenolysis of Iodoarenes

Article abstract of DOI:10.1021/jo0012243

The catalytic hydrodeiodination reaction using molecular hydrogen and Pd/C has been revisited. It is shown, for the first time, that the chemoselectivity of this reaction is controlled by the high affinity of the iodinated compound for the catalyst. This reaction is compatible with most easily reducible functional groups (nitro, aldehyde, olefin, etc.). Using this reaction, the first general method for tritium labeling of 3-(trifluoromethyl)-3-phenyldiazirine is described.

Full text of DOI:10.1021/jo0012243

J . Org. Chem. 2000, 65, 7183-7186
7183
Con sequ en ces of Affin ity in Heter ogen eou s Ca ta lytic Rea ction s:
High ly Ch em oselective Hyd r ogen olysis of Iod oa r en es
Yves Ambroise,† Charles Mioskowski,† Ge´rald Dje´ga-Mariadassou,‡ and Bernard Rousseau*,†
Service des Molecules Marque´es, CEA/ Saclay, 91191 Gif sur Yvette, France, and Laboratoire de
Re´activite´ de Surface, Universite´ Pierre et Marie Curie, 4 place J ussieu, 75252 Paris, France
bernard.rousseau@cea.fr
Received August 10, 2000
The catalytic hydrodeiodination reaction using molecular hydrogen and Pd/C has been revisited.
It is shown, for the first time, that the chemoselectivity of this reaction is controlled by the high
affinity of the iodinated compound for the catalyst. This reaction is compatible with most easily
reducible functional groups (nitro, aldehyde, olefin, etc.). Using this reaction, the first general method
for tritium labeling of 3-(trifluoromethyl)-3-phenyldiazirine is described.
In tr od u ction
iodoarene was selectively reduced while the more reactive
double bond was unaffected. This surprising behavior can
be neither explained nor anticipated by the rate constants
for product formation and relies only on substrate affinity
for the palladium catalyst.
Selectivity in organic chemistry is a major asset during
the synthesis of complex molecules, and the development
of chemoselective reactions provides a powerful tool for
organic chemists. Current chemoselective reactions rely
on a simple principle: the selectivity reflects the rate
constants for product formation.1 We report that the
selectivity observed in some heterogeneous catalyst
transformations is not consistent with this principle.
We revisited the catalytic hydrogenolysis of iodoarenes
using hydrogen and Pd/C.2-5 We determined its chemose-
lectivity versus a variety of functional groups that are
known to be easily reduced (e.g., double bonds). When
an iodoaryl compound and an olefin were separately
subjected to the reducing conditions, the double bond was
transformed faster. When these two groups were reacted
together (whether borne by the same molecule or not),
the selectivity obtained was unexpected. The less reactive
This important result points to new synthetic strate-
gies. We illustrated the utility of this reaction by solving
the 3-(trifluoromethyl)-3-phenyldiazirine tritium labeling
problem.6 This template is used for photoaffinity labeling
and cross-linking experiments, which are biochemical
techniques used to investigate structural and functional
properties of biological systems. Tritium incorporation in
the 3-(trifluoromethyl)-3-phenyldiazirine template re-
quires the introduction of the radioactive element at an
early stage of the synthesis, thus increasing the number
of radioactive steps. Under our conditions, we were able
to introduce tritium in the final step of the synthesis.
Resu lts a n d Discu ssion
† CEA/Saclay.
Under the following conditionssmethanol, 10% Pd/C,
triethylamine, 10 equiv, H2, 1 atmshydrogenolysis of
iodobenzene was complete in 40 min, of bromobenzene
in 14 min, and of chlorobenzene in 6 min. Under the same
‡ Universite´ Pierre et Marie Curie.
(1) (a) Yamakawa, M.; Ito, H.; Noyori, R. J . Am. Chem. Soc. 2000,
122, 1466-1478. (b) Carren˜o, M. C.; Ruano, J . L. G.; Urbano, A.;
Remor, C. Z.; Arroyo, Y. J . Org. Chem. 2000, 65, 453-458. (c) Saito,
S.; Nakagawa, S.; Koizumi, T.; Hirayama, K.; Yamamoto, Y. J . Org.
Chem. 1999, 64, 3975-3978. (d) Shibata, I.; Moriuchi-Kawagami, T.;
Tanizawa, D.; Suwa, T.; Sugiyama, E.; Matsuda, H.; Baba, A. J . Org.
Chem. 1998, 63, 383-385. (e) Yasuda, M.; Hayashi, K.; Katoh, Y.;
Shibata, I.; Baba, A. J . Am. Chem. Soc. 1998, 120, 715-721. (f)
Ohkuma, T.; Ooka; H.; Ikariya, T.; Noyori, R. J . Am. Chem. Soc. 1995,
117, 10417-10418. (g)Yanagisawa, A.; Inoue, H.; Morodome, M.;
Yamamoto, H. J . Am. Chem. Soc. 1993, 115, 10356-10357.
(2) Removal of an iodine atom temporarily attached to a molecule
in order to modify its reactivity: (a) Ahluwalia, V. K.; Prakash, C.;
Gupta, R. Tetrahedron 1982, 38, 609-611. (b) Ahluwalia, V. K.; J olly,
R. S.; Tehim, A. K. Tetrahedron 1982, 38, 3673-3677. (c) Kende, A.
S.; Ebetino, F. H.; Otha, T. Tetrahedron Lett. 1985, 26, 3063-3066.
(d) Weeratunga, G.; Horne, S.; Rodrigo, R. J . Chem. Soc., Chem.
Commun. 1988, 721-722. (e) Nishiyama, S.; Suzuki, Y.; Yamamura,
S. Tetrahedron Lett. 1989, 30, 379-382. (f) Donnelly, D. M. X.; Molloy,
D. J .; Reilly, J . P.; Finet, J . P. J . Chem. Soc., Perkin Trans. 1 1995,
2531-2534. (g) Fukuyama, T.; Liu, G. J . Am. Chem. Soc. 1996, 118,
7426-7427.
(5) Tritium and deuterium labeling: (a) Chauvie`re, G.; Rousseau,
B.; Pillon, F.; Pe´rie´, J . J . Labeled Compd. Radiopharm. 1998, 41, 47-
51. (b) Darula, Z.; Peter, A.; Toth, G. J . Labeled Compd. Radiopharm.
1997, 39, 817-826. (c) Kalindjian, S. B.; Harper, E. A.; Pether, M. J .
Bioorg. Med. Chem. Lett. 1996, 6(10), 1171-1174. (d) Alcaraz, M. L.;
Peng, L.; Klotz, P.; Goeldner, M. J . Org. Chem. 1996, 61, 192-201. (e)
Seltzman, H. H.; Carroll, F. I.; Burgess, J . P.; Wyrick, D.; Burch, D. F.
J . Chem. Soc., Chem. Commun. 1995, 1549-1550. (f) Latli, B.; Casida,
J . E. J . Labeled Compd. Radiopharm. 1995, 36, 147-156. (g) Fuka-
zawa, Y.; Usui, S.; Tanimoto, K.; Hirai, Y. J . Am. Chem. Soc. 1994,
116, 8169-8175. (h) Vader, J .; Kaspersen, F.; Sperling, E.; Schlachter,
I.; Terpstra, A. J . Labeled Compd. Radiopharm. 1994, 34, 845-870.
(i) Yamamoto, M.; Dolle, V.; Warnock, W.; Diyizou, Y.; Yamada, M.;
Nakatani, Y.; Ourisson, G. Bull. Soc. Chim. Fr. 1994, 131, 317-329.
(6) (a) Hori, N.; Iwai, S.; Inoue, H.; Ohtsuka, E. J . Biol. Chem. 1992,
267, 15591-15594. (b) Brunner, J . Annu. Rev. Biochem. 1993, 62, 483-
514. (c) Delfino, J . M.; Schreiber, S. L.; Richards, F. M. J . Am. Chem.
Soc. 1993, 115, 3458-3474. (d) Kotzyba-Hibert, F.; Kapfer, I.; Goeldner,
M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1296-1312. (e) Hesterkamp,
T.; Hauser, S.; Luetcke, H.; Bukau, B. Proc. Natl. Acad. Sci. U.S.A.
1996, 93, 4437-4441. (f) Grassi, D.; Lippuner, V.; Aebi, M.; Brunner,
J .; Vasella, A. J . Am. Chem. Soc. 1997, 119, 10992-10999. (g)
Hatanaka, Y.; Hashimoto, M.; Kanaoka, Y. J Am. Chem. Soc. 1998,
120, 453-454. (h) Weber, T.; Brunner, J . J . Am. Chem. Soc. 1995, 117,
3084-3095.
(3) Preparation of monoiodinated compounds from their dihaloge-
nated analogues: (a) Numazawa, M.; Kimura, K.; Ogata, M.; Nagaoka,
M. J . Org. Chem. 1985, 50, 5421-5423. (b) Horne, S.; Weeratunga,
G.; Rodrigo, R. J . Chem. Soc., Chem. Commun. 1990, 39-41.
(4) Synthesis of new molecules of biological interest: (a) Li, T.; Zeng,
Z.; Estevez, V. A.; Baldenius, K. U.; Nicolaou, K. C. J oyce, G. F. J .
Am. Chem. Soc. 1994, 116, 3709-3715. (b) Ishiwata, H.; Sone, H.;
Kigoshi, H.; Yamada, K. Tetrahedron 1994, 50, 12853-12882.
10.1021/jo0012243 CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/13/2000

7184 J . Org. Chem., Vol. 65, No. 21, 2000
Ambroise et al.
when the iodo compound had completely disappeared
from the reaction mixture (dotted line 9). The same
competitive reaction was carried out with bromobenzene
versus trans-1-phenyl-1-propene. In this case, no selectiv-
ity was observed. The olefin reduction and the bromoben-
zene hydrogenolysis took place simultaneously. These
experiments showed that the iodobenzene delayed the
double-bond reduction. As a matter of fact, if the iodo-
benzene is highly adsorbed on the catalyst surface, its
reduction kinetics will not depend on its concentration
in solution but rather on its concentration on the pal-
ladium surface. The number of catalytic sites being
constant, the transformation rate of iodobenzene would
also be constant throughout the reaction. The graphical
representation of its time-dependent conversion would
then be a straight line, and this is indeed what is seen
in Figure 2. These experiments show that the selectivity
can be explained by a marked adsorption of the aryl
iodide on the catalyst. The catalytic sites would then be
totally occupied, thereby preventing access by the olefin
and hence explaining the selectivity.
F igu r e 1. Schematic overview of the different steps in a
surface reaction.
conditions, trans-1-phenylstyrene was completely reduced
in 3 min. These results suggest that selective hydro-
genolysis of a carbon-halogen bond in the presence of a
double bond is unlikely. To solve this problem, we focused
our attention on the mechanism of this catalytic hydro-
genation. The substrate is first adsorbed on the pal-
ladium surface (Figure 1).7 Then, the adsorbed substrate
reacts with the adsorbed hydrogen to give the product,
which is subsequently released from the catalyst surface.
To modify the selectivity of the hydrogenation reaction,
we focused on the adsorption step rather than on the
reaction step. If two substrates S1 and S2 are competi-
tively reduced, the selectivity could result from the
preferential adsorption of S1 or S2 on the catalyst
surface. If the affinity of S1 for the catalyst is sufficiently
greater than that of S2, then S1 will predominantly cover
the catalyst and prevent the adsorption of S2. In this
case, hydrogenation of S1 will preferentially occur.
Recently, Tundo et al. showed that iodobenzene has a
greater affinity for Pd/C than bromo- and chlorobenzene.8
We imagined that this property could be advantageous
in obtaining high selectivities. This hypothesis was tested
using the iodobenzene and trans-1-phenylpropene tem-
plates (Figure 2).
Deiod in a tion of Bifu n ction a l Com p ou n d s. In the
previous experiments, the competitive hydrogenation was
carried out on two different molecules. We tried to
determine whether the same selectivity could be obtained
when both an aryl iodide and an easily reducible function
are borne by the same molecule (Table 1).
Ta ble 1. Hyd r od eiod in a tion of Bifu n ction a l
Com p ou n d sa
F igu r e 2. Hydrogenation of PhI and trans-1-phenyl-1-propene
with hydrogen and Pd/C catalyst. Full line noncompetitive
reactions of trans-1-phenyl-1-propene (9) and PhI (9). Dotted
line: competitive reactions between trans-1-phenyl-1-propene
(9) and PhI (2).
We first verified that the stirring speed was sufficiently
high (1400 rpm) to ensure that the reduction occurred
under kinetic conditions. trans-1-Phenyl-1-propene and
iodobenzene were hydrogenated separately under the
reaction conditions mentioned above. The double bond
reduction was complete within 3 min, and iodobenzene
was converted into benzene within 40 min (full line 9
and 2, respectively). When these two substrates were
reacted competitively, the reduction of trans-1-phenyl-
1-propene was delayed and iodobenzene reacted first
(dotted line 2). The reduction of the olefin only occurred
a
Same conditions as described in the text.
The chemoselectivity that we previously noted was
confirmed. The hydrogenolysis of the carbon-iodine bond
occurred before the reduction of nitro (entry 1) or alde-
hyde (entry 2) groups. The selectivity again was excellent
for the double bond (entry 3), and to the best of our
knowledge this is the first report of the selective hydro-
genolysis of a carbon-iodine bond in the presence of a
double bond. These experiments show that the selectivity
observed with iodobenzene can be generalized to other
substrates.
Gen er a l Met h od for t h e La b elin g of t h e 3-(Tr i-
flu or om et h yl)-3-p h en yld ia zir in e Gr ou p . We took
advantage of this efficient hydrogenolysis method by
solving the 3-(trifluoromethyl)-3-phenyldiazirine tritium
labeling problem. This template is known to be easily
(7) (a) Boudart, M.; Dje´ga-Mariadassou, G. Kinetics of Heterogeneous
Catalytic Reactions; Princeton University Press: Princeton, 1984; pp
38-76. (b) Norskov, J . K.; Stoltze, P. In Handbook of Heterogeneous
Catalysis; Ertl, G., Kno¨zinger, H., Weitkamp, J ., Eds.; VCH: Wein-
heim, Germany, 1997; Vol. 3, pp 984-991.
(8) Marques, C. A.; Selva, M.; Tundo, P. J . Org. Chem. 1993, 58,
5256-5260.

Chemoselective Hydrogenolysis of Iodoarenes
J . Org. Chem., Vol. 65, No. 21, 2000 7185
reduced by molecular hydrogen and Pd.9 Consequently,
when one designs a photoactivatable probe containing a
tritium nucleus as well as a diazirine group, the radio-
element needs to be introduced early in the synthesis.
This implies several radioactive steps.10 For example, six
radioactive steps were required to label the photoacti-
vatable membrane-spanning phospholipid probe recently
introduced by Schreiber and Richards.6c This multistep
synthesis considerably increases costs, experimental dif-
ficulties, and radioactive waste. Furthermore, the lack
of an efficient tritium-labeling method for 3-(trifluoro-
methyl)-3-phenyldiazirine is a major drawback for the
use of this photolabeling reagent.
Ta ble 2. Ch em oselective Hyd r ogen olysis of Iod in a ted
Com p ou n d 1a in th e P r esen ce of a n Ea sily Red u cible
Su bstr a tea
We synthesized compound 1a , which contains the
photoactivatable group and a carbon-iodine bond (Figure
3). Compounds 1a and 1b were easily synthesized by
means of reported procedures.6h For convenience, all tests
were first carried out with hydrogen rather than tritium.
F igu r e 3. Hydrogenolysis kinetics of 1a .
For the first 50 min of the reaction, the disappearance
of the iodinated compound 1a ([) was accompanied by
the concomitant appearance of 1b (2) in over 97% yield.
As expected, the diazirine reduction only began once 1a
had completely disappeared from the reaction medium
(after 50 min). This behavior can be explained by a
selective adsorption of 1a on the catalyst via its carbon-
iodo bond, as demonstrated above.
a
b
The reaction was carried out as described in the text. De-
termined by HPLC: column Zorbax ODS, mobile phase: H2O/
CH3CN 1/1, UV detection at 220.
We determined the scope and limitations of the deio-
dination of the photoactivatable probe by studying the
competitive hydrogenation of an equimolecular mixture
of 1a and an easily reducible substrate. The method’s
selectivity was evaluated by HPLC to quantify easily
reducible substrate remaining when 1a was converted
into 1b with the best yield. The compounds used and the
results obtained are indicated in Table 2.
The selectivity was excellent for disubstituted conju-
gated double bonds of E (entries 1 and 2) or Z (entry 3)
stereochemistry and for methylene double bonds (entry
4). Nonconjugated double bonds (entry 5), aldehyde (entry
6), and nitro (entry 7) were left unchanged. The protective
O-benzyl group is also very stable under these conditions
(entry 8). A weaker selectivity was, however, observed
for the monosubstituted double bonds (entry 9). The
reduction of the monosubstituted triple bonds to double
bonds was faster than the hydrogenolysis of the carbon-
iodine bond (entry 10). The same was true for the
disubstituted triple bonds but with a lower selectivity
(entry 11). All these reactions were carried out under
simple and fast (<1 h) conditions. When carrying out the
reaction with tritium, [3H]-1b was obtained with an
isotopic purity higher than 95% (determined by mass
(9) Schmitz, E.; Ohme, R. Chem. Ber. 1961, 94, 2166-2173.
(10) (a) Brunner, J .; Spiess, M.; Aggeler, R.; Huber, P.; Semenza,
G. Biochemistry 1983, 22, 3812-3820. (b) Harter, C.; Ba¨chi, T.;
Semenza, G.; Brunner, J . Biochemistry 1988, 27, 1856-1864. (c)
Nakayama, T.; Khorana, H. G. J . Org. Chem. 1990, 55, 4953-4956.
(d) Taki, M.; Kuniyasu, A.; Nakayama, H.; Kanaoka, Y. Chem. Pharm.
Bull. 1991, 39, 1860-1862. (e) Resek, J . F.; Bhattacharya, S.; Khorana,
H. G. J . Org. Chem. 1993, 58, 8, 7598-7601. (f) Latli, B.; Morimoto,
H.; Williams, P. G.; Casida, J . E. J . Labeled Compd. Radiopharm. 1998,
41, 191-199.

7186 J . Org. Chem., Vol. 65, No. 21, 2000
Ambroise et al.
3
spectroscopy). H NMR of [3H]-1b gave a unique singlet
Typ ica l Exp er im en t: Hyd r ogen a tion of 1a (See F igu r e
3). [2-Iodo-4-(3-trifluoromethyl-3H-diazirin-3-yl)phenyl]metha-
nol 1a (25.6 mg, 1 equiv) was dissolved in MeOH (3 mL) and
triethylamine (105 µL, 10 equiv). Pd/C (7.95 mg, 0.1 equiv)
was added, and the reaction mixture was vigorously stirred
at 20 °C under 1 atm of hydrogen. During the reaction, the
pressure of H2 was maintained constant by a continuous
supply of H2 (checked with an accurate manometer). Aliquots
(80 µL) were sampled at 0, 10, 20, 40, 50, 65, 85, and 105 min,
filtered, and analyzed by HPLC (H2O/MeCN ) 1/1, 1 mL/min,
UV detection at 220 nm). The rates of conversion of 1a and
production of 1b were derived from the respective areas by
correlation with standard curves.
in the aromatic region, thus showing that tritium did not
exchange under our catalytic conditions. A detailed
description of the synthesis and characterization of [3H]-
1b, and as well as of two other photoactivatable probes,
will be reported in due course.
Con clu d in g Rem a r k s
We have shown that the rate constants for product
formation neither explain nor allow anticipation of the
chemoselectivity that occurs during a catalytic deiodina-
tion using H2 and Pd/C. We have demonstrated, for the
first time, that the selectivity of this reaction is controlled
by the substrate’s affinity for the catalyst. As an applica-
tion, we developed the first general method for tritium
labeling of the 3-(trifluoromethyl)-3-phenyldiazirine pho-
toactivatable template in the final step. Beyond the
radiolabeling of fragile molecules, we think that this
principle is general, will lead to new chemoselective
reactions, and will find wide application in the areas of
synthetic organic, mechanistic organic, organometallic,
and catalysis chemistry.
This typical procedure was applied for all the hydrogena-
tions described in this paper, including hydrogenation of PhI
versus trans-1-phenyl-1-propene (Figure 2), deiodination of
bifunctional compounds (Table 1), and competitive hydrogena-
tions versus 1a (Table 2).
Ack n ow led gm en t. We warmly thank Dr. Eric Doris
for helpful discussions. We also thank Ms. Florence
Pillon and Mr. Alain Valleix for running important
control experiments and Dr. L. Pichat for critical read-
ing of this paper.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
1
cedures, characterization data, and H and 13C spectra for all
Exp er im en ta l P r oced u r es
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
The palladium on carbon (10% w/w) used in the experiments
was from Aldrich and was used without any treatment.
J O0012243