Water-soluble dendrimers as photochemical reaction media: Chemical behavior of singlet and triplet radical pairs inside dendritic reaction cavities

Article abstract of DOI:10.1021/ja049492w

Water-soluble poly(alkyl aryl ether) dendrimers have been explored for their use as hosts of organic substrates in aqueous media. Prototypical photoreactions, namely, photo-Fries reaction of (a) 1-naphthyl benzoate and (b) 1-naphthyl phenyl ester and α-cleavage reaction of (a) dibenzyl ketones and (b) benzoin alkyl ethers, have been examined. We find that a dendritic microenvironment not only restricts the mobility of radical intermediates but also rigidly encapsulates the substrate, intermediates, and products from "leaking" to the bulk environment. Comparative studies of the same photoreactions in micellar media demonstrate that dendritic media offer much better constrainment than the micelles.

Full text of DOI:10.1021/ja049492w

Published on Web 07/03/2004
Water-Soluble Dendrimers as Photochemical Reaction Media:
Chemical Behavior of Singlet and Triplet Radical Pairs Inside
Dendritic Reaction Cavities
Lakshmi S. Kaanumalle,† Jayaraj Nithyanandhan,‡ Mahesh Pattabiraman,†
Narayanaswamy Jayaraman,*,‡ and V. Ramamurthy*,†
Contribution from the Department of Chemistry, Tulane UniVersity,
New Orleans, Louisiana 70118, and Department of Organic Chemistry, Indian Institute of
Science, Bangalore 560012, India
Received January 28, 2004; E-mail: murthy@tulane.edu
Abstract: Water-soluble poly(alkyl aryl ether) dendrimers have been explored for their use as hosts of
organic substrates in aqueous media. Prototypical photoreactions, namely, photo-Fries reaction of (a)
1-naphthyl benzoate and (b) 1-naphthyl phenyl ester and R-cleavage reaction of (a) dibenzyl ketones and
(b) benzoin alkyl ethers, have been examined. We find that a dendritic microenvironment not only restricts
the mobility of radical intermediates but also rigidly encapsulates the substrate, intermediates, and products
from “leaking” to the bulk environment. Comparative studies of the same photoreactions in micellar media
demonstrate that dendritic media offer much better constrainment than the micelles.
Introduction
these dendrimers endow them the term “static, unimolecular
micelles”.5
A branches-upon-branches architectural motif remains to be
the most appealing structural feature of dendritic macromol-
ecules. Ever since the prominence of dendrimers in the literature1
over a decade ago, the structural features and the associated
molecular and supramolecular properties have attracted a wide
spectrum of investigations in a number of chemical, biological,
and materials related studies involving dendritic macromol-
ecules.2 A densely packed periphery and well-defined cavities
of different sizes at the interior result in the so-called “exo-”
and “endo-” receptor properties in dendrimers, and the benefits
of exo- and endo- receptor properties have been explored in a
number of studies.3 The evolution of microenvironments upon
advancing the generation number of a homologous series of
dendrimers is a characteristic feature of dendritic architecture.
Particularly, the creation of microenvironments is more profound
in those dendrimers that contain a polar exterior and nonpolar
interior. Analogous to such microenvironments in regular
micelles,4 the presence of distinct microenvironments in
Clearly, the possibilities of encapsulation of organic molecules
open up avenues to exploit the dendritic microenvironment in
varied studies. We envisaged that such a microenvironment
could be utilized as reaction media, in a manner comparable to
well-known hosts such as cyclodextrins and micelles.6 We herein
conceptualize this idea by invoking photochemically induced
reactions of organic molecules encapsulated by dendrimers.
Specifically, we describe encapsulation of water-insoluble
organic guest molecules in aqueous dendrimer medium, derived
from poly(alkyl aryl ether) dendrimers, and the photochemical
reactions of these guest molecules inside the dendritic microen-
vironments. Based on photoreactions of four organic molecules
and the photophysics of pyrene, we establish that poly(alkyl
aryl ether) dendrimers provide hydrophobic reaction cavities
similar to that of micelles but constrain the mobility of guest
molecules more than the ones provided by micelles. Given the
current emphasis on “green chemistry”, use of water as the
medium and light as the reagent is most appealing.7 We utilize
in this study dendrimers as vehicles to solubilize organic
† Tulane University.
‡ Indian Institute of Science.
(1) (a) Buhleier, E.; Wehner, W.; Vo¨gtle, F. Synthesis 1978, 155-158. (b)
Tomalia, D. A.; Baker, H.; Dewald, J. R.; Hall, M.; Kallos, G.; Martin, S.;
Roeck, J.; Ryder, J.; Smith, P. Polym. J. (Tokyo) 1985, 17, 117-132. (c)
Newkome, G. R.; Yao, Z.-Q.; Baker, G. R.; Gupta, V. K. J. Org. Chem.
1985, 50, 2003-2004. (d) Hawker, C. J.; Fre´chet, J. M. J. J. Am. Chem.
Soc. 1990, 112, 7638-7647.
(2) (a) Top. Curr. Chem. 1998, 197. (b) Top. Curr. Chem. 2000, 210. (c) Top.
Curr. Chem. 2001, 212. (d) Top. Curr. Chem. 2002, 217.
(3) (a) Smith, D. K.; Diederich, F. Chem. ReV. 1904, 4, 1353-1361. (b) Hecht,
S.; Fre´chet, J. M. J. Angew. Chem., Int. Ed. 2001, 40, 74-91. (c) Bosman,
A. W.; Janssen, H. M.; Meijer, E. W. Chem. ReV. 1999, 99, 1665-1688.
(d) Newkome, G. R.; He, E.; Moorefield, N. Chem. ReV. 1999, 99, 1689-
1746. (e) Inoue, K. Prog. Polym. Sci. 2000, 25, 453-571. (f) Gorman, C.
B.; Smith, J. C. Acc. Chem. Res. 2001, 34, 60-71. (g) Oosterom, G. E.;
Reek, J. N. H.; Kamer, P. C. J.; van Leeuwen, P. W. N. M. Angew. Chem.,
Int. Ed. 2001, 40, 1828-1849.
(4) Moroi, Y. Micelles, Theoretical and Applied Aspects; Plenum: New York,
1992.
(5) (a) Naylor, A. M.; Goddard, W. A., III; Kiefer, G. E.; Tomalia, D. A. J.
Am. Chem. Soc. 1989, 111, 2339-2341. (b) Newkome, G. R.; Moorefield,
C. N.; Baker, G. R.; Johnson, A. L.; Behera, R. K. Angew. Chem., Int. Ed.
Engl. 1991, 30, 1176-1178. (c) Gopidas, K. R.; Leheny, A. R.; Camianati,
G.; Turro, N. J.; Tomalia, D. A. J. Am. Chem. Soc. 1990, 112, 8515-
8522. (d) Hawker, C. J.; Wolley, K. L.; Fre´chet, J. M. J. J. Chem. Soc.,
Perkin Trans. 1 1993, 1287-1297. (e) Pesak, D. J.; Moore, J. S.; Wheat,
T. E. Macromolecules 1997, 30, 6467-6482. (f) Piotti, M. E.; Rivera, F.,
Jr.; Bond, R.; Hawker, C. J.; Fre´chet, J. M. J. Am. Chem. Soc. 1999, 121,
9471-9472.
(6) (a) Photochemistry in microheterogeneous systems; Kalyanasundaram, K.,
Ed.; Academic Press: New York, 1987. (b) Photochemistry in Organized
& Constrained Media; Ramamurthy, V., Ed.; VCH: New York, 1991.
(7) Anastas, P. T.; Kirchhoff, M. M. Acc. Chem. Res. 2002, 35, 686-694.
9
10.1021/ja049492w CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 8999-9006
8999

A R T I C L E S
Kaanumalle et al.
Figure 1. Structures of the dendrimers used in this study.
molecules in water and as restrictive reaction media to control
photoproduct distribution.
the first, second, and third generation dendrimers possess,
respectively, 6, 12, and 24 phenolic hydroxyl groups at their
peripheries. These dendrimers are soluble in water in their
phenoxide form at a pH above 9. Prior to utilizing the
dendrimers as reaction media, knowledge of hydrophilic-
hydrophobic balance and the micropolarity arising due to
hydrophilic exterior and the relatively hydrophobic interior is
essential. To probe the micropolarity of the interior of den-
drimers, we have employed pyrene as the probe.9 Pyrene is
soluble in water only to the extent of 1 × 10-6 M. The
fluorescence spectrum shown in Figure 2 reveals that even under
Results and Discussion
The dendrimers of choice to test them as reaction media are
the poly(alkyl aryl ether) dendrimers reported recently (Figure
1).8 These dendrimers are constituted with phloroglucinol, acting
as the core and branching component, and a pentamethylene
chain, which acts as the linker component. In this constitution,
(8) Nithyanandhan, J.; Jayaraman, N. J. Org. Chem. 2002, 67, 6282-6285.
9
9000 J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004

Dendrimers as Photochemical Reaction Media
A R T I C L E S
Scheme 1
Table 1. Ratio of I1/I3 Bands of Pyrenea in Dendrimers and Water
vances from first to third. For example, while the polarity
experienced by pyrene in aqueous solution of G1 (pH ∼9) is
similar to that of dioxane, that in aqueous solution of G3 (pH
∼9) is similar to that of ethanol. This effect is most likely due
to an increase in the number of hydrophobic cavities as the
dendrimer generation advance and to the better ability of cavities
derived from higher generations to encapsulate the probe
molecule. PAMAM dendrimers are also known to exhibit a
similar property.10
[host], M
I1/I3
I1/I3 for solventsb
2 × 10-4 (G3 dendrimer)
2 × 10-4 (G2 dendrimer)
2 × 10-4 (G1 dendrimer)
water pH > 9
1.18
1.30
1.48
1.70
1.18 (ethanol)
1.35 (methanol)
1.50 (dioxane)
1.87 (water)
a [Pyrene] ) 1 × 10-5 M. b Literature values.
such low concentrations probably pyrene exists as aggregates
in water (note the presence of emission in the region 430-500
nm due to a ground-state dimer). As reported earlier, the
solubility of pyrene in aqueous solution, as determined by UV-
vis spectroscopy, increases by a factor of 5, 8, and 24 for the
first, second, and third generation dendrimers (100 µM each),
respectively.8 Emission spectra displayed in Figure 2 show the
Results presented above clearly suggest that the poly(alkyl
aryl ether) dendrimers, especially the G3, contain hydrophobic
pockets (in water) in which one could conduct chemical
reactions. To explore this possibility, we have carried out four
photochemical reactions, namely, photo-Fries reaction of 1-naph-
thyl benzoate and 1-naphthyl phenylacyl ester and R-cleavage
reaction of dibenzyl ketones and benzoin alkyl ethers. In these
four reactions, a single molecule is photochemically cleaved
into two pieces (radical pair), and dependent on the mobility of
the two reactive species, the composition of the final products
varies. As discussed below, the final products obtained within
the hydrophobic pockets of poly(alkyl aryl ether) dendrimers
are much more selective than those in common organic solvents.
The first reaction examined is the photo-Fries reaction of
1-naphthyl benzoate (1) (Scheme 1).11,12 This molecule upon
irradiation in hexane gives 2-benzoyl 1-naphthol (2) and
4-benzoyl 1-naphthol (3) in the ratio 3:1 (Table 2). However,
in basic aqueous medium (pH ∼9), this molecule undergoes
hydrolysis to yield naphthol and benzoic acid (dark reaction).
When an aqueous solution of G3 dendrimer (2 × 10-4 M) was
added to the aqueous solution (pH ∼9; 5 mL) containing 4 ×
10-4 M of 1-naphthyl benzoate, no hydrolysis occurred even
after 12 h. This suggested that the G3 dendrimer protects the
reactant from water and OH- ions by encapsulating it within
its hydrophobic pockets. However, addition of an aqueous
solution of G1 dendrimer (2.3 × 10-3 M) had led to the thermal
hydrolysis process (Table 2). These observations are consistent
with the hydrophobicity information provided by the probe
pyrene (G1 is less hydrophobic than G3). Irradiation of 1-naph-
thyl benzoate included in an aqueous solution of G3 resulted in
the same two products as in hexane, but the ratio was much
different (19:1 as opposed to 3:1 in hexane; Table 2). Similar
higher selectivity in favor of 2-benzoyl 1-naphthol was observed
Figure 2. Fluorescence spectrum of pyrene in (red) water (pH 9.5) and
(green) aqueous solution of G3 dendrimer (200 µM, pH 9.5).
absence of a ground-state dimer of pyrene when G3 dendrimers
(2 × 10-4 M) were added to the aqueous basic solution. The
absence of the ground-state dimer was also confirmed for
aqueous solutions containing G1 and G2 dendrimers. The absence
of emission due to a ground-state dimer and excimer in
dendrimer media suggests that pyrene molecules are kept
isolated by the latter, and within the excited-state lifetime, pyrene
in the excited singlet state is unable to interact with another
pyrene molecule present in the ground state.
It is well established that the ratio of the intensities of the
vibrational peaks 1 and 3 (I1 to I3) in the fluorescence spectrum
of pyrene is an excellent measure of the polarity of its immediate
environment.9 I1/I3 values measured from the fluorescence
spectra of pyrene in aqueous dendritic media are presented in
Table 1. A comparison of the I1/I3 of dendrimers vs I1/I3 of
common solvents indicates that pyrene experiences an increas-
ingly nonpolar environment as the dendrimer generation ad-
(10) Pistolis, G.; Malliaris, A.; Paleos, C. M.; Tsiourvas, D. Langmuir 1997,
13, 5870.
(11) (a) Miranda, M. A. In CRC Handbook of Organic Photochemistry and
Photobiology; Horspool, W. M., Song, P. S., Eds.; CRC Press: Boca Raton,
FL, 1995; pp 570-578.
(12) (a) Cui, C.; Weiss, R. G. J. Am. Chem. Soc. 1993, 115, 9820-9821. (b)
Pitchumani, K.; Warrier, M.; Cui, C.; Weiss, R. G.; Ramamurthy, V.
Tetrahedron Lett. 1996, 37, 6251-6254. (c) Pitchumani, K.; Warrier, M.;
Ramamurthy, V. J. Am. Chem. Soc. 1996, 118, 9428-9429. (d) Pitchumani,
K.; Warrier, M.; Ramamurthy, V. Res. Chem. Intermed. 1999, 25, 623-
631.
(9) (a) Kalyanasundaram, K.; Thomas, J. K. J. Am. Chem. Soc. 1977, 99, 2039-
44. (b) Dong, D. C.; Winnik, M. A. Photochem. Photobiol. 1982, 35, 17-
21.
9
J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004 9001

A R T I C L E S
Kaanumalle et al.
Scheme 2
Table 2. Product Distribution upon Irradiation of Naphthyl
Table 3. Product Distribution upon Photolysis of Dimethyl
Naphthyl Phenyl Acetate in Various Mediaa,b,c
Benzoatea,b,c
c
c
c
medium
2
3
naphthol
benzil
benzoic acid
medium
5
6
7
8
9
10
11
12
hexane
aq NaOHd
G3
62
1
95
93
9
24
10
62
4
hexane
aq NaOHd
aq NaOHe
G3
35
17
4
21
8
43
69
8
6
37
5
7
27
96
93
49
92
80
83
3
4
7
9
G2
G1
G1
64
65
27
25
G2
G1
d
e
39
4
5
3
1
2
2
NaDCh
NaCh
SDS
99
85
83
1
15
17
NaDCh
NaCh
SDS
2
1
1
2
10
13
2
2
a [G1] ) 22.5 × 10-4 M, [G2] ) 8 × 10-4 M, [G3] ) 2 × 10-4 M, [1]
) 4.0 × 10-4 M. b Mean occupancy numbers: NaDCh, NaCh ) 0.2-0.25
(15-17 monomers per aggregate); SDS ) 0.7-1.0 (70 monomers per
micelle). c For structures, see Scheme 1. d Product distribution resulting from
dark (thermal) reaction of 1.
a [G1] ) 22.5 × 10-4 M, [G2] ) 8 × 10-4 M, [G3] ) 2 × 10-4 M, [4]
) 3.42 × 10-4 M. b Mean occupancy numbers: NaDCh, NaCh ) 0.17-
0.23; SDS ) 0.42-0.67. c For structures, see Scheme 2. d Product distribu-
tion resulting from dark (thermal) reaction. e Product distribution resulting
from thermal reaction and photoreaction.
when an aqueous solution of G2 (8 × 10-4 M) was used as the
host. The higher selectivity observed with G2 and G3 as hosts
suggests that the singlet radical pair, naphthoxyl and benzoyl,
formed upon cleavage of 1-naphthyl benzoate has much less
mobility within the hydrophobic pockets of dendrimers than in
hexane. The short lifetime of the singlet radical pair and probable
high microviscosity of the hydrophobic pockets of the den-
drimers probably do not allow migration of benzoyl radical from
the 1 to 4 position of the naphthoxyl radical.
The second photoreaction we investigated was the cleavage
reaction of 1-naphthyl phenylacyl ester 4 (Scheme 2).13 Pho-
tolysis of 1-naphthyl phenylacyl ester 4 in hexane resulted in a
mixture of seven products, yields of which are presented in
Table 3. Similar to 1-naphthyl benzoate discussed above,
1-naphthyl phenylacyl ester 4 underwent thermal hydrolysis in
basic aqueous medium (pH ∼9; 3.4 × 10-4 M). This thermal
reaction once again was prevented by addition of aqueous
solutions of G2 and G3 dendrimers to the above solution. Most
interestingly, when the irradiation of 1-naphthyl phenylacyl ester
4 in basic aqueous solution containing 8 × 10-4 M G2 or 2 ×
10-4 M G3 was conducted, only two products were obtained
(Table 3). Of the two products (2- and 4- isomers), 2-phenylacyl
1-naphthol 5 predominated the product mixture. The selectivity
obtained in the presence of dendrimers is dramatic, going from
seven products in hexane solution to essentially a single product,
2-phenylacyl 1-naphthol 5, in the presence of dendrimers. The
above product selectivity could be understood based on the
mechanism provided in Scheme 2. Homolytic cleavage of the
ester occurs from their excited singlet state to yield geminate
singlet radical pair A. It is followed by in-cage recombination
(giving rearranged products 5 and 6) or cage escape. Decarbo-
nylation of phenylacyl radical within the cage results in the
radical pair B and outside the cage leads to naphthoxyl and
benzyl free radicals. Reaction of these radicals either within or
outside the cage leads to products 7-10. Diaryl ethane 11 is
the result of coupling of the two dimethyl benzyl radicals outside
the cage.
(13) (a) Gu, W.; Warrier, M.; Ramamurthy, V.; Weiss, R. G. J. Am. Chem.
Soc. 1999, 121, 9467-9468. (b) Gu, W.; Hill, A. J.; Wang, X.; Cui, C.;
Weiss, R. G. Macromolecules 2000, 33, 7801-7811. (c) Gu, W.; Weiss,
R. G. Tetrahedron 2000, 56, 6913-6925. (d) Gu, W.; Weiss, R. G. J.
Photochem. Photobiol. C 2001, 2, 117-137. (e) Gu, W.; Weiss, R. G. J.
Org. Chem. 2001, 66, 1775-1780. (f) Gu, W.; Bi, S.; Weiss, R. G.
Photochem. Photobiol. Sci. 2002, 1, 52-59. (g) Warrier, M.; Kaanumalle,
L. S.; Ramamurthy, V. Can. J. Chem. 2003, 81, 620-631.
Absence of products 7-11 implies that the primary radical
pair A does not give the radical pair B. This suggests that the
recombination of the naphthoxyl and phenylacyl radicals within
9
9002 J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004

Dendrimers as Photochemical Reaction Media
A R T I C L E S
Scheme 3
Table 4. Product Distribution on Photolysis of Benzoin Ethyl
a dendrimer must be rapid compared to their escape from a
dendrimer and decarbonylation of the phenylacyl radical. The
rate of decarbonylation of 1,1′-dimethyl 2-phenylacyl radical
Ethera,b,c
c
medium
14
15
16
17
18
19
hexane
40
13
32
3
56
78
63
82
31
1
4
6
5
5
4
57
30
29
41
17
20
41
in solution has been estimated to be 1.5 × 108 s-1 14
. Therefore,
methanol
benzene
water
2
the rate of recombination of the radical pair A must be faster
than the above rate. Further, the hydrophobic pockets of G2 and
G3 dendrimers must not be “leaky” in the time scale of 6 ×
10-9 s. Selective formation of 2-phenylacyl 1-naphthol (5) from
1-naphthyl phenylacyl ester 4 is consistent with the behavior
of 1-naphthyl benzoate 1 discussed above and suggests that the
reaction cavity of a dendrimer restricts the mobility of the radical
pair A.
10
65
21
41
44
50
25
12
12
aq NaOH
a
G3
G3
8
8
6
5
2
2
1
14
15
16
4
d
7
4
G2
G1
NaDCh
NaCh
SDS
1
7
55
66
39
To address the question whether the hydrophobic pockets of
G2 and G3 dendrimers in water can restrict the mobility of guest
molecules (and reactive intermediates) even on a longer time
scale, we have explored the photochemistry of benzoin ethyl
ether (13)15 and 1-phenyl-3-para-tolyl-propane-2-one (20).16
These two systems cleave from the triplet state to yield triplet
radical pairs. Under such conditions, coupling of the radical
pair can occur only if it can cross to the singlet radical pair or
separate to become free radicals. Results presented below on
these two systems suggest that reaction cavities of dendrimers
are “leak-proof” even on a microsecond time scale.
Benzoin ethyl ether (13) upon irradiation in solution yields
products of Norrish type I and type II processes from the triplet
state (Scheme 3; Table 4).14 Generally, in isotropic solvent
media, products of the Norrish type I process predominate the
product mixture. In solution, R-cleavage leads to benzoyl and
R-alkoxybenzyl triplet radical pair C (Scheme 3) which gener-
ally escapes the cage to yield benzil (14) and pinacol ether 15.
Minor amounts (<10%) of Norrish-Yang cyclization product
18 and deoxy benzoin (19) via γ-hydrogen abstraction are also
formed (Scheme 3). In solution, the rearrangement products,
para-benzoylbenzyl ethyl ether (16) and ortho-benzoylbenzyl
ethyl ether (17), through coupling of the benzoyl and R-alkoxy-
benzyl radical pair C are not obtained. Product distributions in
a [G1] ) 14 × 10-4 M, [G2] ) 11 × 10-4 M, [G3] ) 4.8 × 10-4 M,
[13] ) 2.9 × 10-4 M. b Mean occupancy numbers: NaCh, NaDCh ) ∼0.2;
SDS ) ∼0.6. c For structures, see Scheme 3. d [G3] ) 2.4 × 10-4 M.
neutral and basic aqueous media (pH ∼9) provided in Table 4
clearly indicate that the Norrish type I process is predominant
in these media, and this process does not yield the product of
in-cage reaction, 16 and 17. Irradiation of benzoin ethyl ether
(3 × 10-4 M) in basic aqueous solution (pH ∼9) containing G1
(14 × 10-4 M), G2 (11 × 10-4 M), and G3 (4.8 × 10-4 M)
dendrimers resulted in a product mixture different from that in
neutral or basic aqueous media (Table 4). Several observations
with respect to its photobehavior in dendrimer media are
noteworthy from the results presented in Table 4: (a) Products
of the Norrish type II process which were almost insignificant
(<10%) in organic solvents became the major product (∼70%)
in dendrimer media. (b) The rearrangement products 16 and
17, the ones not obtained in solution, are formed in significant
amounts (∼20%) in the presence of G2 and G3 dendrimers. (c)
Pinacol ether 15, the resultant of an out of cage reaction, a major
product in organic solvents, is formed in negligible amounts
(<2%) in the presence of dendrimers.
Formation of para-benzoylbenzyl ethyl ether and ortho-
benzoylbenzyl ethyl ether in significant amounts (∼20%) in the
presence of dendrimers indicates that R-cleavage reaction does
take place in this media. Absence of pinacol ethers under high
concentrations of dendrimers suggests that the only reaction
open to the benzoyl and R-alkoxybenzyl radical pair is the
recombination to yield the reactant benzoin ethyl ether and the
rearranged benzoylbenzyl ethyl ethers 16 and 17. Formation of
(14) (a) Turro, N. J.; Gould, I. R.; Baretz, H. B. J. Phys. Chem. 1983, 87, 531-
532. (b) Lunazzi, L.; Ingold, K. U.; Scaiano, J. C. J. Phys. Chem. 1983,
87, 529-530.
(15) Lewis, F. D.; Lauterbach, R. T.; Heine, H.-G.; Hartmann, W.; Rudolph,
H. J. Am. Chem. Soc. 1975, 97, 1519-1525.
(16) (a) Robbins, W. K.; Eastman, R. H. J. Am. Chem. Soc. 1970, 92, 2, 6076-
6079. (b) Engel, P. S. J. Am. Chem. Soc. 1970, 92, 6074-6075.
9
J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004 9003

A R T I C L E S
Kaanumalle et al.
Scheme 4
Table 5. Photolysis of 1-Phenyl-3-p-tolyl-propan-2-onea,b,c
Norrish type II products in major amounts in the presence of
dendrimer, we believe, is facilitated by the suppression of
Norrish type I products due to recombination of the benzoyl
and R-alkoxybenzyl radicals to yield the reactant benzoin alkyl
ether. It is quite likely that at the hydrophobic-hydrophilic
interface of a dendrimer-water, benzoin ethyl ether adopts a
conformation that is suited for γ-hydrogen abstraction.17 Further
work is underway to probe this possibility.
cage
c
medium
AA
AB
BB
effectd
21
hexane
21
21
4
51
53
80
69
67
54
60
54
71
29
26
6
16
17
28
21
21
17
0.05
0.06
0.77
0.46
0.42
0.11
0.24
0.18
0.42
aq NaOH
e
G3
G3
10
5
5
3
6
a
9
G2
G1
NaDCh
NaCh
SDS
11
15
14
17
12
The last reaction we have investigated in the context of
probing the mobility of guests within the hydrophobic pockets
of dendrimers is the Norrish type I reaction of 1-phenyl-3-para-
tolyl-propane-2-one (20) (Scheme 4).15 Irradiation of 1-phenyl-
3-para-tolyl-propane-2-one in hexane solution results in R-cleav-
age (yielding the primary radical pair D) to be followed by
decarbonylation (to give the secondary radical pair E). In hexane
solution, no product from the radical pair D has been detected.
The three diaryl ethanes AA, AB, and BB resulting from the
radical pair E are formed in the ratio of 1:2:1 (Scheme 4). In
restricted environments, the rearrangement product 21 is formed
from the radical pair D (Scheme 4).18 When the radical pair E
is held within a cage (with a little translational mobility), the
only product expected is AB.19 The cage effect (AB - AA -
BB)/(AA + AB + BB) and the yield of the rearrangement
product 21 provide information concerning the “leakiness” of
the reaction cavity with respect to the radical pairs D and E.
Results of photolysis of 1-phenyl-3-para-tolyl-propane-2-one
in various media are summarized in Table 5. Clearly, in the
presence of dendrimers, AB is formed in much larger amounts
than in hexane. Observation of higher cage effect (>0.4) in the
presence of dendrimers than that in hexane or water (<0.05)
indicates that the reaction cavity in dendrimers is much more
restrictive than that in isotropic solution. The fact that the cage
effects are less than one in G3 and G2 dendrimers suggests that
either the hydrophobic pockets of dendrimers are not totally
“leak-proof” with respect to the radical pair E or each dendrimer
molecule is multiply occupied. The key experiment of control-
ling the occupancy number by increasing the dendrimer
concentration could not be carried out owing to absorption
problems. At high concentrations (8 × 10-4 M) of the
dendrimer, the presence of a phenyl chromophore on the
8
3
a [G1] ) 30 × 10-4 M, [G2] ) 8 × 10-4 M, [G3] ) 2 × 10-4 M, [20]
) (2.67-3.42) × 10-4 M. b Mean occupancy numbers: NaCh, NaDCh )
∼ 0.11-0.14; SDS ) 0.32-0.37. c For structures, see Scheme 4. d Cage
effect ) (AB - AA - BB)/(AA + AB + BB). e [G3] ) 9 × 10-4 M.
backbone of the dendrimer prevented absorption by the reactant
1-phenyl-3-para-tolyl-propane-2-one. In the limited concentra-
tion region investigated (2 to 9 × 10-4), the cage effect increased
from 0.46 to 0.77. In the absence of further study, we cannot
be certain whether the observed change is due to multiple
occupancy at lower concentration of the dendrimer or due to
some amount of reaction taking place in water. A similar
observation with respect to cage escape was made during
irradiation of benzoin ethyl ether 13 included in G3 (Scheme 3
and Table 4). For example, the yield of pinacol ether 15
decreased from 7% to 1% when the G3 concentration increased
from 2.4 × 10-4 to 4.8 × 10-4M. Based on the behavior of the
radicals from benzoin ethyl ether, we believe that the radical
pair E from 1-phenyl-3-para-tolyl-propane-2-one may also
prefer to stay within the hydrophobic pockets of the dendrimer.
One should note that the escape rate would also depend on the
hydrophobicity of the guest molecule (or the guest reactive
intermediate). Further work is needed to resolve this issue.
As indicated in the Introduction section, dendrimers are often
considered to be “static unimolecular micelles”. Given this
feature, in the context of reaction media, we were interested in
comparing the reaction cavity provided by dendrimers to that
by sodium dodecyl sulfate (SDS), sodium cholate (NaCh), and
sodium deoxycholate (NaDCh) micelles.20 Product distributions
(20) (a) Turro, N. J.; Gra¨tzel, M.; Braun, A. M. Angew. Chem., Int. Ed. Engl.
1980, 19, 675. (b) O’Connar, C. J.; Wallace, R. G. AdV. Colloid Interface
Sci. 1985, 22, 1-111. (c) Small, D. M. AdV. Chem. Ser. 1968, 84, 31-52.
(d) Fisher, L.; Oakenfull, D. Aust. J. Chem. 1979, 32, 31-39. (e) Lindman,
B.; Kamenka, N.; Fabre, H.; Ulmius, J.; Wieloch, T. J. Colloid Interface
Sci. 1980, 73, 556-565. (f) O’Connor, C. J.; Wallace, R. G. AdV. Colloid
Interface Sci. 1985, 22, 1-111 (g) Ju, C.; Bohne, C. J. Phys. Chem. 1996,
100, 3847-3854. (h) Sugioka, H.; Moroi, Y. Biochim. Biophys. Acta 1998,
1394, 99-110. (i) Matsuoka, K.; Kuranaga, Y.; Moroi, Y. Biochim. Biophys.
Acta 2002, 1580, 200-214. (j) Ninomiya, R.; Matsuoka, K.; Moroi, Y.
Biochim. Biophys. Acta 2003, 1634, 116-125. (k) Rinco, O.; Nolet, M.-
C.; Ovans, R.; Bohne, C. Photochem. Photobiol. Sci. 2003, 2, 1140-1151.
(17) (a) Devanathan, S.; Ramamurthy, V. J. Phys. Org. Chem. 1988, 1, 91-
102. (b) de Mayo, P.; Nakamura, A.; Tsang, P. W. K.; Wong, S. K. J. Am.
Chem. Soc. 1982, 104, 6824-6825.
(18) Frederick, B.; Johnston, L. J.; de Mayo, P.; Wong, S. K. Can. J. Chem.
1984, 62, 403-410.
(19) (a) Turro, N. J. Proc. Natl. Acad. Sci. U.S.A. 1983, 80, 609-621. (b) Gould,
I. R.; Turro, N. J.; Zimmt, M. B. In AdVances in Physical Organic
Chemistry; Bethell, D., Ed.; Academic Press: London, 1984; Vol. 20, pp
1-53. (c) Turro, N. J. Chem. Commun. 2002, 2279-2292. (d) Turro, N. J.
Acc. Chem. Res. 2000, 33, 637-646.
9
9004 J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004

Dendrimers as Photochemical Reaction Media
A R T I C L E S
for all four reactions discussed above obtained in micellar media
above critical micelle concentration are included in Tables 2-5.
Clearly the photoreaction is much less selective within con-
ventional micelles than in dendrimers. Irradiation of 1-naphthyl
benzoate in SDS and NaCh micelles gave both 2-benzoyl
1-naphthol (2) and 4-benzoyl 1-naphthol (3) in the ratio 4:1, a
slight improvement from that in hexane (Table 2). However,
within NaDCh micelles, 2-benzoyl 1-naphthol was selectively
obtained. Analogous behavior was observed with 1-naphthyl
phenylacyl ester 4 (Table 3). Within SDS and NaCh micelles,
a mixture of 2-phenylacyl 1-naphthol 5 and 4-phenylacyl
1-naphthol 6 was obtained, whereas, within NaDCh micelles,
2-phenylacyl 1-naphthol alone was obtained. The higher selec-
tivity observed in NaDCh micelles is consistent with the
established higher rigidity of NaDCh micelles in relation to
NaCh micelles.20b-k Photobehavior of benzoin ethyl ether in
all three micelles (occupancy number, i.e., number of molecules
per micelle: in SDS, 0.6; in bile salts micelles, 0.2) are different
from that in dendrimers (Table 4). Most importantly, in all three
micelles, the pinacol ether 15, a product of cage escape, is
formed in significant amounts (>25%). On the other hand, when
dendrimer was used as the medium, no pinacol ether was formed
at a higher dendrimer-to-reactant ratio. This suggests that
dendrimers are much less “leaky” than a conventional micelle.
Photochemistry of 1-phenyl-3-para-tolyl-propane-2-one within
SDS micelles has been extensively investigated.21 The cage
effect (∼0.4) that we recorded in the SDS micelle is close to
the literature value. Both NaCh and NaDCh micelles provided
a lesser cage effect (∼0.20; Table 5). Clearly the cage effects
observed in dendrimers are much higher than those in NaCh
and NaDCh micelles.
Figure 3. Absorption spectrum of dendrimer G3 (see Figure 1 for structure
of the dendrimer).
Experimental Section
Materials: Poly(alkyl aryl ether) dendrimers used for the study were
synthesized and characterized following the literature procedure.8
Sodium dodecyl sulfate (99%), sodium cholate (98%), and sodium
deoxycholate (98%) procured from Sigma-Aldrich were used without
further purification. Naphthyl benzoate (1) and 1-naphthyl phenylacyl
ester 4 were prepared following literature procedures.13g Benzoin ethyl
ether (13) obtained from Sigma-Aldrich was recrystallized twice from
hexane. Substrate 20 was synthesized as described in the literature.23
Inclusion of Reactants within Dendrimers and Photolysis: The
procedure adopted for all substrates were similar and one of them is
described below. Substrate 1 (0.5 mg, 4 × 10-4 M) was added to a
stirred solution of a known amount of dendrimer in 5 mL of aqueous
NaOH ([G1] ) 22.5 × 10-4 M, [G2] ) 8 × 10-4 M, [G3] ) 2 × 10-4
M) . After the solution was stirred for 12 h, it was filtered through a
Whatmann filter paper (medium porosity) to remove any floating
particles. Filtrate was purged with nitrogen for 30 min and then
irradiated in a Pyrex tube with a 450 W medium-pressure Hg lamp.
Irradiation for 2 h resulted in ∼30% conversion in the case of 1, 4,
and 13. In case of substrate 20, the sample in aqueous G1 solution was
irradiated for 4 h, the sample in aqueous G2 solution, for 7 h, and the
sample in aqueous G3 solution, for 12 h, to obtain 30% conversion.
Absorption by the dendrimer was responsible for the low conversion.
As presented above, in the context of “green chemistry”, the
dendrimers are better media than micelles to perform product-
selective photochemistry.22 The presence of small amounts (10-4
M) of dendrimers can dissolve a large amount of organic
molecules in water (pH ∼9). Based on four reactions listed in
this report, we conclude that dendrimers provide a better
hydrophobic environment than micelles and the reaction cavities
of dendrimers incarcerate the reactants and the intermediates
for a much longer time than a conventional micelle. Further,
dendrimers have a higher capacity to solubilize organic com-
pounds than micelles. Following photolysis, products could be
more easily extracted from a dendrimer than from a micelle,
the dendrimer itself can be precipitated and reused by making
the medium neutral. The selectivity in photoproducts obtained
and convenience of their use make dendrimers better media than
conventional micelles to perform selective photoreactions.
However, one should note the dendrimers described in this study
absorb between 220 and 320 nm (Figure 3), and this could
restrict their use with molecules absorbing only in this region.
We are currently exploring the use of water-soluble dendrimers
and other organic hosts as media for photoreactions.
Extraction of Photoproducts and Reactants from Dendrimer
Aqueous Solution: After photolysis, the basic aqueous solution was
acidified with 10% dilute HCl. Reactants and products were extracted
from aqueous solution using an ethyl acetate and acetonitrile (7:3)
solvent mixture, dried over anhydrous Na2SO4, concentrated, and
analyzed on an HP-5890 series II gas chromatograph fitted with an
SE-30 or HP-5 column. A known amount of internal standard was added
before analysis for mass balance studies. For substrates 1, 4, and 20,
dodecane was used as the internal standard, and for substrate 13,
benzophenone was used as the internal standard.
Characterization of Photoproducts: Peaks in the GC traces were
identified by coinjecting with authentic samples which were prepared
by solution irradiation. Spectral data of photoproducts from 1 and 4
prepared by solution irradiation were compared with literature reports.13
Among the photoproducts from 13, benzaldehyde, 14, and 19 were
identified by comparison with the commercially available samples
(Aldrich). Photoproducts 15 and 18 isolated from solution irradiation
and 16 and 17 isolated from dendrimer irradiation were characterized
1
by H NMR and GC-MS.
1H NMR of 15 (400 MHz, CDCl3): δ 0.94 (t, 6 H), δ 2.83-3.65
(m, 4 H), δ 4.30 (s, 2 H), δ 7.09-7.58 (m, 10 H). Mass spectral data
m/z (relative intensity): 165 (4), 152 (2), 136 (9), 135 (100), 107 (66),
79 (48), 77 (32), 51 (10).
(21) (a) Turro, N. J.; Cherry, W. R. J. Am. Chem. Soc. 1978, 100, 7431-7432.
(b) Turro, N. J.; Weed, G. C. J. Am. Chem. Soc. 1983, 105, 1861-1868.
(c) Turro, N. J.; Zimmt, M. B.; Lei, X. G.; Gould, I. R.; Nitsche, K. S.;
Cha, Y. J. Phys. Chem. 1987, 91, 4544-4548. (d) Turro, N. J.; Chow,
M.-F.; Chung, C.-J.; Kraeutler, B. J. Am. Chem. Soc. 1981, 103, 3886-
3891. (e) Turro, N. J.; Anderson, D. R.; Chow, M.-F.; Chung, C.-J.;
Kraeutler, B. J. Am. Chem. Soc. 1981, 103, 3892-3896. (f) Gould, I. R.;
Zimmt, M. B.; Turro, N. J.; Baretz, B. H.; Lehr, G. F. J. Am. Chem. Soc.
1985, 107, 4607-4612.
1H NMR of 16 (400 MHz, CDCl3): δ 1.2 (t, 3 H), δ 3.51 (q, 2 H),
δ 4.57 (s, 2 H), δ 7.40-7.80 (m, 9 H). Mass spectral data m/z (relative
(23) Rabjohn, N., Ed.; Organic Synthesis; Wiley: New York, 1963; Vol. IV, p
(22) Hecht, S.; Frechet, J. M. J. J. Am. Chem. Soc. 2001, 123, 6959-6960.
176.
9
J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004 9005

A R T I C L E S
Kaanumalle et al.
intensity): 240 (M+, 18), 211 (18), 196 (34), 181 (5), 167 (49), 166
(100), 152 (10), 133 (21), 126 (14), 105 (77), 89 (24), 77 (50).
1H NMR of 17 (400 MHz, CDCl3): δ 1.0 (t, 3 H), δ 3.38 (q, 2 H),
δ 4.58 (s, 2 H), δ 7.30-7.80 (m, 9 H). Mass spectral data m/z (relative
intensity): 240 (M+, 9), 211 (59), 194 (28), 181 (3), 165 (25), 152
(12), 133 (100), 105 (28), 77 (49), 51 (17).
1H NMR of 18 (400 MHz, CDCl3): δ 1.4 (d, 3 H), δ 5.2 (q, 1 H),
δ 5.95 (s, 1 H), δ 7.20-7.80 (m, 10 H). Mass spectral data m/z (relative
intensity): 196 (17), 167 (28), 152 (9), 134 (85), 133 (62), 118 (12),
105 (100), 91 (15), 77 (42).
organic layers were washed several times with water and dried over
anhydrous MgSO4. The filtered solution was concentrated and analyzed
on an HP-5890 series II gas chromatograph fitted with an SE-30 or
HP-5 column.
The following conditions were used for GC analysis of the
photoproducts. The temperature of the injection and the detection ports
were maintained at 225 °C and 250 °C.
Substrate 1. Column: HP-5. Temperature program: initial temp, 100
°C; initial time, 1 min; rate, 5 °C/min; final temp, 220 °C; final time,
10 min. Retention times: 1-naphthol, 10.7 min; benzil, 16.8 min;
benzoic acid, 6.8 min; ortho rearrangement product, 26.6 min; para
rearrangement product, 33.8 min.
In the case of 20, the photoproducts AA and BB were commercially
available (Aldrich and Lancaster). Photoproduct AB was identified
based on the GC-MS fragmentation pattern.
Substrate 4. Column: HP-5. Temperature program: initial temp, 100
°C; initial time, 1 min; rate, 5 °C; final temp, 270 °C; final time, 10
min. Retention times: 10, 10.7 min; 11, 16.6 min; 12, 21.8 min; 7,
25.4 min; 4, 27.0 min; 8, 27.4 min; 5, 28.4 min; 6, 31.7 min.
Mass spectral data m/z (relative intensity): 196 (M+, 17), 105 (100),
91 (12), 77 (11).
Rp 21 was identified by isolating the compound from photolysis of
dendrimer samples.
Substrate 13. Column: SE-30. Temperature program: initial temp,
70°C; initial time, 1 min; rate, 5 °C/min; final temp, 175 °C; final time,
30 min. Retention times: 19, 25.5 min; 15, 26.1 min; 14, 28.6 min;
13, 30.0 min; 17, 32.3 min; 18, 34.3 min; 16, 43.4 min.
1H NMR of 21 (400 MHz, CDCl3): δ 2.3 (s, 3 H), δ 2.4 (s, 3 H),
δ 4.19 (s, 2 H), δ 7.09-7.24 (m, 6 H), δ 7.87-7.88 (m, 6 H). Mass
spectral data m/z (relative intensity): 224 (M+, 6), 119 (100), 105 (6),
91 (22), 77 (4), 65 (9).
Substrate 20. Column: SE-30. Temperature program: initial temp,
100 °C; initial time, 1 min; rate, 5 °C/min; final temp, 270 °C; final
time, 10 min. Retention times: AA, 14.3 min; AB, 17 min; BB, 19.3
min; 20, 22.3 min; 21, 24.3 min.
Inclusion of Substrates within Micelles and Photolysis: Stock
solutions of known concentration (3-5 mM) of the reactants 1, 4, 13,
and 20 in CH2Cl2-hexane were prepared. A volume of solutions
corresponding to 0.5-1.0 mg of the substrates was pipetted into a test
tube. The solution was purged with air to remove the solvent, and a
thin film of the substrate was obtained. Weighed amounts of the
surfactants (100-200 mg) were added to this followed by 5 mL of
deionized water (in the case of SDS) or 5 mL of 0.2 M aq. NaCl (in
the case of bile salts). After the solutions were stirred for 3 h, they
were filtered through a Whatmann filter paper, and the filtrate was taken
in a test tube, purged with dry N2 for 15 min, and irradiated in a Pyrex
test tube (40 min for 1 and 4, 10 min for 13, and 20 min for 20). The
conversions were maintained under 30%.
Fluorescence Measurements: Fluorescence spectra were recorded
at room temperature on an Edinburgh FS900CDT steady-state fluori-
meter. The concentration of pyrene aqueous solution used in the
fluorescence measurement is 1 × 10-5 M. The emission spectrum was
recorded by excitation at 335 nm. The excitation spectrum was recorded
at the 385 nm emission wavelength. To the aqueous solution of pyrene,
dendrimer solution was added such that the concentration of dendrimer
was 2 × 10-4 M in solution. The solution was stirred for 3 h prior to
recording the emission/excitation spectrum.
Extraction of Photoproducts and Starting Material: After ir-
radiation, the solutions were diluted with deionized water to a surfactant
concentration well below their critical micelle concentration. The
photoproducts were extracted by using an organic solvent. In the case
of SDS, the extracting solvent was a 10% ether-hexane mixture, and
the separatory funnel was shaken carefully to avoid the formation of
emulsion. The extraction procedure was repeated at least thrice. Ethyl
acetate was used for extraction in the case of bile salt micelles. The
Acknowledgment. V.R. thanks the National Science Founda-
tion for support of the research (CHE-9904187 and CHE-
0213042). N.J. acknowledges financial support from Council
of Scientific and Industrial Research (CSIR), India, and J.N.
thanks CSIR, India, for a research fellowship.
JA049492W
9
9006 J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004