Alkali metal ion controlled product selectivity during photorearrangements of 1-naphthyl phenyl acylates and dibenzyl ketones within zeolites

Article abstract of DOI:10.1139/v03-041

Photochemical behaviors of 1-naphthyl phenyl acylates and dibenzyl ketones included in zeolites have been compared. 1-Naphthyl phenyl acylates while in solution produce eight photoproducts; within NaY it gives a single product. The selectivity is attributed to the restriction brought on the mobility of the primary radical pair by the alkali metal ions present in zeolites. Photochemistry of dibenzyl ketones within NaY reveals that the intersystem crossing in caged radical pairs could be influenced by the heavy alkali metal ions. Structures of complexes among Li⁺ ion and the guest 1-naphthyl phenyl acetates and dibenzyl ketone computed at the B3LYP level have been useful to understand the origin of the observed product selectivity within zeolites.

Full text of DOI:10.1139/v03-041

620
Alkali metal ion controlled product selectivity
during photorearrangements of 1-naphthyl phenyl
acylates and dibenzyl ketones within zeolites
M. Warrier, Lakshmi S. Kaanumalle, and V. Ramamurthy
Abstract: Photochemical behaviors of 1-naphthyl phenyl acylates and dibenzyl ketones included in zeolites have been
compared. 1-Naphthyl phenyl acylates while in solution produce eight photoproducts; within NaY it gives a single
product. The selectivity is attributed to the restriction brought on the mobility of the primary radical pair by the alkali
metal ions present in zeolites. Photochemistry of dibenzyl ketones within NaY reveals that the intersystem crossing in
caged radical pairs could be influenced by the heavy alkali metal ions. Structures of complexes among Li+ ion and the
guest 1-naphthyl phenyl acetates and dibenzyl ketone computed at the B3LYP level have been useful to understand the
origin of the observed product selectivity within zeolites.
Key words: photo-Fries reaction, zeolites, cation–π interaction, spin-orbit coupling, heavy atom effect.
Résumé : On a comparé les comportements photochimiques des phénylacylates de 1-naphtyle et des dibenzylcétones
inclus dans des zéolites. Les phénylacylates de 1-naphtyle en solution conduisent à huit photoproduits alors qu’il ne
s’en forme qu’un lorsqu’ils sont inclus dans NaY. On attribue la sélectivité à la restriction imposée à la paire de radi-
caux primaires par les ions des métaux alcalins présents dans les zéolites. La photochimie des dibenzylcétones dans
NaY révèle que le passage entre systèmes dans une paire de radicaux en cage peut être influencé par les ions des mé-
taux alcalins lourds. Les structures des complexes entre l’ion Li+ et les molécules de phénylacétates de 1-naphtyle et
de dibenzylcétones qui ont été calculées au niveau B3LYP se sont avérées utiles pour comprendre l’origine de la sélec-
tivité des produits qui a été observée pour les réactions dans les zéolites.
Mots clés : réaction de photo-Fries, zéolites, interaction cation–π, couplage spin-orbite, effet d’atome lourd.
[Traduit par la Rédaction] Warrier et al. 631
Introduction
result in both the ortho- and the para-isomers (Scheme 2).
However, the factors that control the outcome of the prod-
ucts vary with the nature of the medium. In solution, it is the
electron densities at various aromatic carbons in the aryloxy
radical which control the regioselectivity. We suggested that
the selectivity within zeolites resulted from the restriction
imposed on the mobility of the aryloxy and the acetyl frag-
ments by the cations present in zeolites (2). Vasenkov and
Frei (4) provided support for the existence of Na+-acetyl
radical interaction within solvent-free NaY through time-
resolved FT-IR spectroscopy. They also confirmed that only
2-acetyl-1-naphthol is formed upon photolysis of 1-naphthyl
acetate within solvent-free NaY. The acetyl radicals gener-
ated by photolysis of 1-naphthyl acetate (reacting from ex-
cited singlet state) within NaY decay following a single
exponential law with a lifetime of 71 ± 15 µs. The acetyl
radical generated from pinacolone (reacting from triplet
state) has a much longer lifetime (315 ± 30 µs). The differ-
ence in lifetime between singlet and triplet radical pairs sug-
gested that the reactivity of the reactive radical generated
from the excited singlet and triplet states of the precursors
Photo-Fries rearrangement of phenylacetate and photo-
Claisen rearrangement of allylphenyl ether yield ortho-
hydroxy and para-hydroxy isomers as products (Scheme 1)
(1). In solution, independent of the polarity of the medium,
one obtains a mixture. On the other hand, within alkali metal
ion exchanged X and Y zeolites, ortho-hydroxy isomers are
obtained as the major products from phenylacetate and allyl-
phenyl ether (2). Selectivity obtained within zeolites was
very high even during the photo-Fries rearrangement of 1-
naphthyl acetate and 1-naphthyl benzoate (Scheme 1) (3).
The results with 1-naphthyl esters suggested that zeolites
could favor the formation of a single product (2-acetyl-1-
naphthol and 2-benzoyl-1-naphthol) by discriminating
among three possible photoproducts, all of which may be
formed with equal probability in isotropic media.
Both photo-Fries and photo-Claisen rearrangements pro-
ceed via a similar mechanism (1). Promotion to the excited
singlet state results in fragmentation of the ester and the
ether. Cage escape, recombination, and hydrogen migration
Received 13 January 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 28 May 2003.
Dedicated to Professor Donald R. Arnold for his contributions to Chemistry.
M. Warrier, L.S. Kaanumalle, and V. Ramamurthy.1 Department of Chemistry, Tulane University, New Orleans, LA 70118 U.S.A.
1Corresponding author (e-mail: murthy@tulane.edu).
Can. J. Chem. 81: 620–631 (2003)
doi: 10.1139/V03-041
© 2003 NRC Canada

Warrier et al.
621
Scheme 1.
Scheme 2.
would behave differently. Further, the long lifetime (µs) of
the acetyl radical suggested that the ortho and (or) para selec-
tivity in photo-Fries reaction would depend on the nature of
the “spin” of the primary radical pair. To test these possibili-
ties and to further explore the selectivity in photo-Fries reac-
tions in naphthyl systems we have exmined the photobehavior
of three 1-naphthyl phenylacyl esters 1a–1c and three
dibenzyl ketones 2a–2c (Scheme 3) (5). The study mainly
consisted of photoproducts analysis. To gain an insight into
the alkali ion binding to the reactants within zeolites density
functional theory (DFT) calculations were performed. Results
of these studies are presented in this report.
Results
Photolysis of 1-naphthyl phenylacyl esters 1a–1c
Photolysis of 1-naphthyl phenylacyl esters 1a–1c in hex-
ane resulted in a mixture of products (Scheme 4), yields of
which are presented in Table 1. The product distributions ob-
tained in this study is identical to the ones reported by Gu
and Weiss (6) recently. Photolysis of 1a–1c within different
cation-exchanged zeolites, however, showed formation of
mainly the ortho- rearrangement product (2-phenylacyl-1-
naphthol (3)) accompanied by minor amounts of 1-naphthol
(8) (Table 2). Percentage conversion for the same length of
© 2003 NRC Canada

622
Can. J. Chem. Vol. 81, 2003
Scheme 3.
Scheme 4.
irradiation and for the same loading level of 1a–1c within
zeolites decreased from light alkali metal ion-exchanged
zeolites (e.g., LiY) to heavy alkali metal ion-exchanged
zeolites (e.g., TlY) (Table 2). Based on our earlier studies
(7), facile intersystem crossing leading the reactive excited
singlet esters into their unreactive triplet state was suspected
for the decrease in reactivity within heavy alkali metal ion-
exchanged zeolites. To ascertain this possibility, emission
spectra of 1a–1c in different alkali metal ion-exchanged
zeolites at 77 K were recorded. As seen in Fig. 1 for 1-
naphthyl phenylacetate, the emission was mainly fluores-
cence in LiY and NaY and mainly phosphorescence in CsY
and TlY. Similar emission behavior was obtained in the case
of 1b and 1c.
Photolysis of dibenzyl ketones 2a–2c
Photolysis of dibenzyl ketones 2a–2c in hexane yielded
only the decarbonylation products 1,2-diphenyl ethane, 2,3-
diphenyl butane, 2,3-diphenyl 2,3-dimethylbutane, respec-
tively (Scheme 4, Table 3). These results are consistent with
the original reports by Engel (8a), Robbins and Eastman
(8b), and Turro and co-workers (8c, 8d). The same products
obtained in hexane were formed upon photolysis of 2a and
2b in light-atom-exchanged zeolites (e.g., LiY and NaY)
(Table 3). Photolysis of 2a and 2b in heavy-atom-exchanged
zeolites such as CsY and TlY resulted in rearrangement
products, 2-methyl-phenyl benzyl ketone and 4-methyl-
phenyl benzyl ketone, and 2-ethyl-phenyl α-methyl benzyl
ketone and 4-ethyl-phenyl α-methyl benzyl ketone in addi-
© 2003 NRC Canada

Warrier et al.
623
Table 1. Product distribution upon photolysis of substrates 1a–1c in hexane.
Product yieldsa
Compound
3
4
5
6
7
8
9
10
6
9
15
9
2
13
2
2
15
1a
1b
1c
59
59
28
7
9
9
11
17
14
1
1
—
5
1
6
aIrradiations were conducted to about 40% conversion. The conversion was achieved in 15 min. All yields
presented are an average of at least five independent runs. Error limits on yields were ±5%.
Table 2. Product distribution upon photolysis of 1-naphthyl phenyl acylates (1a–1c) within zeolites.
Medium
Conversiona (%)
3a (%)
8a (%)
Conversiona (%)
3b (%)
8b (%)
Conversiona(%)
3c (%)
LiY
NaY
KY
RbY
CsY
TlY
27
24
15
9
3
<1
97
97
96
92
—
—
3
3
4
8
—
—
19
26
11
7
2
<1
>99
>99
93
95
87
—
—
7
5
13
—
46
24
35
<1
<1
—
>99
>99
>99
—
—
—
—
aAll irradiations were carried out for 2 h in a hexane slurry. The ratio of products were independent of the conversion in the range 15–90%. All yields
are an average of at least five independent runs. Error limit on yields were ±5%.
Fig. 1. Emission spectra of 1-naphthyl phenylacyl ester (1a) in various cation exchanged zeolites at 77 K.
tion to 1,2-diphenylethane and 2,3-diphenylbutane, respec-
tively (Scheme 4) (9). The relative yields of the
rearrangement products (2-alkyl and 4-alkyl-phenyl benzyl
ketone) with respect to the decarbonylated products were de-
pendent on the rates of decarbonylation of the phenyl acyl
radical which in turn depended on the stability of the
benzylic radicals formed. For example, in the case of
dibenzyl ketone (2a), which forms, after decarbonylation, a
primary benzyl radical, yield of the rearrangement products
was as high as 79% in TlY zeolite (Table 3). In the case of
© 2003 NRC Canada

624
Can. J. Chem. Vol. 81, 2003
Table 3. Product distribution upon photolysis of dibenzyl ketones (2a–2c) in hexane and Y zeolites.
Medium
11aa (%)
12a (%)
13a (%)
11ba (%)
12b (%)
13b (%)
11cb (%)
Hexane
LiY
NaY
KY
RbY
CsY
TlY
100
95
76
90
79
33
21
—
1
2
7
3
—
4
8
17
18
54
22
100
>99
>99
>99
95
—
—
—
—
—
—
7
—
—
—
—
5
>99
>99
>99
>99
>99
>99
>99
13
57
76
75
24
18
aAll yields are an average of at least five independent runs. Error limit on yields were ±5%.
bIn addition to 11c, cumene and α-methylstyrene are formed as minor products. The yield on this column represents all three.
α,α,α′,α′-tetramethyl dibenzyl ketone (2c), where a stable
substituted tertiary benzyl radical is formed, the yield of the
rearrangement products was <1% in TlY (Table 3).
puted values. Nevertheless, the interaction between the al-
kali metal ion and the substrates is expected to be significant
within zeolites.
Discussion
Alkali metal ion interaction with 1-naphthyl phenylacyl
esters and dibenzyl ketones probed through
Details of the photo-Fries reaction of 1-naphthyl phenyl-
acyl esters in solution and polymer medium have recently
been discussed by Gu, Weiss, and co-workers (5). Product
distributions obtained upon photolysis of esters 1a–1c in
hexane could be understood based on the mechanism pro-
vided in Scheme 5. Homolytic cleavage of the esters occurs
from their excited singlet state to yield singlet geminate rad-
ical pair A. It is followed by in-cage recombination (giving
rearranged products 3 and 4) or cage escape. Decarbonyl-
ation of phenylacyl radical within the cage results in the rad-
ical pair B and outside the cage leads to naphthoxy and
benzyl free radicals. Reactions of these two radicals either
within or outside the cage result in products 5–8, and prod-
uct 9 is the result of coupling between two alkyl radicals
outside the cage. In this study we make use of the photo-
Fries reaction of 1-naphthyl phenylacyl esters to establish
that zeolites are capable of influencing the product distribu-
tions in a reaction that gives as many as eight products. Fur-
ther, the reaction allows us to probe the characteristics of
zeolite cages, knowledge of which could be further used to
control photoreactions. All our conclusions are based on
photoproduct distributions and attempts to monitor the rates
of reactions through time-resolved diffuse reflectance laser
techniques were not successful. The following aspects of the
mechanism presented in Scheme 5 are noteworthy. The rela-
tive ratio of products 3 to 4, and 5 to 6 provide information
about the rotational mobility of reactive fragments present in
cages A and B (Scheme 5). The relative yields of products 8
and 9 give an indication about the translational mobility of
radical fragments. The relative selectivity observed in the
photo-Fries products 3 and 4 and the decarbonylated prod-
ucts 5 and 6 provides valuable information concerning re-
strictions imposed by the zeolite on different time scales.
When 1-naphthyl phenylacyl esters 1a-1c were irradiated
in hexane, eight photoproducts (shown in Scheme 4) were
formed and the relative yields remained the same independ-
ent of the conversion in the range 5–40% (Table 1). Upon
photolysis of 1-naphthyl phenylacyl esters 1a–1c within LiY,
NaY, and KY zeolites led essentially to a single product 3
(Table 2). The selectivity was dramatic, going from eight
products in hexane solution to a single product (2-
phenylacyl 1-naphthol (3)) in zeolites. For this to happen,
computations
To gain a better insight on the role of cations (present
within zeolite supercages) on the observed photobehavior of
naphthyl esters and dibenzyl ketones within zeolites, density
functional theory (DFT) calculations on Li+ binding to 1-
naphthyl phenyl acetate and dibenzyl ketone were carried
out (10). The polarized 6-31G* basis set was used for C, H,
O, and Li. In view of the complexities involved in the reac-
tions inside the zeolite cage, the goals of the computational
work were necessarily limited to a few specific aspects.
First, the magnitude of the interaction energies among Li+
and 1-naphthyl phenyl acetate and dibenzyl ketone were
quantified. This provided a measure of the potential role of
different cations in altering the reactivity of 1-naphthyl
phenyl acetate and dibenzyl ketone. In addition to the en-
ergy, the structures of lithium ion – 1-naphthyl phenyl ace-
tate and lithium ion – dibenzyl ketone complexes were
analyzed. Interaction energies of Li+ were determined based
on gas-phase optimized geometries of metal ion complexes
with 1-naphthyl phenyl acetate and dibenzyl ketone.
The computed geometries of 1-naphthyl phenyl acetate
and dibenzyl ketone are shown in Figs. 2 and 3. In both the
cases, two low energy conformers were identified and these
were used for complexation with lithium ion. Intuitively,
binding of Li+ is expected to be higher with the substrate
when it interacts simultaneously with either two aryl rings or
an aryl ring and a carbonyl chromophore. Thus, we have
chosen initial guess geometry with the Li+ positioned be-
tween the two functional groups. During the calculation the
cations were free to move to find the most stable position.
Calculations did not include any structural moiety from the
zeolite. In the case of 1-naphthyl phenyl acetate, three geom-
etry-optimized structures obtained upon metal ion
complexation along with their interaction energies are shown
in Fig. 2. Interaction energies of Li+ to 1-naphthyl phenyl
acetate were almost similar in all three structures. Two ge-
ometry-optimized structures for the lithium ion complex
with dibenzyl ketone were obtained (Fig. 3). Since within a
zeolite, the cations are bound to oxygen counter ions, the
metal ions are likely to interact less effectively with 1-
naphthyl phenyl acetate and dibenzyl ketone than the com-
© 2003 NRC Canada

Warrier et al.
625
Fig. 2. Conformations of 1-naphthyl phenylacyl ester (top 2 structures) and its complex with Li+ (bottom three) as computed at the
RB3LYP/6-31G(d) level.
Fig. 3. Conformations of dibenzyl ketone (top 2 structures) and its complex with Li+ (bottom two) as computed at the RB3LYP/6-
31G(d) level.
the two primary radicals in A formed upon photofragment-
ation must be coupling before the decarbonylation of phen-
ylacyl radical. Thus, two questions need to be addressed:
(a) Why is the ortho-isomer (2-phenylacyl 1-naphthol 3)
alone formed?; and (b) Why do the primary phenylacyl (and
related) radicals not decarbonylate within zeolites, or why is
the coupling of radical in cage A faster than decarbony-
lation?
© 2003 NRC Canada

626
Can. J. Chem. Vol. 81, 2003
Scheme 5.
If one accepts that the two reactive fragments, naphthoxy
and phenylacyl radicals, would be held close to each other
by the intrazeolite cations, preference for ortho coupling to
yield 3a–3c over the para coupling to yield 4a–4c is under-
standable. This suggestion is based on the optimized struc-
tures for the Li+ – 1-naphthyl phenyl acetate complex shown
in Fig. 2. Photofragmentation of the Li+ – 1-naphthyl phenyl
acetate complex present in these structural forms would lead
to naphthoxy and phenylacyl radicals being held close to
each other by the alkali metal ion. Lack of separation en-
forced by the zeolite wall and the alkali metal ion would
place the two radicals, having opposite electronic spins (fa-
vorable for bond formation), close to one another. Due to re-
stricted mobility enforced by the alkali metal ion, the attack
of the acyl fragment would occur at the nearest available po-
sition on the aromatic ring, namely the 2-position of the
naphthoxy ring.
The lack of formation of products 5–9 implies that the pri-
mary radical pair A do not proceed to the radical pair B.
This suggests that the recombination of the naphthoxy and
phenylacyl radicals within a zeolite must be rapid compared
with cage diffusion and decarbonylation of the phenylacyl
radical. The rates of decarbonylation of phenylacyl and 1-
methyl 2-phenylacyl and 1,1′-dimethyl 2-phenylacyl radicals
have been estimated to be 4.8 × 106 s–1, 4.0 × 107 s–1, and
1.5 × 108 s–1, respectively (11). It is important to note that
even in 1c where the intermediate 1,1′-dimethyl 2-phenylacyl
radical decarbonylates ~30 times faster than phenylacyl radi-
cal, no decarbonylation products were obtained within
zeolites and the only product obtained was 2-phenylacyl 1-
naphthol 3c. Given this, the rate of diffusion of the radical
pair A out of the cage must be smaller than the maximum
rate of decarbonylation observed in the case of 1,1′-dimethyl
2-phenylacyl radical (1.5 × 108 s–1). In the absence of data
relating to the rate of decarbonylation within zeolites, we
can only speculate that the lack of products due to decar-
bonylation and radical pair B is most likely due to the fact
that the rate of recombination in the case of cage pair A is
faster than the rates of decarbonylation and cage escape. In-
tuitively then, if the recombination step can be slowed the
decarbonylation could become competitive. This led us to
explore the photochemistry of dibenzyl ketones, where the
same phenylacyl radical would be produced as part of a trip-
let radical pair. Results of these are discussed below.
At this point, a discussion on the lack of reactivity of the
naphthyl esters within heavy alkali metal ion exchanged
zeolites is appropriate. An important observation within CsY
and TlY was that no photoproducts from 1a–1c were ob-
served in these zeolites. Enhanced intersystem crossing effi-
ciency in heavy cation exchanged zeolites and the fact that
these esters are unreactive in their triplet states explain the
lack of reactivity within CsY and TlY. Evidence for inter-
system crossing can be found in the emission spectra for the
ester 1a in different cation exchanged zeolites shown in
Fig. 1. Fluorescence predominates in light atom zeolites like
LiY and NaY. Within KY and RbY, phosphorescence in-
creases at the expense of fluorescence. In TlY there is only
phosphorescence, suggesting complete population of the
triplet state. Consistent with this, naphthyl esters 1a–1c were
unreactive within TlY.
Upon excitation, dibenzyl ketones similar to 1-naphthyl
phenylacyl esters, yield products via two stages of radical
pair formation, α-cleavage to yield primary caged radical
pair A and decarbonylation to give the secondary free radi-
cals (or radical pair B) (Scheme 6) (12). An important dif-
ference among 1-naphthyl esters and dibenzyl ketones is that
© 2003 NRC Canada

Warrier et al.
627
Scheme 6.
the former reacts from S1 to give the singlet radical pair and
the latter from T1 to give the triplet radical pair. The radical
pair A resulting from dibenzyl ketones following inter-
system crossing can undergo coupling to yield the rearrange-
ment products 4-methylphenyl benzyl ketone and 2-
methylphenyl benzyl ketone or decarbonylate (following or
prior to intersystem crossing) to yield the caged radical pair
B or diffuse apart to give two free radicals. Since inter-
system crossing is required for the formation of rearrange-
ment products, relative yields of 4-methyl-phenyl benzyl
ketone and 2-methylphenyl benzyl ketone with respect to
decarbonylated products (diphenylethanes) provides infor-
mation concerning the extent of restriction imposed by zeo-
lite cages on longer time scales (compared with 1-naphthyl
phenylacyl esters chemistry) on the radical pair A. Further,
the dibenzyl ketone systems allow us to probe the influence
of heavy cations on the intersystem crossing process in radi-
cal pairs.
Decarbonylation to give diphenylethanes is the only reac-
tion upon irradiation of dibenzyl ketones 2a–2c in hexane
solution (Scheme 4) (12). Decarbonylation accounted for
>90% of the products when the dibenzyl ketones 2a–2c in-
cluded in zeolites LiY, NaY, and KY were photolyzed. This
behavior was distinctly different from that of the naphthyl
esters 1a–1c which did not decarbonylate within the same
zeolites. We had indicated above that if we could slow the
rate of recombination of the geminate radical pair it might
be possible to observe decarbonylation of phenylacyl radi-
cals within zeolites. Results with the dibenzyl ketones sug-
gest that we have achieved this goal. Because of the triplet
character of the radical pair, even close proximity and lack
of mobility are unable to favor the recombination of the rad-
ical pair over the decarbonylation process. The fact that the
“spin” prevents the recombination of the radical pair is evi-
dent from the dependence of the product distribution on the
alkali metal ion present in zeolites. Higher yields of rear-
rangement products (4-methylphenyl benzyl ketone and 2-
methylphenyl benzyl ketone) from dibenzyl ketone were ob-
tained within heavy alkali metal ion exchanged zeolites (Ta-
ble 3). Similar dependence of product distribution on the al-
kali metal ion was observed in the case of α,α′-dimethyl-
dibenzyl ketone 2b. Significantly, the ketone 2c gave only
2,3-diphenyl-2,3-dimethyl butane, α-methyl styrene, and
cumene in all zeolites.
Turro and Zhang (13) rationalized the variations in the
yield of the rearrangement products from 2a within MX
zeolites on the basis of “cation size effect”. Within X
zeolites the yield of the rearranged products 4-methylphenyl
benzyl ketone and 2-methylphenyl benzyl ketone were de-
pendent on the size of the cation; larger the size, higher the
yield. They proposed that smaller free volume within a cage
would restrict the mobility of enclosed radicals and thus pro-
long the lifetime of the triplet radical pair. Long lifetime
would favor intersystem crossing and recombination to yield
the rearranged products. Thus, cations with larger sizes are
predicted to yield more rearrangement products. This feature
is unlikely to play a role in Y zeolites, which contain fewer
cations. The supercage of X zeolites contains four type II
and ~3.5 type III cations whereas Y zeolites contain four
type II cations and no type III cations. Absence of type III
cations makes more room for the guest molecules. Also, if
cage free volume was a major factor, cation size would in-
fluence product distributions accordingly. Ionic radii of dif-
ferent cations are shown in Table 4. Product distributions in
KY and TlY provide an ideal case to compare and discuss
size effects on product formations during the photolysis of
2a. Both cations K+ and Tl+ are of the same size but differ in
their spin orbit coupling parameters by 2 orders of magni-
tude (Table 4). Rearrangement products are produced in
much larger yields in TlY than in KY. This suggests that size
is not the most important parameter in deciding selectivity
within Y zeolites.
A comparison of the results observed with dibenzyl ke-
tones and 1-naphthyl phenylacyl esters provides an insight
into how a zeolite may be influencing the reactivity of inter-
mediates generated within its cages. First, let us compare the
behavior of dibenzyl ketone 2a and 1-naphthyl phenylacyl
ester 1a. Both systems upon photolysis generate phenylacyl
© 2003 NRC Canada

628
Can. J. Chem. Vol. 81, 2003
Table 4. Ionic radii and spin-orbit coupling constants for differ-
ent cations exchanged into the zeolites.
mobility of the radical pair and provide a mechanism for
spin interconversion.
Ionic radius
Spin-orbit coupling constantsb for
Cation
of the cationa
the corresponding atom ζ cm–1
Summary
Li+
Na+
K+
Rb+
Cs+
Tl+
0.86
1.12
1.44
1.58
1.84
1.40
0.23
11.5
38
160
370
Photochemistry of a series of 1-naphthyl phenylacyl esters
and dibenzyl ketones were investigated in different cation
exchanged zeolites. Naphthyl esters which give eight prod-
ucts in isotropic solution give essentially a single product
within light alkali metal ion exchanged zeolites. Consistent
with the known triplet reactivity of these molecules, within
heavy alkali metal ion exchanged zeolites no photoproducts
were obtained. Most importantly, within light alkali metal
ion exchanged zeolites these molecules do not undergo
decarbonylation. The reaction stops at the stage of the pri-
mary radical pair and observed products are formed from
this radical pair. Alkali metal ion binding to the reactant es-
ter in a cooperative fashion to two parts of the reactant mole-
cule is suggested to favor recombination of the primary
radicals (A) at the nearest possible site, namely the 2-
position of the naphthoxy ring. Such effects fail to stop
decarbonylation when the primary radical pair is formed in
the triplet state. A comparison of the photobehavior of
dibenzyl ketones with that of naphthyl esters reveals this
facet of the mechanism. Results presented here suggest that
it is possible to steer the course of a photoreaction and direct
the formation of a specific photoproduct by using zeolites as
reaction media.
3410
aSee F.A. Cotton and G. Wilkinson. Advanced Inorganic Chemistry. V
ed. Wiley-Interscience, New York. 1988. p. 124 and 209.
bNumbers taken from S.L. Murov, I. Carmichael, and G.L. Hug.
Handbook of photochemistry. Marcel Dekker, New York. 1988. pp. 338–
341.
radical. In NaY while 1a quantitatively rearranged to 3, 2a
gave only 10% of the rearranged products 12 and 13
(Scheme 4). Clearly, in the later case, the phenylacyl radical
decarbonylates before it has a chance to recombine to form
the rearranged products. This difference, we believe, is the
result of the spin state of the geminate radical pair. When a
mechanism to spin interconversion is provided as with TlY,
the yields of the rearranged products from 2a increased to
~79%, suggesting that the spin barrier is the main reason for
the lack of rearrangement in 2a. An insight into the process
is gained by comparing the behaviors of naphthyl ester 1c
and dibenzyl ketone 2c. While the former gives the rear-
ranged product 3 quantitatively, the latter gives only the cor-
responding decarbonylated product 11c. In this case due to
the presence of α-methyl groups, the decarbonylation occurs
30 times faster than in unsubstituted phenyl acyl radical
(compare the rates of decarbonylation of phenylacyl and
1,1′-dimethyl 2-phenylacyl radicals: 4.8 × 106 s–1 and 1.5 ×
108 s–1) (11). Since we would have detected 1% of the rear-
ranged product by GC, we believe that the rate of Tl+ in-
duced ISC in the geminate pair (A3 to A1, Scheme 6) must
be <1 × 106 s–1. Thus, a comparison of the results observed
with dibenzyl ketones and naphthyl esters suggest that lack
of products due to decarbonylation in the case of naphthyl
esters 1a–1c is due to the fact that the primary radical pair A
is formed in the singlet state. Based on the computed struc-
tures for lithium ion – dibenzyl ketone complexes (Fig. 3),
one would expect that the radical pairs resulting from frag-
mentation of dibenzyl ketone would be held close to each
other within a zeolite cage. The structures of the complexes
being similar in dibenzyl ketone and 1-naphthyl phenyl ace-
tate one would expect the primary radical pair to be posi-
tioned in a similar manner from the two systems within a
zeolite cage facilitating the formation of the rearranged
products. The lack of rearranged products in the dibenzyl
ketone system within LiY and NaY suggests that restricting
the mobility of the radical pair near one another alone is not
sufficient. Coupling between the two radicals can occur only
when they have the singlet spin. In dibenzyl ketone systems
where the radical pair is produced in the triplet state the lim-
iting factor for coupling is the spin. When the mechanism
for spin interconversion becomes available as within heavy
cation exchanged zeolites the rearranged products are
formed. Thus, the cations serve two roles: they restrict the
Experimental
NaY zeolite was obtained from Zeolyst International, the
Netherlands. Monovalent cation exchanged (Li+, K+, Rb+,
Cs+, and Tl+) zeolites were prepared by stirring 10 g of NaY
with 100 mL of a 10% solution (in water) of the correspond-
ing metal nitrate for 12 h with continuous refluxing. The ze-
olite was filtered and washed thoroughly with distilled water.
This procedure was repeated three times. The zeolite was
then heated in an oven at 120°C for about 6 h to obtain the
cation exchanged zeolite used for photolysis. The exchange
levels were estimated by ICP analysis and they were in the
range 60–85%.
1-Naphthyl phenylacyl esters 1a and 1b were gifts from W.
Gu and R.G. Weiss of Georgetown University. 1-Naphthyl
phenylacyl ester 1c was synthesized as per Scheme 7 (14).
Step a
Lithium diisopropyl amide (0.02 mol) in dry THF
(100 mL) was stirred under nitrogen at –78°C. To this 2 g
(0.015 mol) of 2-phenyl propionaldehyde in dry THF was
added dropwise with continuous stirring. To the resulting or-
ange dispersion 2.7 g (0.02 mol) of methyl iodide was added
dropwise. The mixture was stirred for 12 h and then poured
over an ice water mixture and extracted with three 20 mL
portions of diethyl ether. The ether layer was dried with
anhyd MgSO4, filtered, and evaporated to give a crude mix-
ture of the product aldehyde (14). Pure 14 was obtained by
purification using column chromatography (silica gel and
hexane); yield: 35%. MS data: 148 (M+, 6.5), 119 (97.8),
103 (14.1), 91 (100), 77 (28.3), 65 (8.7), 51 (27.2), 41
(54.3).
© 2003 NRC Canada

Warrier et al.
629
Scheme 7.
obtained by column chromatography eluting with 2% ethyl
acetate in petroleum ether. The product was a colorless oil;
Step b
Silver oxide was prepared by adding a solution of 5.1 g
(0.03 mol) of AgNO3 in 10 mL of water to a solution of
2.4 g (0.06 mol) of NaOH in 10 mL of water. Continuous
shaking during the addition ensured complete reaction and
resulted in a brown semisolid mixture. This mixture was
cooled with an ice bath and 2 g (0.014 mol) of the aldehyde
14 was added slowly with continuous stirring. The oxidation
reaction was allowed to proceed for 2 h. The black silver
suspension was removed by suction filtration and was
washed with warm water. The combined filtrate was allowed
to cool and the washings were acidified with concentrated
hydrochloric acid precipitating about 1.8 g of the corre-
sponding acid (15); yield: 70%. MS data: 164 (M+, 23.5),
119 (100), 103 (12.2), 91 (73.5), 77 (20.4), 65 (6.1), 51
(17.3), 41 (30.6).
1
yield 31%. H NMR (CDCl3, 400 MHz) δ: 1.35, 1.45 (d,
3H), 3.9, 4 (q, 1H), 7–7.5 (10 H, m). 13C NMR (CDCl3,
100 MHz) δ: 18.5, 19, 51, 51.7, 140.2, 141, 141.8, 143.2,
210.3, 211.3. MS data m/e (relative intensity): 238 (M+, 4.4),
133 (10.3), 128 (2.9), 105 (100), 89 (1.5), 77 (17.6), 63
(1.5), 51 (5.9), 39 (2.9).
Synthesis of α,α,α′,α′-tetramethyl dibenzyl ketone (2c) (15)
Dibenzyl ketone (2a) (1 g, 0.005 mol) dissolved in 15 mL of
dry THF was added dropwise to the cooled solution (–10°C) of
potassium hydride (0.8 g, 0.02 mol) in 15 mL dry THF. The
resulting light yellow suspension was allowed to stir for
30 min and then 1.24 mL (0.02 mol) of methyl iodide was
added to the solution dropwise. The white suspension
formed was allowed to stir for 12 h at room temperature,
acidified with dil. HCl, and extracted with diethyl ether. The
ether layer was dried with anhyd MgSO4 and evaporated to
give a crude product mixture. Pure 2c was obtained by col-
umn chromatography, eluting on a silica gel column with
2% ethyl acetate in petroleum ether. The product was a
Step c
The acid, 15 (2 g) (0.012 mol), 1.76 g (0.012 mol) of 1-
naphthol, 2.5 g (0.012 mol) of DCC, and 0.15 g (0.0012 mol)
of DMAP were mixed together in 50 mL of dry methylene
chloride. The mixture was stirred under nitrogen for 12 h.
The resulting suspension was then filtered to remove the
white residue. The filtrate was washed with water and aque-
ous acetic acid (5%). The resulting methylene chloride solu-
tion was then dried with anhyd MgSO4 giving a crude
mixture of the product ester 1c. The mixture was purified by
column chromatography, eluting with petroleum ether, then
with 5% methylene chloride in petroleum ether, and finally
with 1% ethyl acetate in petroleum ether. Yield: 23%. The
resulting ester was a thick viscous colourless liquid. 1H
NMR (CDCl3, 400 MHz) δ: 1.89 (s, H), 7.18 (1H, d), 7.35–
7.55 (7H, m), 7.6 (2H, d), 7.73 (2H, d), 7.85 (2H, d). 13C
NMR (CDCl3, 100 MHz) δ: 26.4, 47, 118, 121, 125.5, 126
(Ar-C), 126.5, 127, 127.3, 128, 128.8, 134.8, 144, 146.8,
175.2. MS data m/e (relative intensity): 290 (M+, 14.7), 144
(75), 119 (100), 115 (41.2), 103 (10.3), 91 (51.5), 77 (13.2),
63 (10.3), 41 (32.4).
1
white solid; yield 81%. H NMR (CDCl3, 400 MHz) δ: 1.3
(s, 12 H), 7.18–7.3 (10H, m, Ar-H). 13C NMR (CDCl3,
100 MHz) δ: 28 (-CH3), 53.2 (-C(CH3)2), 125.8 (Ar-C),
126.5 (Ar-C), 128.5 (Ar-C), 144.5 (Ar-C), 213.4 (C=O). MS
data m/e (relative intensity): 266 (M+, 0.9), 147 (6.8), 119
(100), 103 (4.5), 91 (45.5), 77 (9), 51 (4.5), 41 (18).
Identification of photoproducts
Photoproducts from 1a and 1b have been identified previ-
ously by W. Gu and R.G. Weiss of Georgetown University.
The spectra of the photoproducts 3a–10a and 3b–10b agreed
with the reported data. Authentic samples provided by Gu
and Weiss were used to identify the GC peaks. Irradiation of
1c in hexane gave 1-naphthol, 9c, and 3c as the major prod-
ucts. An authentic sample obtained from Aldrich was used
to identify 1-naphthol. The spectral data for 9c and 3c are
1
shown below: For 9c: H NMR (CDCl3, 400 MHz) δ: 1.34
Dibenzyl ketone (2a) was used as obtained from Aldrich.
Dibenzyl ketones 2b and 2c were synthesized following the
procedure described below.
(s), 7.06–7.5 (10 H, m). 13C NMR (CDCl3, 100 MHz) δ:
25.2, 43.6, 125.5, 126.8, 128.6, 146.8. MS data m/e (relative
intensity): 238 (M+, 0.5), 128 (0.5), 119 (100), 91 (50), 77
(9.2), 65 (2.6), 51 (3.9), 41 (17). The other photoproducts
from ester 1c were assigned by comparing their GC reten-
tion times with similar products from the photolysis of esters
1a and 1b.
Synthesis of α,α′-dimethyl dibenzyl ketone (2b) (15)
Dibenzyl ketone (2a) (1 g, 0.005 dissolved in 15 mL of
dry THF was added dropwise to the cooled solution (–10°C)
of pottassium hydride (0.4 g, 0.01 mol) in 15 mL dry THF.
The resulting light yellow suspension was allowed to stir for
30 min and then 0.62 mL (0.01 mol) of methyl iodide was
added dropwise. The white suspension formed was allowed
to stir for 12 h, acidified with dil. HCl and extracted with di-
ethyl ether. The ether layer was dried with anhyd MgSO4
and evaporated to give a crude product mixture. Pure 2b was
Photoproducts upon photolysis of dibenzylketones namely
diphenylethane (11), cumene (17), and α-methyl styrene (18)
were compared to the original samples obtained from
Aldrich. Products 14 and 19 were prepared by preparative
photolysis of 2b and 2c, respectively, in hexane. Spectral
data for 14: 13C NMR (CDCl3, 400 MHz) δ: 18.5, 19, 46.5,
© 2003 NRC Canada

630
Can. J. Chem. Vol. 81, 2003
47.5, 126.2, 127.9, 128.5, 146. MSl data m/e (relative inten-
sity): 210 (4.4), 106 (8.9), 105 (100), 91 (3.3), 77 (15.6), 65
(2.2), 51 (5.6), 39 (3.3). For 19: 1H NMR (CDCl3,
400 MHz) δ: 1.34 (s), 7.06–7.5 (10 H, m). 13C NMR
(CDCl3, 100 MHz) δ: 25.2, 43.6, 125.5, 126.8, 128.6, 146.8.
MS data m/e (relative intensity): 238 (0.5), 128 (0.5), 119
(100), 91 (50), 77 (9.2), 65 (2.6), 51 (3.9), 41 (17).
134.5, 142, 149.8, 200. MS data m/e (relative intensity): 238
(1.9), 165 (1.3), 133 (100), 119 (3.7), 105 (22.2), 91 (5.6),
77 (18.2), 51 (5.6), 39 (3.7). For 16: light orange colored
1
solid. H NMR (CDCl3, 400 MHz) δ: 1.25 (t, 3H), 1.6 (d,
3H), 2.7 (q, 2H), 4.7 (q, 1H), 7.2–7.4 (7H, m), 7.95 (2H, d).
13C NMR (CDCl3, 100 MHz) δ: 16, 20, 29, 48, 127, 128.2,
128.5, 129, 130.2, 134.5, 142, 149.8, 200. MS data m/e (rel-
ative intensity): 238 (1.6), 165 (1.6), 133 (100), 105 (21.8),
89 (4.7), 77 (15.6), 51 (6.3), 39 (3.1).
Synthesis of 2-methyl phenyl benzyl ketone (12) and 4-
methyl phenyl benzyl ketone (13)
Photolysis
In a small round-bottomed flask, 0.5 g (0.003 mol) of the
corresponding acid chloride (2-methyl benzoyl chloride or 4-
methyl benzoyl chloride) was added to 15 mL dry THF. This
mixture was kept stirring at –10°C under a nitrogen atmo-
sphere. A benzyl magnesium chloride (0.0025 mol) suspen-
sion in THF was added dropwise to the above solution. The
light yellow solution was allowed to stir for 12 h at room
temperature (25°C) followed by addition of satd NH4Cl so-
lution in water and extraction with three 20 mL portions of
diethyl ether. The ether layers were combined, dried with
anhyd MgSO4, and evaporated to give a crude product mix-
ture. Pure 12 or 13 was then obtained by column chromatog-
raphy, eluting on a silica gel column with 2% ethyl acetate
Known amounts of substrate (3 mg) and activated (450°C,
air oven) zeolite (300 mg) were stirred together in hexane
for at least 6 h, filtered, and washed with excess hexane. GC
analysis of the hexane layer made sure that all substrate was
loaded into the zeolite. The loading level was kept about 1
molecule per 12 supercages for the esters 1a–1c and 1 mole-
cule in 7 supercages for the ketones 2a–2c. The “loaded” ze-
olite sample was transferred to fresh hexane solvent (10 mL)
and photolyzed (Pyrex tubes, 450 W medium pressure mer-
cury lamp) as a hexane slurry. A continuous nitrogen purge
was maintained through the slurry during the irradiation of
ketones 2a–2c. For the esters 1a–1c, 30% conversion was
achieved upon 2 h of photolysis in zeolites LiY, NaY, and
KY. The reaction proceeded very slowly in zeolites RbY and
CsY and was almost completely absent in TlY. A 30% con-
version was achieved after 2 h of irradiation for ketone 2a,
1 h of irradiation for ketone 2b, and 30 min of irradiation for
ketone 2c. GC analysis of the hexane layer of the irradiated
slurry confirmed that all the products and the reactant re-
mained within the zeolite. After irradiation, the samples
were extracted with a 5% water in THF solution and ana-
lyzed on a GC (Hewlett-Packard 5890 (FID detector), col-
umn SPB-5 (30 m, 0.32 mm i.d., 1.0 mm film thickness)).
The conditions used for the GC run for esters 1a–1c were:
initial temperature (100°C), initial time (1 min), rate (5°C
per min), final temperature (250°C), final time (10 min). The
GC conditions used for ketone 2a were: initial temperature
(100°C), initial time (1 min), rate (2°C per min), final tem-
perature (160°C), final time (10 min), rate A (10°C per min),
final temperature A (250°C), final time A (1 min). GC con-
ditions used for ketones 2b and 2c were: initial temperature
(100°C), initial time (1 min), rate (5°C per min), final tem-
perature (250°C), final time (10 min). In general, mass bal-
ance as followed by using a calibration compound on a GC
was found to be ~85%. The calibration compound used for
the esters 1a–1c was 1-naphthyl acetate and for the ketones
2a–2c was phenyl phenyl acetate.
1
in petroleum ether. For 12: H NMR (CDCl3, 400 MHz) δ:
2.45 (s), 4.23 (s), 7.2–7.4 (8H, m), 7.74 (1H, d). 13C NMR
(CDCl3, 100 MHz) δ: 21.3, 48.5, 125.5, 126.8, 128.5, 130,
132, 133, 134.5, 136.6, 137.6, 138.8, 201.5. MS data m/e
(relative intensity): 210 (2.7), 165 (1.1), 119 (100), 91
(69.4), 65 (48.6), 51 (8.3), 39 (25). For 13: 1H NMR
(CDCl3, 400 MHz) δ: 2.42 (s), 4.28 (s), 7.22–7.38 (7H, m),
7.94 (2H, d). 13C NMR (CDCl3, 100 MHz) δ: 21.7, 45.4,
127.6, 129.4, 129.6, 130.1, 130.2, 135, 135.6, 144.9, 198.5.
MS data m/e (relative intensity): 210 (2.7), 165 (1.4), 119
(100), 91 (48.6), 65 (37.8), 51 (6.8), 39 (17.6).
Synthesis of 2-ethyl phenyl α-methyl benzyl ketone (15)
and 4-ethyl phenyl α-methyl benzyl ketone (16)
Ethyl benzene (10 g) (0.09 mol) and 14.5 g (0.094 mol) of
phenyl acetyl chloride were placed in a small round-
bottomed flask. Into a conical flask, 12.5 g (0.09 mol) of
AlCl3 was weighed out and this solid was added with fre-
quent shaking to the contents of the round-bottomed flask.
The mixture turned dark green with evolution of HCl gas
during the addition. After addition was complete, a reflux
condenser was attached and the round-bottomed flask was
heated on a water bath for 3 h. The contents of the flask
were poured into a mixture of crushed ice and concd HCl
with continuous stirring. The resultant water layer was ex-
tracted with five 25 mL portions of diethyl ether. The ether
layers were combined, dried with anhyd MgSO4, and evapo-
rated to give a crude product mixture. The non-methylated
precursors to 15 and 16 could be isolated by column chro-
matography, eluting on a silica gel column with a 5% ethyl
acetate in petroleum ether mixture. They were then methyl-
ated using the procedure used for the ketones 2b and 2c de-
scribed earlier. Purification by silica gel chromatography,
eluting with 2% ethyl acetate in petroleum ether afforded
Absorption spectra
Absorption spectra were recorded using a Shimadzu 2101
PC UV–vis spectrophotometer with a diffuse reflectance ac-
cessory attachment. The dried zeolite samples were packed
into a 2 mm quartz cuvette (in a dry box) and sealed with
teflon tape. Conversion of the reflectance spectra to the ab-
sorption spectra was carried out using a Kubelka–Munk pro-
gram supplied with the instrument.
1
pure 15 and 16. Spectral data for 15: white colored solid. H
NMR (CDCl3, 400 MHz) δ: 1.25 (t), 1.6 (d), 2.7 (q), 4.7 (q),
7.23–7.45 (8H, m), 7.76 (1H, d). 13C NMR: (CDCl3,
100 MHz) δ: 16, 20, 29, 48, 127, 128.2, 128.5, 129, 130.2,
Emission spectra
Emission spectra were recorded for the three naphthyl es-
ters 1a–1c. The spectra were recorded on a Edinburgh FS-
© 2003 NRC Canada

Warrier et al.
631
3. K. Pitchumani, M. Warrier, C. Cui, R.G. Weiss, and V.
Ramamurthy. Tetrahedron Lett. 6251 (1996).
4. (a) S. Vasenkov and H. Frei. J. Am. Chem. Soc. 120, 4031
(1998); (b) S. Vasenkov and H. Frei. J. Phys. Chem. A, 104,
4327 (2000).
5. (a) W. Gu, M. Warrier, V. Ramamurthy, and R.G. Weiss. J.
Am. Chem. Soc. 121, 9467 (1999); (b) M. Warrier, N.J. Turro,
and V. Ramamurthy. Tetrahedron Lett. 41, 7163 (2000).
6. (a) W. Gu and R.G. Weiss. Tetrahedron, 56, 6913 (2000);
(b) W. Gu and R.G. Weiss. J. Photochem. Photobiol C, 2, 117
(2001).
900 CDT spectrofluorimeter. Solid zeolite samples were
taken in quartz ESR tubes placed in a quartz cylindrical
Dewar, and the emissions collected at right angles were re-
corded. The ester samples were excited at 290 nm. The fluo-
rescence λmax was 340 nm and the phosphorescence λmax
was 525 nm. Lifetime measurements were carried out on an
Edinburgh FL-900 CDT single photon counter using a hy-
drogen filled ns flashlamp (40 KHz) as the light source. The
samples were excited at 290 nm, and the emitted photons
were collected at the peak maximum of the fluorescence
bands of the adsorbed substrates. The lamp profile was col-
lected with ludox as the scattering medium. The observed
decays were fitted using the distribution or the exponential
analysis program supplied with the instrument. The lifetimes
were fitted to an exponential fit. The suitability of the fit
was ascertained by a χ2 value close to unity. The decays
within the zeolites were usually biexponential.
7. V. Ramamurthy, J.V. Caspar, D.F. Eaton, E.W. Kuo, and D.R.
Corbin. J. Am. Chem. Soc. 114, 3882 (1992).
8. (a) P.S. Engel. J. Am. Chem. Soc. 92, 6074 (1970); (b) W.K.
Robbins and R.H. Eastman. J. Am. Chem. Soc. 92, 6076
(1970); (c) N.J. Turro and W.R. Cherry. J. Am. Chem. Soc.
100, 7431 (1978); (d) N.J. Turro and B. Kraeutler. Acc. Chem.
Res. 13, 369 (1980).
9. Such rearrangement products were first reported by Turro, de
Mayo, and co-workers: (a) N.J. Turro, D.R. Anderson, and B.
Kraeutler. Tetrahedron Lett. 21, 3 (1980), (b) B. Frederick, L.J.
Johnston, P. de Mayo, and S.K. Wong. Can. J. Chem. 62, 403
(1984).
Computations
Geometry optimizations were carried out using the re-
stricted hybrid Hartree–Fock density functional theory
(RB3LYP) with Becke three parameter exchange functional
in conjunction with correlation functional by Lee, Yang, and
Parr (LYP). The 6-31G(d) basis set was used for C, H, O, Li.
Stationary points have been characterized as true minima by
frequency calculations. All of the calculations were carried
out using the Gaussian 98 series of programs (10).
10. M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery,
Jr., R.E. Stratmann, J.C. Burant, S. Dapprich, J.M. Millam,
A.D. Daniels, K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi,
V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C.
Adamo, S. Clifford, J. Ochterski, G.A. Petersson, P.Y. Ayala,
Q. Cui, K. Morokuma, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz, B.B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham,
C.Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P.M.W. Gill, B. Johnson,W. Chen, M.W. Wong, J.L. Andres,
C. Gonzalez, M. Head-Gordon, E.S. Replogle, and J.A. Pople.
Gaussian 98, Revision A.9. Gaussian, Inc., Pittsburgh, Penn.
1998.
Acknowledgement
It is a pleasure to thank Professors R.G. Weiss and N.J.
Turro for very useful and stimulating discussions. We also
thank W. Gu and R.G. Weiss for providing authentic sam-
ples at the beginning and at various stages of the project.
The authors thank the National Science Foundation (CHE-
9904187 and CHE-0212042) for support of the research
summarized here.
11. (a) L. Lunazzi, K.U. Ingold, and J.C. Scaiano. J. Phys. Chem.
87, 529 (1983); (b) N.J. Turro, I.R. Gould, and B.H. Baretz. J.
Phys. Chem. 87, 531 (1983).
12. (a) I.R. Gould, N.J. Turro, and M.B. Zimmt. Magnetic field
and magnetic isotope effects on the product of organic reac-
tions. In Advances in physical organic chemistry. Vol. 20.
Edited by V. Gold and D. Bethell. Academic Press, H.B.
Jovanovich, London. 1984. pp. 1; (b) N.J. Turro. Pure Appl.
Chem. 67, 199 (1995); (c) N.J. Turro. Acc. Chem. Res. 33,
637 (2000); (d) N.J. Turro. Pure Appl. Chem. 58, 1219 (1986).
13. N.J. Turro and Z. Zhang. Tetrahedron Lett. 28, 5637 (1987).
14. (a) E. Campaigne and W.M. Leuser. In Organic syntheses Coll.
IV. Edited by N. Rabjohn. John Wiley and Sons, New York.
pp. 919; (b) A. Hassner and V. Alexanian. Tetrahedron Lett.
4475 (1978).
References
1. (a) D. Bellus. Adv. Photochem. 8, 109 (1971); (b) D. Bellus
and P. Hrdlovic. Chem. Rev. 67, 599 (1967); (c) V.I. Stenburg.
Org. Photochem. 1, 127 (1967); (d) M.A. Miranda. In Hand-
book of organic photochemistry and photobiology. Edited by
W.M. Horspool and P.S. Song. CRC Press, Boca Raton. 1995.
pp. 570.
2. (a) K. Pitchumani, M. Warrier, and V. Ramamurthy. J. Am.
Chem. Soc. 118, 9428 (1996); (b) K. Pitchumani, M. Warrier,
and V. Ramamurthy. Res. Chem. Intermed. 25, 623 (1999);
(c) W. Gu, M. Warrier, B. Schoon, V. Ramamurthy, and R.G.
Weiss. Langmuir, 16, 6977 (2000).
15. A.A. Millard and M.W. Rathke. J. Org. Chem. 43, 1834
(1978).
© 2003 NRC Canada

This article has been cited by:
1. Cheng Yang, Yoshihisa InoueSupramolecular Photochirogenesis 115-153. [CrossRef]
2. Neil G. Berry, Lisa Iddon, Mazhar Iqbal, Xiaoli Meng, Prabha Jayapal, Caroline H. Johnson, Jeremy K. Nicholson, John C.
Lindon, John R. Harding, Ian D. Wilson, Andrew V. Stachulski. 2009. Synthesis, transacylation kinetics and computational
chemistry of a set of arylacetic acid 1β-O-acyl glucuronides. Organic & Biomolecular Chemistry 7:12, 2525. [CrossRef]
3. Selvanathan Arumugam. 2008. Nafion as an efficient reaction medium for diastereoselective photochemical reactions. Tetrahedron
Letters 49:15, 2461-2465. [CrossRef]
4. Arun Kumar Sundaresan, V. Ramamurthy. 2008. Consequences of controlling free space within a reaction cavity with a remote
alkyl group: photochemistry of para-alkyl dibenzyl ketones within an organic capsule in water. Photochemical & Photobiological
Sciences 7:12, 1555. [CrossRef]
5. Lakshmi S. Kaanumalle, Corinne L. D. Gibb, Bruce C. Gibb, V. Ramamurthy. 2007. Photo-Fries reaction in water made selective
with a capsule. Organic & Biomolecular Chemistry 5:2, 236. [CrossRef]
6. Selvanathan Arumugam, Lakshmi S. Kaanumalle, V. Ramamurthy. 2006. Alkali Ion Exchanged Nafion as a Confining Medium
for Photochemical Reactions†. Photochemistry and Photobiology 82:1, 139. [CrossRef]