Value of zeolites in asymmetric induction during photocyclization of pyridones, cyclohexadienones and naphthalenones

Article abstract of DOI:10.1039/b702572f

Two strategies, namely chiral inductor and chiral auxiliary approaches, have been examined within zeolites with the aim of achieving asymmetric induction during the photocyclization of cyclohexadienone, naphthalenone and pyridone derivatives. Within zeoli

Full text of DOI:10.1039/b702572f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Value of zeolites in asymmetric induction during photocyclization of
pyridones, cyclohexadienones and naphthalenones†
Karthikeyan Sivasubramanian, Lakshmi S. Kaanumalle, Sundararajan Uppili and V. Ramamurthy*
Received 19th February 2007, Accepted 26th March 2007
First published as an Advance Article on the web 23rd April 2007
DOI: 10.1039/b702572f
Two strategies, namely chiral inductor and chiral auxiliary approaches, have been examined within
zeolites with the aim of achieving asymmetric induction during the photocyclization of
cyclohexadienone, naphthalenone and pyridone derivatives. Within zeolites, enantioselectivity as high
as 55% and diastereoselectivity as high as 88% have been obtained. The observed stereoselectivities are
significant given the fact that these reactions gave very little stereoselectivities in isotropic solution
media. The results obtained on the photocyclization of dienones, naphthalenones and N-alkyl
pyridones within zeolites compliment our earlier investigations on the photocyclization of tropolone
derivatives, the geometric isomerization of 1,2-diphenylcyclopropanes and 2,3-diphenyl-1-benzoyl
cyclopropanes, and the Norrish type II reaction of a-oxoamides, phenyl adamantyl ketones, phenyl
norbornyl ketones and phenyl cyclohexyl ketones. With the help of these examples, we have established
the importance of zeolite and its charge compensating cations in effecting asymmetric induction in
photochemical reactions.
Introduction
The chiral sources that have been employed to achieve stereose-
lectivity in photochemical reactions include circularly polarized
light, chiral sensitizers, chiral solvents, chiral substituents, chiral
host–guest assemblies and chiral crystalline environments.1–8 In
our laboratory, an approach based on solid host–guest assemblies
using zeolite as the solid host to induce asymmetric induction
during photochemical transformations has been developed.9 We
have previously demonstrated that the confined cavities of a
zeolite and charge compensating cations present within them could
be used to induce enantio- and diastereoselectivity in products
of photoreactions such as geometric isomerization, hydrogen
abstraction and cyclization.10 To establish generality of the zeolite-
based approach, we have investigated excited state reactions of 3,3-
dimethyl-4-oxo-3,4-dihydronaphthalene derivatives (1a, 1b, 2a–2e,
2g, 2i, 2j, 2m–2q, 2s, 2t), 2,4 cyclohexadienone derivatives (1c, 1d,
4a, 4e–4h, 4j, 4n, 4o, 4q, 4r, 4s) and N-alkyl pyridone derivatives
(1e–1j, 6e, 6g, 6i–6q) (Schemes 1–4) within MY zeolites where M is
an alkali ion. Results obtained with these examples are presented
in this report.
Scheme 1 Naphthalenone, dienone and pyridone derivatives investigated
by the chiral inductor approach.
a chiral inductor within an achiral zeolite. Photoproducts were
analyzed by either GC or HPLC using chiral columns. The
first of the two enantiomeric/diasteromeric peaks on GC/HPLC
traces was assigned as A and the second peak as B. In the
present manuscript, we disclose the results that we believe would
convince the readers that the zeolite-based strategy is a viable
(although not general) approach to achieve asymmetric induction
in photoreactions.
Results
In order to establish the value of zeolites during asymmetric induc-
tion in photoreactions, we have pursued two approaches, namely
chiral inductor approach—i.e. inclusion of an achiral reactant
within a chirally-modified zeolite and chiral auxiliary approach—
i.e. inclusion of a reactant molecule covalently appended with
Chiral inductor approach
Structures of four achiral dienones and six pyridones examined
in the context of chiral inductor approach are provided in
Scheme 1. It has been established earlier that naphthalenone11
and 2,4-cyclohexadienone12 derivatives upon irradiation in highly
polar solvents such as trifluoroethanol undergo oxa di-p-methane
rearrangement from a pp* excited triplet state (Scheme 2). We
have previously shown that the lowest excited state of a,b-enones
Department of Chemistry, University of Miami, Coral Gables, Fl 33124,
USA
† Electronic supplementary information (ESI) available: Experimental de-
tails including synthesis, characterization of reactants and photoproducts
and analysis conditions for photoproducts. See DOI: 10.1039/b702572f
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1569–1576 | 1569
©

View Article Online
Scheme 3 Optically pure chiral inductors used.
chiral b-lactams. The zeolite used for the current study is the
commercially available achiral Y zeolite. The chiral environment
was created by inclusion of chiral inductor molecules into the
supercages of MY zeolite. Experimental procedures including
adsorption, irradiation, extraction and analysis are described in
the experimental section.
In the chiral inductor approach, in order to maximize the
chances of the reactant molecule being closer to the chiral
inductor molecule, we used a higher ratio of chiral inductor to
the reactant (10 : 1). Structures of the chiral inductors used in this
study are shown in Scheme 3. Results obtained in the presence
of these chiral inductors are presented in Table 1. Irradiation of
various dienone and pyridone derivatives within chirally-modified
NaY zeolites, resulted in moderate enantiomeric excess (ee) in the
photoproducts. The importance of chiral inductors and cations
present in a zeolite during asymmetric induction was realized from
the following observations: (a) irradiation of dienones (1a–1d) and
pyridones (1e–1j) in MY zeolite without a chiral inductor gave
Scheme 2 Photochemistry of chiral auxiliary appended naphthalenone,
dienone and pyridone derivatives. X in structures 2 and 4 corresponds
to COR* and in structure 6 corresponds to CH2COR*. The structures
of chiral auxiliaries (R* in X) used during chiral auxiliary approach are
provided in Scheme 4.
is pp* within alkali ion exchanged MY zeolites.13 Consistent
with this, oxa di-p-methane rearrangement products alone were
obtained upon irradiation of 1a–d included within a chirally-
modified zeolite. In agreement with the known photobehavior
of pyridones (Scheme 2),14–19 excitation of N-alkyl pyridone
derivatives (1e–1j) adsorbed within chirally-modified zeolites gave
Scheme 4 The structures of chiral auxiliaries used. See Scheme 2 to track the structures to which these groups would be attached.
1570 | Org. Biomol. Chem., 2007, 5, 1569–1576 This journal is The Royal Society of Chemistry 2007
©

View Article Online
Table 1 Enantioselectivity in the photoproducts obtained within MY
(2) and cyclohexadienone derivatives (4), upon irradiation in
solution gave products of oxa di-p-methane rearrangement in less
than 13% diastereomeric excess (de). The same substrates when
irradiated within NaY zeolite gave products in moderate to high
diastereoselectivity (30–81%) (Tables 2 and 3). Similar enhance-
ment in de (∼88%) was observed during the photocyclization of
pyridone derivatives (6) within MY zeolites (Table 4).
Like in the chiral inductor approach, the diastereoselectivity
obtained within MY zeolites varied with the nature of the charge-
compensating cations. For example, the naphthalenone derivative
2g gave the photoproduct (3g) with de of 42% (B) in LiY; 48% (B)
in NaY; 81% (B) in KY and 46% (B) in RbY. Yet another notable
feature was that the nature of the cation not only controlled the
extent of diastereoeselectivity but in some cases it also controlled
the diastereomer being enhanced. For pyridone derivatives 6l and
6m, LiY and NaY gave one diastereomer in excess whereas in KY,
RbY and CsY gave the other diastereomer in excess (6l: LiY 48%
(B), NaY 25% (B), KY 76% (A), RbY 88% (A) and CsY 82% (A);
6m: LiY 32% (B), NaY 50% (B), KY 36% (A), RbY 25% (A) and
CsY 29% (A)). The dienone derivatives (4f, 4h, 4o and 4q) also
behaved in a similar fashion (Tables 3 and 4). These derivatives
zeolitesa,b,c
Substrate
Zeolite
Chiral inductor
% Enantiomeric excess
1a
1a
1a
1a
1b
1b
1b
1b
1c
NaY
NaY
NaY
NaY
NaY
NaY
NaY
NaY
NaY
(−)-Ephedrine
13A
10B
17A
15B
18B
15A
17B
14A
25A
(+)-Ephedrine
(+)-Pseudoephedrine
(−)-Pseudoephedrine
(−)-Ephedrine
(+)-Ephedrine
(+)-Pseudoephedrine
(−)-Pseudoephedrine
(1S, E)-(−)-Camphor
quinine-3-oxime
(−)-Ephedrine
1d
1d
1d
1d
1e
1f
1g
1h
1i
NaY
NaY
NaY
NaY
NaY
NaY
NaY
NaY
KY
30B
28A
20B
26A
3B
6B
6B
22B
53B
40B
50B
(+)-Ephedrine
(−)-Pseudoephedrine
(+)-Pseudoephedrine
(−)-Norephedrine
(−)-Norephedrine
(−)-Norephedrine
(−)-Norephedrine
(−)-Ephedrine
1i
1j
RbY
NaY
(−)-Ephedrine
(−)-Norephedrine
a A refers to first peak eluting from GC/HPLC column. b The optical
antipode of the chiral inductor gave the other enantiomer in excess.
c The loading, irradiation and extraction procedures are provided in the
experimental section.
Table 2 Diastereoselectivity in the photoproducts obtained within MY
zeolites for naphthalenone derivatives (2)a,b
Substrate
CH3CN
NaY
KY
MY
2a
2b
2c
2d
2e
2g
2i
0
9B
3B
13B
6
10A
46A
57B
48B
45B
48B
35B
13B
21A
12A
0
8B
60A
34B
35B
23B
81B
58B
13B
43A
58A
45A
41B
30B
12A
57B
31A (LiY)
48A (RbY)
24B (CsY)
34B (RbY)
35B (CsY)
46B (RbY)
45B (CsY)
25B (LiY)
10A (RbY)
20A (RbY)
10A (RbY)
5B (RbY)
7B (RbY)
15A (RbY)
30B (LiY)
racemic products; (b) irradiation of pyridone 1j in NaY zeolite
with (Si : Al = 40), a zeolite with lesser number of cations gave only
2% enantioselectivity, while the same substrate in NaY (Si : Al =
2.4) with a higher number of cations gave 50% enantioselectivity.
Also dienone 1c while in NaY gave 30% ee, within ephedrine-
modified Y-sil (Si : Al > 285), a zeolite with no cation, gave racemic
products; (c) the enantioselectivity was also dependent on the
nature of the alkali ion present in a zeolite. For example, in the case
of 1c, irradiation in (−)-ephedrine-modified alkali ion exchanged
Y zeolites led to an ee of 0% in LiY, 32% in NaY, 26% in KY, 5%
in RbY, 0% in CsY and for 1i, 23% in LiY, 5% in NaY, 53% in KY,
40% in RbY, and 12% in CsY; (d) consistent with the suggestion
that interaction between the cation, chiral inductor and reactant
molecules plays a crucial role, inclusion of water molecules that
would coordinate to the cation, into (−)-ephedrine-modified
NaY zeolite reduced the enantioselectivity to 0% from 25% in
the case of 1d; (e) variation of the irradiation temperature had a
distinct effect on the enantioselectivity obtained from the product
formed from cyclohexadienone derivative (1d) (chiral inductor =
(−)-ephedrine). Decreasing the temperature from 25 ◦C to −55 ◦C
increased the enantioselectivity from 30 (B) to 49 (B).
0
3B
8B
9A
0
2A
8A
8A
9A
6B
2j
2m
2n
2o
2p
2q
2s
2t
5B
11B
48A
19B
a A refers to first peak eluting from GC/HPLC column. b The loading,
irradiation and extraction procedures are provided in the experimental
section.
Table 3 Diastereoselectivity in the photoproducts obtained within MY
zeolites for cyclohexadienone derivatives (4)a,b
Substrate
CH3CN
NaY
RbY
MY
4a
4e
4f
4g
4h
4j
4n
4o
4q
4r
4s
2B
0
0
5
0
5B
2
12B
38B
25B
39B
13B
8A
22B (CsY)
31B (CsY)
40A (LiY)
17A (LiY)
34A (LiY)
39A (LiY)
32A (LiY)
25A (LiY)
35B (KY)
60A (KY)
25B (KY)
73A
59A
41A
20A
53A
22A
25A
27A
59B
Chiral auxiliary approach
We admit that the enantioselectivity obtained by the chiral induc-
tor approach with the substrates (1a–1j) was moderate at best.
One way to enhance the chiral environment is to covalently link
the chiral inductor to the substrate of interest so that the achiral
reactant and chiral inductor components of the same molecule
would stay close to each other. We call this chiral auxiliary
approach. The chiral auxiliaries used were optically pure chiral
alcohols, amines, amino alcohols and amino acid methyl ester
derivatives (Scheme 4). Such chiral auxiliary linked naphthalenone
0
4A
2A
4B
29A
11A
7B
40B
17B
43A
18B
a A refers to first peak eluting from GC/HPLC column. b The loading,
irradiation and extraction procedures are provided in the experimental
section.
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1569–1576 | 1571
©

View Article Online
Table 4 Diastereoselectivity in the photoproducts obtained for pyridone
derivatives (6) within MY zeolitesa,b
and 88% in RbY zeolite, whereas substrate 6i which has an alkyl
chiral auxiliary (cyclohexylethylamine) gave the b-lactam product
in 24% de in LiY zeolite. The above comparison revealed that the
functional group present in a chiral auxiliary has a role to play in
the chiral induction process.
Also to ensure that the observed enhancement in diastereoselec-
tivity within zeolite is not due to any experimental artifacts, the 1 :
1 diastereomeric mixture of photoproducts isolated from solution
irradiation of the naphthalenone derivative 2g was included within
NaY by stirring in hexane for 12 hours. The photoproducts
were then extracted with acetonitrile and analyzed. The analysis
indicated the ratio of photoproducts to be 1 : 1, confirming that
no selective inclusion or extraction of photoproducts took place.
Substrate
CH3CN
NaY
KY
MY
6e
6g
6i
1B
2B
2B
3B
3B
3B
44B
8A
5A
84B
11B
25B
75B
42A
1A
74B
67B
76A
38B (RbY)
32A (RbY)
24A (LiY)
64B (CsY)
60B (RbY)
88A (RbY)
82A (CsY)
32B (LiY)
53A (RbY)
28B (LiY)
33B (LiY)
34A (RbY)
6j
6k
6l
6m
6n
6o
6p
6q
1A
2A
1A
2B
3A
50B
14B
33B
4A
36A
16B
10B
8A
21A
78A
a A refers to first peak eluting from GC/HPLC column. b The loading,
irradiation and extraction procedures are provided in the experimental
section.
Discussion
We have used two methodologies, namely ‘chiral inductor’ and
‘chiral auxiliary’, to conduct enantio- and diastereoselective
photocyclization of 3-dimethyl-4-oxo-3,4-dihydro-naphthalene-1-
carboxylic acid derivatives (1a, 1b, 2a–2e, 2g, 2i, 2j, 2m–2q, 2s,
2t), 2,4-cyclohexadienone derivatives (1c–1e, 4a, 4e–4h, 4j, 4n,
4o, 4q, 4r, 4s) and N-alkyl pyridone derivatives (1e–1j, 6e, 6g,
6i–6q) (Schemes 1–4). All three systems gave similar enantio-
and diastereoselectivities by chiral inductor and chiral auxiliary
approaches, and the extent of selectivity was dependent on the
nature and number of cations within a zeolite, presence of
water and the structure of chiral inductor/chiral auxiliary. These
observations suggest that the model used to understand the asym-
metric induction within zeolites should take into consideration
the existence of interaction between the reaction site, the chiral
component and the cation present within a zeolite.
Although the products in the three systems are different,
the primary step that determines the chirality involves similar
motions (Scheme 5). Naphthalenones and cyclohexadienones in
their excited triplet state undergo 1,3-bonding through disrotatory
motion of the p-orbitals inward or outward to yield the triplet
diradical intermediate that rearranges to the final product. Pyri-
dones yield the b-lactams from their excited singlet state via 1,4-
bonding through disrotatory motion of the p-orbitals inward or
outward by a concerted 4e-cyclization process. Once the molecule
is adsorbed on the zeolite surface, the outward motion that would
push the newly forming cyclopropane ring (naphthalenones and
cyclohexadienones) or b-lactam ring towards the surface would
not be sterically favored, and therefore once the molecule is
adsorbed on a surface the cyclization is expected to proceed only
in LiY and NaY gave one diastereomer in excess, whereas in KY,
RbY and CsY gave the other diastereomer in excess (Table 3).
Water, as observed in the case of the chiral inductor approach,
reduced the de. The diastereoselectivities obtained by irradiation
of naphthalenone derivatives (2b–2e, 2i and 2g) and pyridone
derivative (6l) under dry and wet conditions are provided in
Table 5. These results clearly illustrate the role of free cations
in controlling the extent of diastereoselectivity.
Once again, the importance of the cation became evident when
diastereoselectivity was monitored for substrate (6l) in zeolites
having Si : Al ratio of 6 (cations per unit cell = 22), 15 (cations
per unit cell = 9) and 40 (cations per unit cell = 3.4). The
diastereoselectivity observed for substrate 6l within KY (Si : Al
= 2.4) was 76%. This value decreased to 6% when KY (Si : Al
= 40) was used. Zeolite KY with Si : Al ratio of 6 and 15 gave
diastereoselectivity of 20% and 10%, respectively. When pyridone
derivative 6l adsorbed onto silica that does not have any cations
was irradiated the de was 3%.
Not all chiral auxiliaries were equally effective. Perusal of
Tables 2 to 4 suggests that chiral auxiliaries having an aromatic
group give better diastereoselectivity than the ones with an
alkyl group. For example, irradiation of substrate 2g gave the
photoproduct in 81% de in KY zeolite, whereas irradiation of 2i
gave the photoproduct in only 58% de in KY zeolite. Dienone and
pyridone derivatives also exhibited similar behavior. For example,
photoirradiation of pyridone derivatives 6g, 6k and 6l gave the b-
lactam photoproduct in 42% de in KY zeolite, 67% in KY zeolite
Table 5 Effect of co-adsorbent (water) on diastereoselectivity in the photoproductsa,b
Zeolite
Substrate
% Diastereoselectivity (dry)
% Diastereoselectivity (wet)
NaY
NaY
NaY
NaY
NaY
KY
2b
2c
2d
2e
2i
48A
57B
48B
45B
35B
58B
48B
76A
16A
11B
25B
21B
20B
9B
2i
NaY
KY
2g
6l
15B
1A
a A refers to first peak eluting from GC/HPLC column. b The loading, irradiation and extraction procedures are provided in the experimental section.
1572 | Org. Biomol. Chem., 2007, 5, 1569–1576
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
zeolite gave 53% excess of B enantiomer. On using the optical
antipode of chiral inductor (+)-ephedrine, 50% excess of A
enantiomer was obtained suggesting that the system is well
behaved. Though appreciable enantioselectivity was obtained with
pyridone derivatives (1h, 1i and 1j), their alkyl counter parts,
namely N-methyl pyridone (1e), N-ethyl pyridone (1f) and N-
propyl pyridone (1g), yielded very low enantioselectivity (3–6%)
(Table 1). Furthermore, in the aryl-substituted 1h, 1i and 1j the one
with ethyl and propyl chain gave higher enantioselectivity than the
one with single methylene unit (Table 1, ee in 1j ∼ 1i > 1h).
To probe the role of the phenyl group and the chain length in
controlling the extent of stereoselectivity, we performed computa-
tions on substrates 1f, 1h, 1i and 1j at RB3LYP level using 6–31G*
basis set with Gaussian 98 A.11 suite of programs.20 Geometry
optimized structures of the substrates were allowed to interact
with Li+ cation and the cation-bound complexes were reoptimized.
During the optimization, the metal ion was free to move to find
the most stable position. The optimized computed structures of
Li+ complexes with 1f, 1h, 1i and 1j are provided in Fig. 1. The
computed structures indicate that the primary interaction of cation
(Li+) with the alkyl-substituted N-ethyl pyridone (1f) is with the
carbonyl group via cation-dipolar interaction (binding affinity
= 66.31 kcal mol−1). However, in the cases of aryl-substituted
pyridones (1h–1j), the cation (Li+) interacted simultaneously with
both the phenyl group (cation–p interaction) and the carbonyl
group (cation-dipolar type interaction) (binding affinity = 76.66
kcal mol−1 for 1h, 85.83 kcal mol−1 for 1i and 85.11 kcal mol−1
for 1j). In terms of the origin of the difference in the extent of
enantioselectivity between the two systems, a comparison of the
structures of 1f–Li+ and 1h–Li+ is revealing. The best ee for 1f
is 6%, while for 1h is 22%. In 1f, the cation interacts only with
carbonyl chromophore leaving the ethyl substituent free. On the
other hand, in 1h the cation ties up the molecule by interacting with
the carbonyl and the phenyl group. Clearly when cation rigidifies
the reactant molecule as a whole, the chiral inductor is able to force
it to prefer one mode of adsorption at the expense of the other
(Fig. 1, Scheme 6). As indicated by the results of 1i and 1j, stronger
Scheme 5 Photocyclization of cyclohexadienone and pyridone.
by disrotation of the p-orbitals inward (Scheme 6). However,
even this one-way rotation of the p-orbitals would yield equal
amounts of the optical isomers as the reactant molecules would
show no preference for adsorption from the two pro-chiral faces
(Scheme 6 without chiral inductor). Interaction between the chiral
inductor/chiral auxiliary and the reactant part (naphthalenone,
cyclohexadienone and pyridone) could bias the adsorption process
and this we believe is the origin of asymmetric induction within
zeolites. We show below with the help of computational results that
interaction between the chiral and reactant components mediated
by cation most likely controls the asymmetric induction process.
We recognize the fact that our analysis is an ‘after the fact’ event
and the cartoon model that we envision to understand the results
lacks predictive power.
Scheme 6 Adsorption of cyclohexadienone and N-alkyl or aryl pyridone
on a surface.
Since all three systems behaved in a similar manner, to
avoid repetitive discussion, we analyze the results obtained from
photocyclization of N-alkyl pyridone derivatives within zeolites.
Irradiation of substrate 1i within (−)-ephedrine-modified KY
Fig. 1 Interaction of Li+ with N-ethyl pyridone (1f), N-benzyl pyridone
(1h), N-ethyl phenyl pyridone (1i) and N-propyl phenyl pyridone (1j). The
structures have been computed at RB3LYP/631G* level using Gaussian
98. B. A. refers to binding affinity.20
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1569–1576 | 1573
©

View Article Online
binding leads to higher ee. In both cases, binding affinity is ∼85
kcal mol−1, and the highest ee in both cases is ∼50% (Table 1).
We had shown previously through solid-state NMR (CP–MAS)
and computational results that Li+ bound cyclohexyl phenyl
ketone binds to chiral inductors such as norephedrine in a very
specific manner.21 We believe that a similar ternary complex
is formed between the reactant pyridone, Li+ and the chiral
inductor. Since Li+ is bound both to the carbonyl and the phenyl
groups, the chiral inductor would be favored to interact from the
same side where the substituent is present. Approach from this
side, due to steric considerations, would force the pyridone to
selectively adsorb from the less hindered pro-chiral face, leading to
enantiomeric excess in the product. In the case of 1f where the Li+
interacts only with the carbonyl chromophore, the chiral inductor
most likely would bind to the carbonyl of the pyridone from the
unsubstituted side, leading to no preference in the adsorption
mode. This intuitive model explains the high ee in 1h, 1i and 1j
and the low values in 1e, 1f and 1g.
Linking the chiral auxiliaries to the cyclohexadienone, naph-
thalenone and pyridone derivatives did not lead to appreciable
chiral induction in solution (Tables 2, 3 and 4) suggesting a lack
of communication between the two parts of the molecule in this
media. On the other hand, irradiation of the same substrates
as zeolite complexes resulted in a significant enhancement in
diastereoselectivity. For example, 2g gave 81% (KY zeolite) while
4f and 6l, gave 73% (NaY zeolite) and 88% (RbY zeolite) de,
respectively (Tables 2, 3 and 4). These results indicated that the
confinement provided by the zeolite framework along with cation–
organic interaction was able to exert a greater control over the
mode of cyclization.
Fig. 2 Interaction of Li+ with substrates 6g, 6k and 6l. The structures
have been computed at RB3LYP/631G* level using Gaussian 98. B. A.
refers to binding affinity.
As observed in the chiral inductor approach, the substrates
appended with a chiral auxiliary containing an aryl group gave
higher diastereoselectivity as compared to chiral auxiliary with
an alkyl substituent. The importance of aryl-substituted chiral
auxiliary in achieving greater selectivity is demonstrated by the
following example. Substrates 6g, 6k and 6l, that have an aryl
group, when irradiated within a zeolite gave the b-lactam product
in 42% (KY), 67% (KY) and 88% (RbY) de, respectively. The
substrate 6i, that has alkyl chiral auxiliary (cyclohexyl ethyl
amine), gave the b-lactam product in 24% de in LiY zeolite. To
obtain an insight into the differential behavior of alkyl- and aryl-
substituted chiral auxiliaries, computations at RB3LYP/631G*
level were performed.20 The computation suggested that for the
substrates 6g, 6k and 6l, the cation (Li+) interacted with the
carbonyl group of the chiral auxiliary, carbonyl group of the
pyridone and the phenyl group of the chiral auxiliary through
cation–p interaction (Fig. 2) (binding affinities of 6g, 6k and 6l
are 89.74 kcal mol−1, 92.66 kcal mol−1 and 96.36 kcal mol−1). This
type of interaction, that ties up the reactant part and the chiral
information together, suppresses the conformational flexibility
of the chiral auxiliary. When the phenyl group is replaced by a
cyclohexyl group (example 6i), the cation–p interaction between
the cation and the phenyl group is switched off (Fig. 3) and the
de decreased. The cation now interacts primarily with either the
pyridone part (binding affinity = 65.36 kcal mol−1) or the carbonyl
group of the chiral auxiliary (binding affinity = 63.36 kcal mol−1).
This type of interaction does not restrict the flexibility of the
chiral auxiliary. From this comparison it is clear that mobility
of chiral auxiliary should be restricted for it to be effective. In
Fig. 3 Interaction of Li+ with substrate 6i. The structures have been
computed at RB3LYP/631G* level using Gaussian 98. B. A. refers to
binding affinity.
the presence of a chiral auxiliary, especially when its mobility
is restricted, as in 6g, 6k and 6l, and it is tied to the reactant site
through cation–CO–aromatic interactions, the two prochiral faces
would not be equivalent. Under such conditions, the pyridone
part of the molecule is expected to adsorb on the zeolite surface
preferentially through one pro-chiral face, leading to significant
asymmetric induction. However, when the flexibility of the chiral
auxiliary is not restricted, as in 6i, selective adsorption is less likely
and de is expected to be low.
As observed in substrates 6g, 6k and 6l, the number of
methylene units connecting the chiral carbon of the amine to the
1574 | Org. Biomol. Chem., 2007, 5, 1569–1576
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
phenyl group in the chiral auxiliary has an effect on the extent of
diastereoselectivity. The prominent difference between these three
substrates lies in the number of methylene units (two in 6l, one
in 6k and zero in 6g) that links the chiral carbon of the amine
to the phenyl group in the chiral auxiliary. With the increase in
chain length, the phenyl group would be in a better arrangement
to interact with the cation that is bound to the carbonyls of the
pyridone and amide units. Consistent with this, the binding affinity
values for 6g, 6k and 6l to Li+ showed an increase with the number
of methylene units (binding affinity of 6g, 6k and 6l are 89.74 kcal
mol−1, 92.66 kcal mol−1 and 96.36 kcal mol−1, Fig. 2). Stronger
interaction translates into more restriction of the chiral auxiliary
and better de in the photoproduct. The phenomenon observed
here is similar to that observed during chiral inductor approach
with substrates 1h, 1i and 1j, with the only difference being the
chiral source is within the molecule. Though the calculations were
performed with Li+ cations, the other charge-compensating alkali
metal ions are also expected to have a similar type of interaction
with the organic molecule as with the case of Li+ cations.
inductor were taken in 4 ml of acetonitrile solvent. The samples
were then irradiated using a 450 W medium pressure mercury arc
lamp in a water-cooled immersion well. The chiral inductor was
removed by performing a column using silica gel with hexane–
ethyl acetate solvent mixture. The stereoselectivity obtained in
the product was measured using HPLC/GC. The conditions used
for measuring the enantiomeric excess are provided in Table S1
(see ESI†). The enantiomeric or diastereomeric excesses were
determined using the formula:
[(area of A − area B)/(area of A + area of B)] × 100.
A and B refer to the first and second peak of the product
enantiomers/diastereomers.
Chiral inductor approach within zeolites
Chiral inductor (10–15 mg or 20–30 mg) and substrates (1a–
1j) (1.5–2.0 mg or 2.0–2.5 mg) were dissolved in 0.5 ml of
dichloromethane followed by the addition of 15 ml of hexanes.
MY (Li+, Na+, K+, Rb+ and Cs+) zeolite (150 mg or 300 mg)
◦
activated at 500 C for 8 h was added with stirring. The loading
As mentioned earlier, the nature of the cation not only controlled
the extent of diastereoselectivity but in examples 4f, 4g, 4h, 4o, 4q,
6l, and 6m, it also controlled the diastereomer being enhanced. It
has been reported in the literature that Li+ and Na+ bind differently
to amino acids such as glycine, valine and arginine compared
to K+. Li+ binds to these molecules preferentially through CO,
NH2 (N-, O-coordination) whereas K+ binds to the oxygens of
the COOH group (O-, O-coordination).22–24 This could be the
reason for observing cation-dependent diastereomeric switch in
these substrates. Results along the same lines were obtained dur-
ing the photoisomerization of 2b, 3b-diphenylcyclopropane-1a-
carboxylic acid derivatives25 and c-hydrogen abstraction reaction
of a-oxoamides within zeolites.26
The results presented in this report on the photocycliza-
tion of dienones, naphthalenones and N-alkyl pyridones within
zeolites compliment our earlier investigations on the pho-
tocyclization of tropolone derivatives,27 the geometric iso-
merization of 1,2-diphenylcyclopropanes25 and 2,3-diphenyl-1-
benzoyl cyclopropanes,28 and the Norrish type II reaction of
a-oxoamides,29 phenyl adamantyl ketones,30 phenyl norbornyl
ketones and phenyl cyclohexyl ketones.31 With the help of these
examples, we have established the importance of zeolite and
its charge compensating cations in effecting chiral induction in
photochemical reactions. In spite of this, we are still unable to
formulate rules that could help us to predict the outcome of a
photoreaction within a zeolite.
level of substrate was kept at one molecule for every 10 supercages.
A higher ratio of the chiral inductor was employed to maximize
the chances of every reactant molecule being adjacent to the chiral
inductor within the supercage. After stirring the slurry for 12 h,
the supernatant was analyzed for reactant and chiral inductor.
Fresh hexanes were added after removing the old hexanes under
nitrogen. The sample was then irradiated using a 330 nm cut-
off uranyl filter for substrates 1a–1d and a pyrex cut-off filter for
substrates 1f–1j. The irradiated slurry was filtered, and washed
again with hexane. The reactant and the photoproducts were
extracted from the zeolite by stirring with acetonitrile for 12 h. The
extract was concentrated and the reactant and its photoproducts
were separated from the chiral inductors by a microcolumn
(silica gel) using hexane–ethyl acetate as eluent. The analyses
were conducted by GC/HPLC fitted with chiral columns. All the
samples were irradiated to achieve 30% conversion to products.
Chiral auxiliary approach—solution photolysis
2 mg of the substrate (2 and 4) to be irradiated was dissolved in
5 ml of trifluoroethanol, degassed with nitrogen, followed by 15–20
minutes irradiation. Conversion to photoproduct was monitored
on a gas chromatograph, HP-5 capillary column. 2mg of the
substrate (6) to be irradiated was dissolved in 5 ml of acetonitrile,
degassed with nitrogen, followed by 2 h of irradiation with a pyrex
cut-off filter. Conversion to photoproduct was monitored on a gas
chromatograph, HP-5 capillary column.
Experimental
Zeolite photolysis
Loading and photolysis procedure
The substrates (2, 4 and 6) (1.5–2.0 mg or 2.0–2.5 mg) were
dissolved in 0.5 ml of dichloromethane followed by addition of
15 ml of hexanes. MY (Li+, N◦a+, K+, Rb+ and Cs+) zeolite (150 mg
or 300 mg) activated at 500 C for 8 h was added with stirring.
The loading level of substrate was kept at one molecule for every
10 supercages. After stirring the slurry for 12 h, fresh hexanes
were added to the slurry after removing the old hexanes under
nitrogen. The supernatant was analyzed for the reactant. The
sample was then irradiated using an appropriate filter (330 nm
for substrates 2, 4 and pyrex cut-off filter for substrates 6) filtered,
All photolysis experiments were carried out by using a 450 Watt
medium pressure mercury arc lamp placed in a water-cooled pyrex
immersion well.
Chiral inductor approach—solution reaction
The solution reactions were performed in acetonitrile solvent. The
typical irradiation procedure is as follows. For enantiomeric excess
studies, 1.5 mg of the substrates along with 15 mg of the chiral
This journal is
The Royal Society of Chemistry 2007
Org. Biomol. Chem., 2007, 5, 1569–1576 | 1575
©

View Article Online
washed again with hexanes. Substrates 2 and 4 were irradiated for
10 min. and substrates 6 were irradiated for 2 h. The reactant and
the photoproducts were extracted from the zeolite by stirring with
acetonitrile. The extract was concentrated and the analyses were
conducted in HPLC/GC fitted with chiral columns.
12 J. Griffiths and H. Hart, J. Am. Chem. Soc., 1968, 90, 3297.
13 J. Shailaja, P. H. Lakshminarasimhan, A. R. Pradhan, R. B. Sunoj, S.
Jockusch, S. Karthikeyan, S. Uppili, J. Chandrasekhar, N. J. Turro and
V. Ramamurthy, J. Phys. Chem. A, 2003, 107, 3187.
14 E. J. Corey and J. Streith, J. Am. Chem. Soc., 1964, 86, 950.
15 R. C. De Selms and W. R. Schleigh, Tetrahedron Lett., 1972, 34, 3563.
16 R. Matsushima and K. Terada, J. Chem. Soc., Perkin Trans. 2, 1985, 9,
1445.
17 F. Toda and K. Tanaka, Tetrahedron Lett., 1988, 29, 4299.
18 L.-C. Wu, C. J. Cheer, G. Olovsson, J. R. Scheffer, J. Trotter, S.-L. Wang
and F.-L. Liao, Tetrahedron Lett., 1997, 38, 3135.
Computational methods
Full geometry optimizations were carried out primarily using the
hybrid Hartree–Fock density functional theory (RB3LYP) with
Becke three parameter exchange functional in conjunction with
correlation function by Lee, Yang and Parr. The 6–31G* basis
set was used for C, H, O, N and Li. Stationary points have
been characterized as true minima by frequency calculations. All
the calculations were performed using Gaussian 98 A.11 suite of
quantum chemical programs.20
19 K. Tanaka, T. Fujiwara and Z. Urbanczyk-Lipkowska, Org. Lett., 2002,
4, 3255.
20 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,
P. Salvador, J. J. Dannenberg, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul,
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, M. Challacombe, P. M. W. Gill, B. G.
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.11), Gaussian, Inc., Pittsburgh, PA, 2001.
Acknowledgements
V. R. thanks the NSF for financial support (CHE-0212042)
21 (a) J. Shailaja, L. S. Kaanumalle, S. Karthikeyan, A. Natarajan, K. J.
Ponchot, A. R. Pradhan and V. Ramamurthy, Org. Biomol. Chem.,
2006, 4, 1561; (b) S. Hoyau and G. Ohanessian, Chem.–Eur. J., 1998, 4,
1561–1569.
References
1 Y. Inoue, in ‘Chiral Photochemistry’, ed. Y. Inoue and V. Ramamurthy,
Marcel Dekker Inc., New York, 2004.
2 Y. Inoue, Chem. Rev., 1992, 92, 741.
22 R. A. Jockusch, A. S. Lemoff and E. R. Williams, J. Am. Chem. Soc.,
3 H. Rau, in ‘Chiral Photochemistry’, ed. Y. Inoue and V. Ramamurthy,
2001, 123, 12255.
Marcel Dekker Inc., New York, 2004.
23 T. Wyttenbach, M. Witt and M. T. Bowers, J. Am. Chem. Soc., 2000,
122, 3458.
4 B. Grosch and T. Bach, in ‘Chiral Photochemistry’, ed. Y. Inoue and
V. Ramamurthy, Marcel Dekker Inc., New York, 2004.
5 Y. Inoue, H. Ikeda, M. Kaneda, T. Sumimura, S. R. L. Everitt and T.
Wada, J. Am. Chem. Soc., 2000, 122, 406.
6 J.-P. Pete and N. Hoffmann, in ‘Chiral Photochemistry’, ed. Y. Inoue
and V. Ramamurthy, Marcel Dekker Inc., New York, 2004.
7 J. R. Scheffer, in ‘Chiral Photochemistry’, ed. Y. Inoue and V. Rama-
murthy, Marcel Dekker Inc., New York, 2004.
24 R. A. Jockusch, W. D. Price and E. R. Williams, J. Phys. Chem. A,
1999, 103, 9266.
25 J. Sivaguru, R. B. Sunoj, T. Wada, Y. Origane, Y. Inoue and V.
Ramamurthy, J. Org. Chem., 2004, 69, 6533.
26 A. Natarajan and V. Ramamurthy, Org. Biomol. Chem., 2006, 4, 4533.
27 A. Joy, L. S. Kaanumalle and V. Ramamurthy, Org. Biomol. Chem.,
2005, 3, 3045.
8 K. Tanaka and F. Toda, in ‘Organic Solid-State Reactions’, ed. F. Toda,
Kluwer Academic Publishers, Dordrecht, Netherlands, 2002, pp. 1–46.
9 J. Sivaguru, A. Natarajan, L. S. Kaanumalle, J. Shailaja, S. Uppili, A.
Joy and V. Ramamurthy, Acc. Chem. Res., 2003, 36, 509.
10 V. Ramamurthy, A. Natarajan, L. S. Kaanumalle, S. Karthikeyan,
J. Sivaguru, J. Shailaja and A. Joy, in ‘Chiral Photochemistry’, ed.
V. Ramamurthy and Y. Inoue, Marcel Dekker Inc., New York, 2004.
11 H. Hart and R. K. Murray, J. Org. Chem., 1970, 35, 1535.
28 J. Sivaguru, R. B. Sunoj, T. Wada, Y. Origane, Y. Inoue and V.
Ramamurthy, J. Org. Chem., 2004, 69, 5528.
29 A. Natarajan, K. Wang, V. Ramamurthy, J. R. Scheffer and B. Patrick,
Org. Lett., 2002, 4, 1443.
30 A. Natarajan, A. Joy, L. S. Kaanumalle, J. R. Scheffer and V.
Ramamurthy, J. Org. Chem., 2002, 67, 8339.
31 A. Natarajan, V. Ramamurthy and J. T. Mague, Mol. Cryst. Liq. Cryst.,
2006, 456, 71.
1576 | Org. Biomol. Chem., 2007, 5, 1569–1576
This journal is
The Royal Society of Chemistry 2007
©