Communication between surfaces by electron relay in a doubly heterogeneous photochemical reaction [11]

Full text of DOI:10.1021/ja990316a

10436
J. Am. Chem. Soc. 1999, 121, 10436-10437
Scheme 1
Communication between Surfaces by Electron Relay
in a Doubly Heterogeneous Photochemical Reaction
Mohamed Ayadim, Jean L. Habib Jiwan, and
Jean Ph. Soumillion*
Laboratory of Photochemistry
Catholic UniVersity of LouVain, 1 Place Louis Pasteur
B-1348 LouVain La NeuVe, Belgium
ReceiVed February 1, 1999
ReVised Manuscript ReceiVed July 15, 1999
Photochemical reactions using sensitizers covalently linked to
the surface of silica beads have already been reported.1-6
Supported photosensitizers used as a suspension in a liquid offer
obvious advantages: ease of photoproducts separation, analysis
simplification, recycling of the sensitizer, and circumventing a
poor solubility in the reaction medium are among these benefits.
In addition to those obvious advantages, it has been shown that
the organization of sensitizers on the silica surface plays an
important role.6
Scheme 2
In this paper, we show that the surface of organically modified
silica beads may be photoactivated using a sensitizer in the
solution together with an electron relay: an amine is unhooked
from a sulfonamide substrate attached to the silica. Furthermore
it was found that a doubly heterogeneous sensitized system in
which the sensitizer and the sulfonamide substrate are attached
to different silica beads works even better: an electron relaying
shuttle ensures communication between the beads. This “synaptic
like” system, which has not been explored as yet, opens the way
to new kinds of photochemical processes or investigations. This
surface to surface communication is, to our knowledge, the first
example of a photoinduced electron transfer organic reaction
photosensitized on one surface and relayed to another one. A
nonphotochemical system related to our work may however be
cited: an alcohol was shown to be alternately reduced and
oxidized on the surfaces of silica and alumina beads charged with
appropriate redox reagents.7
sulfonamide by itself and, on the other hand, the photochemical
reaction is very inefficient in its absence. The hydride is probably
involved in the sensitizer regeneration and a salt effect helping
ion pair separation is also possible.
Four different experimental systems (Scheme 2) were used:
homogeneous reaction (system A), heterogeneous system with
sensitizer (system B) or substrate (system C) covalently grafted
on silica, and the doubly heterogeneous system with sensitizer
and substrate grafted on different silica beads (system D). (See
Table 1 for results.)
That the homogeneous method (A) allows a complete depro-
tection with a 100% transformation into amine and sulfinic acid
can be seen in runs 1 and 3. The sulfonamide group is selectively
cleaved while the amide function of 3a remains unaffected: the
concentration of recovered amine correctly matches the sulfon-
amide consumption.
In this paper the photoreductive deprotection of sulfonamides8,9
is taken as a test reaction for the study of heterogeneous
conditions: in our example, 1,4-dimethoxynaphthalene (1a)
sensitizes the cleavage of the sulfonamides 2 and 3 in the presence
of potassium borohydride coreductant. Scheme 1 embodies this
photodeprotection from which â-phenethylamine (âA) is recov-
ered together with p-toluene sulfinic acid. The first step of the
mechanism is without doubt a photoinduced electron transfer
(PET) reaction from the excited sensitizer toward the accepting
sulfonamide: the fluorescence of excited 1a is quenched by 2 or
3a with Stern-Volmer constants () product of quenching rate
constant by the 1a* lifetime) of 19 and 71 M-1, respectively.
The difference between the two values is due to the better electron
accepting character of 3a. The exact role played by the hydride
is almost unknown: it is unable to perform the cleavage of the
If the sensitizer concentration is raised above 1.5 × 10-4 M,
the situation rapidly deteriorates: the sensitizer is consumed and
secondary products are found, explaining bad yields such as in
run 2. The secondary reactions were found to involve excited
and ground state 1a. By fixing the sensitizer on a surface these
reactions are suppressed. Runs 6 and 7 (system B) show that the
reaction works, even if slower (compare runs 1 and 6 using 2 as
substrate). The fluorescence quenching of 1b by 3a has been
measured with a Stern-Volmer constant of 14 M-1 showing a
decrease of reactivity when going to the heterogeneous system
B. However, in this system and as shown in run 8, the sensitizer
concentration limit is no longer operative as a benefit of using
an immobilized sensitizer. In this run, the sensitizer has been
filtered, washed, and successfully recycled: the efficiency of the
grafted sensitizer remained unchanged. According to the estima-
tion of the 1b silica loading, a turnover of the sensitizer of 80 is
found as a minimum.
(1) Horner, L.; Klaus, J. Justus Liebigs Ann. Chem. 1981, 792-810.
(2) Avnir, D.; Wellner, E.; Ottolenghi, M. J. Am. Chem. Soc. 1989, 111,
2001-2003.
(3) Julliard, M.; Legris, M.; Chanon, M. J. Photochem. Photobiol. A: Chem.
1991, 61, 137-152.
(4) Julliard, M.; Chanon, M. Bull. Soc. Chim. France 1992, 129, 242-
246.
(5) Julliard, M. New J. Chem. 1994, 18, 243-250.
(6) (a) Ayadim, M.; Soumillion, J. Ph. Tetrahedron Lett. 1995, 36, 4615-
4618 (b) Ayadim, M.; Soumillion, J. Ph. Tetrahedron Lett. 1996, 37, 381-
384.
(7) Kim, B.; Regen, S. L. Tetrahedron Lett. 1983, 24, 689-690.
(8) Umezawa, B.; Sawaki, S.; Oshino, O. Chem. Pharm. Bull. 1970, 18,
182-185.
(9) Hamada, T.; Nishida, A.; Yonemitsu, O. J. Am. Chem. Soc. 1986, 108,
140-145.
The sulfonamide substrate has been attached to the surface of
a silica and used in system C. Run 9 again shows the adverse
effect of an increase of 1a concentration in solution, with a large
amount of consumed sensitizer. In run 10, traces of products are
10.1021/ja990316a CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/23/1999

Communications to the Editor
J. Am. Chem. Soc., Vol. 121, No. 44, 1999 10437
Table 1. Homogeneous and Heterogeneous Photosensitized
of the naphthalene anion-radical. Comparison between these runs
shows that the relaying system works and significantly enhances
the yield. It is worth mentioning that the homogeneous reaction
(system A) works better in the absence of the electron relaying
N (see runs 4 and 5): the reaction yield is lowered by a factor of
about 3 in the presence of N, and this means that N effectively
intercepts 1a* but that the shuttling efficiency is insufficient for
maintaining the reaction yields. Efficient electron relay cosensi-
tizations are known11,12 and their functioning was analyzed.13 It
is striking to see that even when not efficient in homogeneous
reactions, electron relay systems may be useful in heterogeneous
systems.
Experiment 15 clearly shows that a doubly heterogeneous
sensitized system may be used (system D), with even better results
than in system C. Of course, all the reaction conditions necessary
to keep the N radical anion working must be met (runs 12-18):
the remaining silanol groups of 1c and 3c were removed by
trimethylsilylation and the methanol solvent was replaced by
DMF. Under those conditions, the obtained deprotection yield
was probably only limited by the sulfonamides remaining
accessible on the surface of 3c. Longer irradiation times did not
liberate larger amounts of amines and this shows that the available
surface sulfonamides of 3c are completely photocleaved. It is not
unreasonable to think that a significant fraction of the grafted
sulfonamide may be buried in holes closed by the silylating
treatment: we have shown, in another work, that on a highly
loaded grafted silica about one-third of the grafted molecules
remain accessible.6b The same masking phenomenon may deac-
tivate a fraction of the sensitizer in 1c. If this is true, our doubly
heterogeneous system is still more efficient than expected.
In this reaction, two heterogeneous electron transfers are taking
place successively. What is presented here is of interest in systems
where a surface needs to be modified by a remote photoactivation
as in cases where the sensitizer and the surface are chemically
incompatible. In the present work, the surface is not sensitive to
the direct action of the sensitizer and is made sensitive to its
remote activation by way of a shuttling electron carrier.
Cleavage of Sulfonamidesa
runb
sens.c
time (h)
SAc,d
relaye
aminec,f,g
1 (A)
2 (A)
3 (A)
4 (A)
5 (A)
6 (B)
7 (B)
8 (B)
9 (C)
10 (C)
11 (C)
12 (D)
13 (D)
14 (D)
15 (D)
16 (D)
17 (D)
18 (D)
1a (0.5)
1a (1.5)
1a (0.7)
1a (0.5)
1a (0.5)
1b (0.6)
1b (0.6)
1b (1.8)h
1a (14)i
1a (0.5)
1a (0.5)
1b (0.6)
1b (0.6)
1b (0.6)
1c (<0.6)
-
35
4.5
28
0.5
0.5
20
28
45
5.5
23
23
28
28
28
28
28
28
28
2 (15)
2 (36)
3a (50)
3a (30)
3a (30)
2 (15)
3a (48)
3a (50)
3b (31)
3c (<20)
3c (<20)
3b (31)
3b (31)
3b (31)
3c (<20)
3c (<20)
3c (<20)
3c (<20)
-
-
-
-
+
-
-
-
-
-
+
-
+
+
+
+
+
-
15
3.0
50.0
7.5
2.5
1.5
47.0
50.0
0.2
<0.04
2.5
0
0
0
7.3
0
1c (<0.6)
1c (<0.6)
0j
0
a 5 mL reaction mixture irradiated with black light phosphor lamps
(350 nm, 12 W, 15 cm from sample). Degassing under argon. Solvent:
MeOH in runs 1-3, 6-9, 12, and 13. DMF in other runs. b See Scheme
2. c µmol in the 5 mL reacting volume. d SA ) sulfonamide. e Naph-
thalene electron relay; 500 µmol in reacting volume. f Recovered
â-phenethylamine. g In runs 1 and 3-8: 100% recovery of amine versus
consumed sulfonamide. In run 2: 30%. h Sensitizer successfully reused
after this run. i 66% of sensitizer consumed after reaction. j No KBH4
in this run.
Scheme 3
only obtained and a considerable drop of the reactivity is observed
(compare runs 4 and 10). Remembering that a PET reaction needs
contact between the reactants, the lifetime of excited 1a (around
6 ns)10 is probably not sufficient to allow it to reach the surface
bearing the immobilized sulfonamide: the deactivation of the
excited state is much faster than its diffusion to the solid particles.
A solution to this particular problem may be a shuttling
molecule. This one should accept an electron from excited 1a
and carry it to the substrate anchored to the silica beads. Excited
1a is quenched by naphthalene (N) through a PET process10 with
a Stern-Volmer constant of about 2 M-1. This makes N a good
candidate for the electron relay, working as shown in Scheme 3
with an anion-radical of sufficient lifetime.
Acknowledgment. The authors thank the FNRS for the financial
support. M. Ayadim acknowledges the Morocco Ministry of Higher
Education and Catholic University of Louvain for fellowships. J.L.H.J.
is “Chercheur Qualifie´” from the FNRS. The authors are indebted to Dr.
O. B. Nagy for helpful discussions.
Supporting Information Available: Experimental details, including
the systhesis proedures for 1b, 3a, and 3b (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
JA990316A
(11) Arnold, D. R.; Snow, M. S. Can. J. Chem. 1988, 66, 3012-3026.
(12) Schaap, A. P.; Lopez, L.; Gagnon, S. D. J. Am. Chem. Soc. 1983,
105, 663-664.
(13) Vauthey, E.; Pilloud, D.; Haselbach, E.; Suppan, P. Chem. Phys. Lett.
1993, 215, 264-268.
Runs 10 and 11 were performed in a nonprotic solvent and
the grafted silica was trimethylsilylated to avoid the destruction
(10) Habib Jiwan, J. L.; Soumillion, J. Ph. J. Phys. Chem. 1995, 99, 14223-
14230.