Triplet photochemistry within zeolites through heavy atom effect, sensitization and light atom effect

Article abstract of DOI:10.1016/S0040-4020(03)00865-2

Methods to generate triplets of organic molecules within zeolites have been established by employing the Zimmerman rearrangement of barrelenes, oxa-di-π-methane rearrangement of β,γ-unsaturated ketones and photodimerization of acenaphthylene as probe reac

Full text of DOI:10.1016/S0040-4020(03)00865-2

Tetrahedron 59 (2003) 5763–5772
Triplet photochemistry within zeolites through heavy atom effect,
sensitization and light atom effect
*
K. Pitchumani, M. Warrier, Lakshmi S. Kaanumalle and V. Ramamurthy
Department of Chemistry, Tulane University, 6400, Freret Street, New Orleans, LA 70118, USA
Received 7 February 2003; revised 23 May 2003; accepted 27 May 2003
Abstract—Methods to generate triplets of organic molecules within zeolites have been established by employing the Zimmerman
rearrangement of barrelenes, oxa-di-p-methane rearrangement of b,g-unsaturated ketones and photodimerization of acenaphthylene as probe
reactions. The two methods, heavy cation effect and triplet sensitization, are well established solution techniques and these work well within
zeolites. The Zimmerman rearrangement of dibenzobarrelene is enhanced even within Liþ and Naþ exchanged zeolites and these are
believed to be the result of slowing of the rearrangement to dibenzocyclooctatetraene from S1 through cation-p interaction. The methods
described here provide an opportunity to explore the control afforded by the zeolite environment on triplet reactions.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
state giving semibullvalenes 3, 4 and 7. From excited singlet
state these molecules yield cyclooctatetraene derivatives 2
and 6. Acenaphthylene upon excitation undergoes photo-
dimerization to give both cis and trans dimers. Of the two,
the trans dimer (10) is derived only from the triplet state
while the cis dimer (9) from both the triplet and excited
singlet states. The product distributions obtained in these
three systems within a zeolite would therefore, be a
reflection of the extent of intersystem crossing under a
given condition. Employing the photochemistry of b,g-
unsaturated ketones 11, 14, and 177 we were able to
establish the limitations of the zeolite based heavy cation
technique. These molecules yield products of 1,3-acyl (12,
15 and 18) and 1,2-acyl shifts (13, 16 and 19; oxa-di-p-
methane rearrangement)) predominantly from S1 and T1
states, respectively (Scheme 1). The a-cleavage leading to
1,3-acyl migration occurs both from S1 and T1 of npp
character while oxa-di-p-methane rearrangement leading to
1,2-acyl shift occur essentially from ppp triplet. Since no
significant heavy cation effect was observed with b,g-
unsaturated ketones, we felt it important to establish a
general method to generate triplets of molecules within
zeolites. To this effect we have explored the well-known
triplet sensitization technique within zeolites.8 Results of
heavy cation and triplet sensitization methods to generate
reactive triplets within zeolites are presented below.
Early work by Kasha and Lewis had postulated the
existence of a phosphorescent triplet state and the important
role it plays in photochemical processes of organic
molecules.1 The use of external heavy atoms to increase
singlet–triplet transition probabilities, first reported by
Kasha in 1952, quickly became a common practice.2
Spin–orbit coupling, induced by heavy atoms, provides
the crucial mechanism, which allows the otherwise
forbidden interconversion between singlet and triplet states.
Molecules that are employed to influence the singlet–triplet
interconversion include oxygen, alkyl halides, organo-
metallic compounds, and rare gases such as xenon.3 We
reasoned that if a heavy atom perturber and an organic
molecule could be ‘closeted’ within a constrained environ-
ment, such as zeolites, the interactions between them could
be strengthened, leading to stronger heavy atom effects. In
this context, reports on enhanced intersystem crossing and
phosphorescence yields in the case of aromatics, olefins and
azoalkanes included in heavy cation (e.g. Csþ, Tlþ)
exchanged faujasites are note worthy.4
By employing the Zimmerman rearrangement of dibenzo-
barrelene (1) and benzobarrelene (5)5 and photodimeriza-
tion of acenaphthylene (8)6 as probe reactions we show in
this report that the heavy cation technique could also be
used to control photoproduct distributions within zeolites
(Scheme 1). For dibenzobarrelene and benzobarrelene, the
Zimmerman rearrangement proceeds solely from the triplet
2. Results
Results of photolysis of barrelenes 1 and 5 in various media
are presented in Tables 1–3. In isotropic solution medium,
there is a preference for cyclooctatetraene (2 and 6). In
Keywords: benzobarrelene; zeolites; Zimmerman rearrangement.
*
Corresponding author. Tel.: þ1-504-862-8135; fax: þ1-504-865-5596;
e-mail: murthy@tulane.edu
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00865-2

5764
K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
Scheme 1.
presence of a triplet sensitizer (acetone for 1 and
acetophenone for 5), under conditions where only the
sensitizer absorbed the light only semibullvalenes 3, 4 and 7
were obtained. Zeolite loaded barrelenes were irradiated in
the form of solid powder and hexane-slurry. For the solid
irradiations, reactant ‘loaded’ zeolite samples were trans-
ferred to quartz/pyrex tubes and photolyzed directly as a
powder. For the slurry irradiations, the loaded zeolite
samples were transferred into fresh hexane and photolyzed
as a hexane slurry with continuous stirring. Cation
exchanged X zeolites were used as media.
dibenzobarrelenes 1b and 1c (Table 3). Ability of Liþ and
Naþ in bringing out triplet chemistry decreased when water
was used as the co-adsorbent. In ‘wet’ zeolites the yield of
the triplet product gradually increased with the atomic
number of the cation (Liþ,Naþ,Kþ,Rbþ,Csþ,Tlþ).
As opposed to dibenzobarrelene, benzobarrelene did not
yield significant amounts of semibullvalene 7 within Liþ
and Naþ exchanged zeolites. Also in this case (benzo-
barrelene) only Tlþ was able to induce significant triplet
reactivity.
In Table 4 results of solid state irradiation of acenaphthylene
included in alkali ion exchanged Y zeolites are presented.
Photochemistry of acenaphthylene in solution has been
extensively investigated.6 Direct excitation gives the cis
dimer while triplet sensitization yields both cis and trans
dimers in the ratio ,1:2 (methanol). From the table, it is
clear that the ratio of the trans to the cis dimer depends on
the cation. Essentially, the cis dimer was obtained in LiY
while in RbY a mixture of trans and cis dimers were formed
The ratio of products derived from excited singlet and triplet
states varies with the cations present within a zeolite.
Mainly triplet derived products were obtained from 1 and 5
within Tlþ exchanged zeolites. Unexpectedly, in the case of
dibenzobarrelene higher yields of the triplet products were
obtained even within LiX and NaX as compared to KX
(Liþ.Naþ.Kþ,Rbþ,Csþ,Tlþ). As shown in Table 3,
this unusual phenomenon also occurs with two substituted

K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
Table 1. Product distribution upon irradiation of dibenzobarrelene (1a)
5765
Medium
Conversion (%)
Slurrya (%)
Solida (%)
2a
3a
2a
3a
Acetonitrile
Acetone
LiX
NaX
KX
RbX
CsX
TlX
77
0
23
100
4(23)b
6(20)
17(30)
16(26)
24(41)
98(88)
33{80}c
38{72}
53{57}
25{31}
13{17}
,1{,1}
67{20}
62{28}
47{43}
75{69}
87{83}
.99{.99}
18{57}
25{44}
38{42}
16{24}
6{12}
82{43}
75{56}
62{58}
84{76}
94{88}
.99{.99}
,1{,1}
Sensitizer
Acetophenone in KY
Loading leveld
25
40
85
5
11
25
95
89
75
1
2
5
99
98
95
4-Methoxy acetophenone in KY
25
40
85
,1
4
13
.99
96
87
,1
,1
,1
.99
.99
.99
a-Aminoaceto phenone HCl
25
40
8
16
92
84
5
10
95
90
Xanthone in KY
25
40
85
5
8
20
95
92
80
2
2
23
98
98
77
a
Slurry irradiations (hexane as solvent) were carried out for 2 h. Solid state irradiations were carried out for 20 h. Conversions are comparable since all
irradiations were conducted under identical conditions.
Numbers in bracket are for solid state irradiations.
Numbers in parentheses are for zeolite co-adsorbed with water.
Loading levels refer to the average number of supercages per guest molecule. For example, a value of 25 indicates one probe molecule and one sensitizer
molecule per 25 supercages.
b
c
d
Table 2. Product distribution upon photolysis of benzobarrelene (5)
Medium
Conversion (%)
Slurrya,b (%)
Solida,b (%)
6
7
6
7
Hexane
Sensc
LiX
NaX
KX
RbX
CsX
TlX
96
0
4
100
10
5
11
10
14
92
70(31)
64(25)d
74(13)
76(20)
63(25)
79(25)
90
95
89
90
86
8
95
93
82
84
82
5
5
7
18
16
18
95
Sensitizer
Acetophenone in KY
Loading levele
13
25
40
17
27
47
83
73
53
6
8
12
94
92
88
4-Methoxy acetophenone in KY
13
25
40
10
34
51
90
66
49
1
2
20
99
98
80
a-Aminoaceto phenone Hcl in KY
13
25
39
54
61
46
18
63
82
37
Xanthone in KY
13
25
40
21
44
68
79
56
32
5
18
28
95
82
72
a
The product ratios are independent of % conversion within the estimated error limits of ^2% and represent an average of at least 5 independent runs.
Slurry irradiations were conducted in hexane for 2 h. Solid irradiations were carried out for 8 h.
Acetophenone sensitization in hexane solution.
Numbers in parentheses are for solid state irradiations.
Loading levels refer to the average number of supercages per guest molecule. For example, a value of 25 indicates one probe molecule and one sensitizer
molecule per 25 supercages.
b
c
d
e

5766
K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
Table 3. Product distribution upon irradiation of dibenzobarrelenes 1b and
1c under dry conditions
found to depend on the % of Csþ ion content in NaY
(Table 4). While in NaY the ratio of cis to trans dimer was
25, that in 8% Csþ exchanged NaY was 3.3. These
observations suggest that the heavy cation such as Rbþ
and Csþ favor the production of triplet acenaphthylene.
Medium
Conversion (%)a,b
Slurryc
Solidc
2b/c 3b/c 4b/c 2b/c 3b/c 4b/c
Reactant 1b
Acetonitrile
Acetone
LiX
NaX
KX
RbX
CsX
TlX
Unlike reactants 1, 5 and 8 there was no significant
difference in product distribution upon photolysis of b,g-
unsaturated ketones 11, 14 and 17 included in KX and TlX
(Table 5). In these systems even the heavy Tlþ ion failed to
produce significant amount of oxa-di-p-methane rearrange-
ment product that is believed to derive only from T1 (ppp).
This prompted us to examine the triplet sensitization
technique within zeolites. Triplet sensitization experiments
were conducted for substrates 1, 5, 11, 14 and 17 in KY
zeolite using acetophenone, 4-methoxy acetophenone,
xanthone and a-aminoacetophenone hydrochloride as
sensitizers. In these experiments, only two products were
formed, one from S1 and the other from T1. To monitor the
efficiency of triplet energy transfer all samples were
irradiated under identical conditions for the same length
of time. We wish to highlight that sensitization occurs even
under very low loading levels and sensitization is a reliable
technique by which triplets of organic molecules could be
produced within a zeolite. Results of these sensitization
experiments are included in Tables 1, 2 and 5. In these
tables, loading level corresponds to the number of super-
cages available for one molecule of the substrate and the
sensitizer. For substrates 1 and 5, it was evident that even at
very ‘low’ loading levels like 85 (one molecule of the
reactant and one molecule of the sensitizer per 85 super-
cages) yield of the triplet product was still very high within
zeolites, suggesting efficient sensitization. 40-Methoxy
acetophenone was found to be the best sensitizer among
the ones investigated. For example, even at a low loading
level of 85, 87% triplet product yield was obtained during
the photolysis of dibenzobarrelene (1) as a hexane slurry
(Table 1). For the same loading level, triplet product yields
were 75, 84 and 80 with acetophenone, a-amino aceto-
phenone hydrochloride and xanthone as sensitizers
(Table 1). In general, sensitization was more effective in
the solid state than in hexane slurry. For example, at a
loading level of 25, with 40-methoxy acetophenone as the
sensitizer, yield of the triplet product from 5 was 98% in the
95
–
4
1
45
5
13
10
14
19
55
74
58
39
66
78
40(29)
36(24)
43(15)
46(36)
54(11)
34
21
29
51
20
3
36
60
68
22
26
29
38
31
57
72
35
2
1
21
2
,1 .99
,1 .99
Reactant 1c
Acetonitrile
Acetone
LiX
NaX
KX
RbX
CsX
TlX
73
25
35
8
3
1
2
–
65
71
55
44
78
82
12(8)
14(2)
68(14)
73(47)
37(37)
87(52)
21
45
55
18
12
30
51
67
12
10
66
46
23
77
83
4
3
10
11
7
4
6
,1 .99
,1 .99
a
Loading level refers to number of available supercages per single
molecule of the substrate/sensitizer. For direct irradiations, loading level
was kept near 25 (20–25), i.e. one molecule per 25 supercages.
Numbers in bracket are for solid state irradiation.
Slurry irradiations (hexane as solvent) were carried out for 2.5 h. Solid
state irradiations were carried out for 20 h. Conversions are comparable
since all irradiations were conducted under identical conditions.
b
c
in nearly 1:1 ratio. Consistent with the conclusion that the
trans dimer comes from the triplet state, irradiation of the
RbY samples under oxygenated conditions gave only the cis
dimer. Also, when the triplet quencher ferrocene was
co-included (along with acenaphthylene) within RbY the
trans dimer was selectively quenched (Table 4). These
suggested that while in LiY acenaphthylene reacts only
from S1, within RbY dimerization occurs from both S1 and
T1. Further confirming the role of heavy cation in favoring
the trans dimerization the ratio of trans to cis dimer was
Table 4. Photodimerization of acenaphthylene within zeolites dependence
of cis to trans dimer ratio on the condition of irradiation
Zeolite
Conditiona,b
cis/trans Dimer ratio
Hexane solution, 5 mg in 5 mL
cis only
16(25)c
2.3(25)c
1.6(25)c
0.6(25)c
1.1(6.0)c
1.9
LiY
NaY
KY
RbY
CsY
RbY
RbY
RbY
RbY
NaY
NaY–CsY
NaY–CsY
NaY–CsY
Table 5. Product distribution upon irradiation of b,g-unsaturated ketones
11, 14 and 17 within zeolites under direct and sensitized conditions
Reactant
Conditions for irradiation
Product (%)
Ferrocene, 20d
Ferrocene, 10d
Ferrocene, 5d
Ferrocene, 2d
0% Csþ
3.2
4.3
8.0
25
22
18
3.3
12
13
11
KX
TlX
.99
89
, 1
11
0.1% Csþþ
KX with 4-methoxy acetophenone
3
97
1.0% Csþ
8.0% Cs
15
.99
.99
18
16
, 1
, 1
82
a
14
17
KX
TlX
Loading level of acenaphthylene in all cases was maintained at one
molecule per five supercages.
All irradiations were conducted as solid samples under deaerated
conditions; for exception see footnote c.
Numbers in parenthesis refers to irradiation under oxygen-saturated
conditions.
The number refers to loading level, which is defined as number of
supercages per molecule. The number 10 means 1 molecule per 10
supercages.
b
c
d
KX with 4-methoxy acetophenone
18
75
79
19
19
25
21
81
KX
TlX
KX with 4-methoxy acetophenone

K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
5767
solid state, while it was only 66% in the slurry mode
(Table 2). As shown in Table 5, the triplet sensitization
technique satisfactorily worked also for b,g-unsaturated
ketones 11, 14 and 17. Thus triplet sensitization is fairly
general within zeolites. To examine whether the sensitiz-
ation occurs by a dynamic or a static process, the triplet
lifetime of the sensitizer 40-methoxy acetophenone was
monitored (by0 following the decay of the triplet–triplet
absorption of 4 -methoxy acetophenone, 350–450 nm) with
increasing concentration of the quencher benzobarrelene.
The triplet lifetime decreased in a linear fashion from 16 to
2.1 ms when the loading level of benzobarrelene was varied
between 0 molecule and 1 molecule per five supercages.
intersystem crossing from S1 to T1 would be enhanced in
these media. Supporting the conclusion that the trans dimer
is derived from T1, its formation within RbY is quenched by
the triplet quenchers oxygen and ferrocene (Table 4). The
possibility of an enhanced heavy atom effect within the
constrained zeolite cages has been previously established
through photophysical studies.4 Examples presented above
establish that the same effect could also be employed to
control photoproduct distributions.
While the heavy cation effect worked nicely with molecules
containing aromatic and olefinic chromophores, it failed in
the case of enones 11, 14 and 17. In these cases, the relative
yields of products due to oxa-di-p-methane rearrangement
are an indication of the effectiveness of the heavy cations.
Similar product distributions obtained within KX and TlX
suggest that the heavy cation Tlþ is unable to enhance the
rate of intersystem crossing with respect to a-cleavage
reaction from S1. Given that the primary intermediate
formed via the a-cleavage process is an allyl radical, the
rate of a-cleavage in b,g-unsaturated ketones 11, 14 and 17
(Scheme 1) from S1 must have high rates. Further,
considering the rules of intersystem crossing rates (S1 to
T1) in organic molecules formulated by El Sayed, absence
of heavy cation effect in b,g-unsaturated ketones is not a
surprise.10 The rate of intersystem crossing from S1 to
nearby triplet is inherently larger when the states involved
have different characters (npp and ppp). In b,g-unsaturated
ketones the intersystem crossing is believed to be between
S1 (npp) and triplet of mixed npp and ppp character.
3. Discussion
3.1. Heavy atom effect
As seen in Table 2, in the case of benzobarrelene, the
percentage of benzosemibullvalene (triplet product)
increased from 5% in NaX to 92% in TlX. Similarly, for
1a, within wet zeolites the yield of dibenzosemibullvalene
(triplet product) increased from 20% in LiX to 99% in TlX
under slurry irradiations and from 43 to 99% under solid
irradiation conditions (Table 1). Similar observations were
also made with substituted dibenzobarrelenes (Table 3). The
observed abundance of semibullvalenes as photoproducts in
heavy cation exchanged zeolites, we believe, is the result of
enhanced intersystem crossing rates in presence of heavy
cations such as Csþ and Tlþ. The increase in intersystem
crossing (and the yield of triplet product) between Liþ and
Tlþ exchanged zeolites is consistent with the difference in
spin–orbit coupling parameter for the corresponding atoms
(Li: 0.23 cm21 and Tl: 3410 cm21).9 Examination of the
photobehavior (especially under wet conditions) of dibenzo-
barrelene (Table 1) in various alkali ion exchanged zeolites
reveal that the relative yield of the triplet product is
dependent on the spin–orbit coupling parameter of the
cation. The amount of triplet product increases in the order
Liþ,Naþ,Kþ,Rbþ,Csþ,Tlþ. It is important to note
that the spin–orbit coupling parameter for the correspond-
ing atoms also increase in the same order (0.23, 11.5, 38,
3.2. Triplet sensitization within zeolites
By using the heavy cation effect we could not improve the
yields of the triplet products in the case of b,g-unsaturated
ketones 11, 14 and 17 (Scheme 1) within zeolites. This
prompted us to explore a general method that would enable
one to produce triplets of all classes of molecules within
zeolites. With this in mind we have examined the triplet
sensitization of barrelenes and b,g-unsaturated ketones
within KY (Tables 1, 2 and 5). As seen from the results
presented in Tables 1 and 2 sensitization is very efficient
within zeolites.11 The triplet sensitization technique satis-
factorily worked also for b,g-unsaturated ketones 11, 14 and
17. Loading level in the tables refers to the number of
supercages per reactant and sensitizer molecules. Energy
transfer was efficient (87% triplet product from 1a) even at
loading levels of one molecule of the sensitizer and
substrate in 85 supercages. Generally, solid state
irradiations were more efficient than slurry irradiations.
For example, at the same loading level of 85, with 40-
methoxy acetophenone as the sensitizer, solid irradiation of
1a gave 99% of the triplet product while slurry irradiation
yielded only 87%. Of the four sensitizers used 40-methoxy
acetophenone was most effective. A point to note is that the
only other product formed under the sensitization conditions
was that from S1.
160, 370 and 3410 cm21
,
respectively).9 Similar
observations were made with dibenzobarrelenes 1b and 1c
(Table 3). However, in the case of benzobarrelene only Tlþ
was most effective in bringing about triplet reactivity.
Results obtained with acenaphthylene establish unequivo-
cally that the heavy cations present in zeolites could be used
to control triplet production. Photolyses of dry solid
inclusion complexes of acenaphthylene in various cation
(Liþ, Naþ, Kþ, Rbþ, and Csþ) exchanged Y zeolites gave
the cis and trans dimers. As seen in Table 4, the cis to trans
dimer ratio is dependent on the cation. Absence of the trans
dimer in LiY and NaY is consistent with the solution
behavior in which the intersystem crossing yield from S1 to
T1 is reported to be near zero (F¼0.00^0.02).6 Exclusive
formation of the cis dimer within LiY and NaY suggests that
in these media the excited singlet state is trapped by the
ground state acenaphthylene prior to intersystem crossing to
the triplet state. Formation of the trans dimer in K, Rb and
CsY is consistent with the expectation that the rate of
Qualitatively, there could be three factors contributing to the
observed efficiency of energy transfer within zeolites:
intrazeolite diffusion of the donor and the acceptor, long
triplet lifetime of the sensitizers and selective donor–
acceptor aggregation in the same supercage. Importance of

5768
K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
diffusion in the process of energy transfer is revealed by the
transient studies where the lifetime of the triplet of 40-
methoxy acetophenone was monitored in presence of
benzobarrelene. The triplet lifetimes were 16, 14, 12, 8.1
and 2.1 ms at 0, 0.02, 0.04, 0.07 and 0.2 molecules per
supercage. If the energy transfer occurs via a static process
the lifetime of the donor 40-methoxy acetophenone would
not be influenced by variation in the loading level of
benzobarrelene. The fact that the triplet lifetime of 40-
methoxy acetophenone within KY was dependent on the
loading level of benzobarrelene suggests that the energy
transfer process was taking place predominantly via a
diffusion-controlled processes. At a loading level of 1
molecule of the sensitizer and 1 molecule of the acceptor in
the ‘light atom effect’ did not operate in the case of
benzobarrelene.
Any discussion of the likely origin of the light atom effect in
the case of dibenzobarrelene should also address absence of
it in the case of benzobarrelene. The increased yield of triplet
product could either be due to the increased rate of
intersystem crossing from S1 to T1 or due to the decreased
rate of the rearrangement in S1. In the absence of any known
mechanisms by which light cations could increase the rate
of intersystem crossing from S1 to T1, we believe that
increased yield of triplet product formation is due to the
decreased rate of rearrangement of dibenzobarrelene to
dibenzocyclooctatetraene from S1. Proposed mechanisms
for conversion of dibenzobarrelene to dibenzocyclooctate-
traene shown in Scheme 2 involve initial benzo–vinyl
bridging.5,13 On the other hand, the rearrangement of
benzobarrelene to benzocyclooctatetraene could proceed
via either benzo–vinyl bridging or vinyl–vinyl bridging
(Scheme 2).14 Clearly benzobarrelene has an additional
pathway (vinyl–vinyl bridging), which is lacking in
dibenzobarrelene.
12
˚
80 cages, the two are expected to be separated by ,40 A.
At this loading level the two molecules would have to travel
˚
,15 A before energy transfer can occur. The details of
energy transfer phenomenon within zeolites will be
addressed in a future publication.
The importance of intrazeolite diffusion was suggested also
by the results of different sensitizers under solid and slurry
conditions. Both xanthone and a-amino acetophenone
hydrochloride would not be expected to be very mobile
within a zeolite; the former due to size and the latter due to
ionic interaction with the surface. Implying that diffusion is
important for efficient energy transfer the yields of triplet
product from 5 at a loading level of 13 under identical
conditions of irradiation were 61 and 79% with a-amino
acetophenone hydrochloride and xanthone as sensitizers,
while with 40-methoxy acetophenone the yield was 90%.
Due to reduced mobility of the acceptor and the sensitizer in
presence of solvent molecules sensitization was less
effective during slurry irradiations.
In order to explore whether the cation–aromatic/olefin p
interaction within zeolites is responsible for the decreased
reactivity in S1, we computed the structures of Liþ complex
with benzobarrelene and dibenzobarrelene at RB3LYP/
6-31G(d) level.15 The computed structures are shown in
Figure 1. In the case of benzobarrelene two structures were
identified. The one where Liþ is interacting with the aryl
part is much more stable than the one in which Liþ is
interacting with the vinyl part. Further, Liþ binds prefer-
entially to the phenyl rather than co-operatively to both
phenyl and vinyl chromophores. Assuming that within
zeolites the same trend prevails, benzobarrelene would be
bound within zeolites through Liþ· · ·phenyl interaction and
the vinyl parts would not be influenced by the cation. Such
type of ion-bound benzobarrelene would be expected to
react equally efficiently through vinyl–vinyl bridging as the
3.3. Light atom effect
Based on the photobehavior of aromatics and olefins we
expected that dibenzobarrelene would not yield products
from the triplet state in light cation exchanged zeolites. On
the contrary, we isolated larger than expected amounts of
dibenzosemibullvalenes, the triplet derived products, from
three dibenzobarrelene systems in LiX and NaX than in KX.
For example, during solid state irradiation of 1a included in
LiX, NaX and KX the yields of dibenzosemibullvalene were
82, 75 and 62%, respectively. A similar trend was observed
also during slurry irradiations in X zeolites (67, 62 and 47%)
(Table 1). Perusal of Table 3 indicates that substituted
dibenzobarrelenes 1b and 1c also show a similar trend. The
decreasing trend (as opposed to the increasing trend based
on heavy cation effect) in the triplet product yield from
LiþX to KþX zeolites in all three systems suggested that the
cation binding to dibenzobarrelenes is likely to play an
important role in this process. Importance of cation–
dibenzobarrelene interaction became apparent when the
irradiation was conducted in presence of co-adsorbed water
molecules. As would be expected based on classical heavy
atom effect the yield of the triplet product increased: LiX,
43%; NaX, 56%; KX, 58%; RbX, 76%; CsX, 88% and TlX
99% (solid irradiation). The above phenomenon of
increased triplet product formation in presence of Liþ and
Naþ is unusual and in the absence of a better term we call
this the ‘light atom effect’. An important point to note is that
Scheme 2.

K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
5769
Figure 1. Optimized structures for Liþ complexes of dibenzobarrelene (top) and benzobarrelene (bottom) at RB3LYP/6-31G(d) level. Binding affinities are
included at the bottom. Two structures were identified in each case.
unbound benzobarrelene. Because of this one would not
expect the light cations to influence the rate of the
rearrangement of benzobarrelene to benzocyclooctate-
traene. This prediction is consistent with the observed
results. On the other hand, in the case of dibenzobarrelene
two structures nearly of equal energies were computed
(Fig. 1). In one Liþ interacts selectively with one phenyl
group and in the other co-operatively with both phenyl
groups. Reactivity of the phenyl groups in both structures
would be lower than that in free dibenzobarrelene. We
believe that such a reduction in reactivity could enhance the
triplet yield and thereby favor the formation of benzo-
semibullvalene at the expense of dibenzocyclooctatetraene.
The computed binding energies of the alkali ion to
dibenzobarrelene decrease in the order Liþ.Naþ.Kþ
(Fig. 1). One would expect the influence of the cation to
decrease in the same order as the binding energy. The yields
of dibenzosemibullvalene within dry LiX, NaX and KX
follow the expected trend (Table 1). The above gas phase
computational data provide a clue to what is likely to be
responsible for the observed enhancement of triplet
products within light alkali cation exchanged zeolites.
While one might question the validity of using gas-phase
computational data to understand the chemistry that occurs
within a more complex environment, we believe that this is
a good starting point.
operates in the case of carbonyl systems.16,17 Light cations
such as Liþ and Naþ preferentially bind to the carbonyl
chromophore and by so doing alter the nature of the lowest
excited state. For example alkali ion bound acetophenone
has ppp as the lowest triplet whereas, the cation free
acetophenone has npp state as the lowest triplet. Similar
phenomenon was also established in the case of a,b-
unsaturated enones.17 No detailed calculations were per-
formed to establish that this is occurring in b,g-unsaturated
ketones. A likelihood of this possibility has been suggested
earlier.18
4. Conclusion
Heavy atom effect, light atom effect and sensitization
methods have been explored to generate the excited triplet
states of organic molecules within zeolites. Heavy atom
effect as expected worked well with two systems whose
excited states have p,pp character. Sensitization, a tech-
nique commonly employed to generate the triplet states of
organic molecules in isotropic media works well with
several systems within zeolites. The light atom effect unique
to dibenzobarrelenes, having limited potential, is mechan-
istically interesting. Our studies in zeolites continue to bring
out the importance of alkali ion-organic interactions in
controlling excited state chemistry.
We have recently established that light cation effect also

5770
K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
5. Experimental
5.4. Direct irradiation under ‘wet’ conditions
NaY and NaX zeolites were obtained from Aldrich.
Monovalent cation exchanged (Liþ, Kþ, Rbþ, Csþ and
Tlþ) zeolites were prepared by stirring 10 g of NaY or NaX
with 100 mL of a 10% aqueous solution of the
corresponding metal nitrate for 12 h with continuous
refluxing. The zeolites were filtered and washed
thoroughly with distilled water. This procedure was
repeated for three times. The cation exchanged zeolite
was then heated0 in an oven at 1208 C for about 6 h.
Acetophenone, 4 -methoxy acetophenone, xanthone and
a-aminoacetophenone hydrochloride and ferrocene were
used as obtained from Aldrich. Acenaphthylene (Aldrich)
was recrystallized thrice from ethanol prior to use.
Commercially available metal nitrates were used for
exchanging the zeolites.
For experiments where ‘wet’ zeolites were used, water was
co-adsorbed into the reactant loaded zeolite sample using
the following procedure: The dry zeolite adsorbed with the
reactant molecules was placed on a piece of weighing paper
(10 cm£10 cm) on a balance. A 50 mL beaker full of water
was placed next to the zeolite sample. After 7 mg of water
was absorbed by the zeolite (corresponding to a loading
level of 3 molecules of water per supercage), the sample was
transferred to fresh hexane, purged with nitrogen and
photolyzed as hexane slurry. Rest of the procedure was the
same as above.
5.5. Sensitization experiments
Known amount of the sensitizer and activated (4508C, air
oven) zeolite (300 mg) KY were stirred in hexane, filtered,
washed with excess hexane and dried under reduced
pressure (1024 Torr). In case of a-aminoacetophenone
hydrochloride the sensitizer was first exchanged into the
zeolite using the cation exchange procedure described under
materials. Above sensitizer loaded zeolite was used to
adsorb the reactants 1 and 2. Known amounts of the
reactants and 300 mg of the zeolite–sensitizer composite
were stirred for 6 h, filtered, washed with hexane and dried.
The loading level of the sensitizer to the substrate was
always maintained at 1:1. The dried samples were then
irradiated (450 W medium pressure mercury lamp, pyrex
vessels) as a solid (solid irradiation) or as slurry in hexane
(slurry irradiation). The hexane slurry was purged
thoroughly with nitrogen before photolysis. Irradiations
were carried out to about 30% conversion in all cases. This
conversion was achieved in 2 h with slurry irradiations, 8 h
with the solid irradiations. Rest of the procedure was the
same as above.
Dibenzobarrelenes (1a–c), benzobarrelene (5), 5-nor-
bornen-2-one (11), bicyclo[2.2.2]octenone (14), and
3-methyl-3-(cyclopent-1-enyl)butan-2-one (17) were pre-
pared following literature procedures.19 Spectral data of
photoproducts from 1a–c, 5, 8, 11, 14 and 17 prepared by
solution irradiation were compared with that of authentic
samples reported in the litearture.6,20 The photoproducts
thus prepared and identified were used to identify the peaks
in the GC traces.
5.1. Photolysis procedures
The procedures adopted for the irradiation of barrelenes,
and b,g-enones were the same. Therefore, a general
procedure is provided below.
5.2. Direct irradiation—slurry mode
The reactants (2–5 mg) were stirred with 300 mg of
activated (5008C overnight under aerated conditions)
MþY or MþX zeolite in 10 mL of hexane for about 10 h.
The zeolite was washed twice with 10 mL portions of
hexane and the combined washings were concentrated and
analyzed. Absence of the reactants in the washings was
taken to indicate that it has been included into the zeolite.
Reactant–MþY complex was irradiated, as a slurry, in 4 mL
of hexane after purging with nitrogen for 15 min. It was then
extracted with 20 mL of diethyl ether for a period of 5 h and
the concentrated ether extract was analyzed by GC
(Hewlett–Packard 5890 series II fitted with HP5 capillary
column). Absence of either the starting material or the
photoproducts in the hexane portion after irradiation,
excluded the possibility of any of them escaping the cage
during the slurry irradiation. The material balance in all
cases was .90%.
5.6. Irradiation of acenaphthylene included in zeolites
Acenaphthylene and dried zeolites (furnace 5008C over-
night) were stirred in trimethylpentane for about 10 h.
Slightly yellowish solids were collected by filtration,
washed with hexane and dried under nitrogen stream.
These were degassed under vacuum (10–4 mm) and
irradiated with two 450 W mercury lamps for 2 h. Samples
were stirred by rotation every 30 min. Acenaphthylene
dimers were extracted with ether and drops of con. HCI
stirring overnight. Samples were analyzed by GC. Samples
were also irradiated under oxygen. For this purpose, oxygen
was admitted to the degassed samples on a vacuum line.
Irradiation of acenaphthylene in presence of ferrocene:
complexes of acenaphthylene (5 mg) in hexane with Rb Y
(250 mg) was prepared by stirring in hexane for an hour.
The color due to acenaphthyelene molecules (ANY) were
not seen in solution and this observation led us to believe
that all acenaphthyelene molecules were adsorbed within
zeolites. At this stage ferrocene was added from stock
solution corresponding to 0, 2, 4, 6, 8, 10 mg and stirred.
Solid complexes were collected by usual filtration etc. and
diffuse reflectance spectra recorded. Spectra revealed the
presence of both ANY and ferroecene inside. Samples
degassed in two tubes (125 mg each) and irradiated for
5.3. Direct irradiation—solid phase
For solid irradiations, the zeolite after stirring in hexane was
filtered and washed with hexane. The solid complex thus
obtained was degassed on a vacuum line (1024 Torr) for
more than 10 h and the dry powder was irradiated by
exposing new surface every 30 min by manually rotating
the sample tube. Rest of the procedure was the same as
above.

K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
5771
3.5 h. Dimer isolated by acid/ether extraction and analyzed
by GC.
9904187 and CHE-0212042) for support of the research
summarized here. We thank J. R. Scheffer for providing
information on synthesis of substituted dibenzoberrelenes
1b and 1c and spectral data of photoproducts and L. J.
Johnston and K. J. Thomas for the time resolved quenching
data. Initial samples provided by J. R. Scheffer allowed us to
initiate the study on 1b and 1c. V. R. thanks J. R. Scheffer
and L. J. Johnston for useful discussion during the course of
the work and for their continued collaboration.
Laser flash photolysis of zeolite samples: laser flash
photolysis studies were carried out at National Research
Council of Canada, Ottawa in the laboratory of L. J.
Johnston. The experimental set up and procedure was
provided by L. J. Johnston and K. J. Thomas. The laser flash
photolysis system used for diffuse reflectance studies was
equipped with a Lumonics EX-510 excimer laser (XeCl,
308 nm, 6 ns/pulse; ,30 mJ/pulse) for sample excitation
(located at the laboratory of L. J. Johnston, National
Research Council of Canada, Ottawa). A pulsed 75-W
xenon lamp with a PTI housing and power supply was used
as the monitoring beam. The diffusely reflected analyzing
beam was collected and focused on the entrance slit
(typically 1.5 mm) of a Digikron 240 monochromator. It
is important to ensure that the specular reflections from both
the lamp and laser do not strike the collection lense. A Burle
4840 photomultiplier tube in a six-dynode stage housing
was attached to the exit slit of the monochromator. A home
built computer controlled power supply was used with the
photomultiplier. The signal from the photomultiplier was
connected via 93 ohm cable (with two 93-ohm terminators)
to a back-off circuit, which measures the light intensity
before the laser pulse and then offsets the voltage to zero.
The signal then goes to a Tektronix 7912 AD digitizer
equipped with 7A16P amplifier and 7B90P time base plug-
ins. The digitizer is connected via a GPIB interface to a PDP
11/55 computer for data storage and processing. A Stanford
Research Systems digital delay/pulse generator (DG235)
provided TTL trigger pulses to control the timing for the
laser, lamp and digitizer. Scimetrics Instruments Labmate
data acquisition and control system is used to read voltages
and provide TTL pulses to open and close shutters.
References
1. (a) Lewis, G. N.; Kasha, M. J. Am. Chem. Soc. 1944, 66, 2100.
(b) Lewis, G. N.; Kasha, M. J. Am. Chem. Soc. 1945, 67, 994.
(c) Kasha, M. Chem. Rev. 1948, 41, 401.
2. Kasha, M. J. Chem. Phys. 1952, 20, 71.
3. (a) Evans, D. J. J. Chem. Soc. 1957, 1351. (b) Evans, D. J.
J. Chem. Soc. 1959, 2753. (c) Evans, D. J. J. Chem. Soc. 1960,
1735. (d) Tsubomura, H.; Mulliken, R. S. J. Am. Chem. Soc.
1960, 82, 5966. (e) Evans, D. J. J. Chem. Soc. 1987, 1961.
(f) McClure, D. S. J. Chem. Phys. 1949, 17, 905. (g) Yuster, P.;
Weissman, S. L. J. Chem. Phys. 1949, 17, 1182. (h) McGlynn,
S. P.; Reynolds, M. J.; Daigre, G. W.; Christodoyleas, N. D.
J. Phys. Chem. 1962, 66, 2499. (i) McGlynn, S. P.; Azumi, T.;
Kasha, M. J. Chem. Phys. 1964, 40, 507. (j) Giachino, G. G.;
Kearns, D. R. J. Chem. Phys. 1970, 52, 2964. (k) Giachino,
G. G.; Kearns, D. R. J. Chem. Phys. 1970, 523, 3886.
(l) Kavarnos, G.; Cole, T.; Scribe, P.; Dalton, J. C.; Turro, N. J.
J. Am. Chem. Soc. 1971, 93, 1032. (m) Wright, M. R.; Frosch,
R. P.; Robinson, G. W. J. Chem. Phys. 1960, 33, 934.
(n) Grabowska, A. Spectrochim. Acta 1963, 19, 307.
4. (a) Ramamurthy, V.; Caspar, J. V.; Corbin, D. R.; Eaton, D. F.
J. Photochem. Photobiol. 1989, 50, 157. (b) Caspar, J. V.;
Ramamurthy, V.; Corbin, D. R. Coord. Chem. Rev. 1990, 97,
225. (c) Ramamurthy, V.; Caspar, J. V.; Eaton, D. F.; Kuo,
E. W.; Corbin, D. R. J. Am. Chem. Soc. 1992, 114, 3882.
(d) Ramamurthy, V.; Caspar, J. V.; Corbin, D. R. Tetrahedron
Lett. 1990, 31, 1097. (e) Ramamurthy, V.; Caspar, J. V.;
Corbin, D. R.; Schlyer, B. D.; Maki, A. H. J. Phys. Chem.
1990, 94, 3391. (f) Uppili, S.; Marti, V.; Nikolaus, A.; Adam,
W.; Engel, P. S.; Turro, N. J.; Ramamurthy, V. J. Am. Chem.
Soc. 2000, 122, 11025.
The zeolite samples were dried before the laser experiments.
The zeolite samples were taken in 3£7 mm2 quartz cells and
were purged with either nitrogen or oxygen immediately
prior to laser excitation. These samples were moved in the
cell holder and shaken frequently in order to provide fresh
zeolite surface and minimize complications due to product
formation. For each transient profile, five or more laser shots
were averaged. Benzobarrelene possesses no significant
absorption at 308 nm. The absorption due to the substrate
was minimal in all studies. This was confirmed by exciting
zeolite samples containing the substrate at the highest
concentration in the absence of sensitizer. Such experiments
did not yield any signal indicating that no transients are
formed by the direct excitation of the substrates.
p-Methoxyacetophenone was used as the sensitizer for the
energy transfer experiments. The zeolite used was KY for
all these experiments. All the measurements were repeated
until consistent readings were obtained. The concentration
of benzobarrelene for these studies were found to be from
0.02 and 0.2 molecules per supercage. The concentration of
sensitizer was kept at 0.02 molecules per supercage for all
quencher concentrations.
5. (a) Zimmerman, H. E. Org. Photochem. 1991, 11, 1.
(b) Zimmerman, H. E.; Armesto, D. Chem. Rev. 1996, 96,
3065. (c) Ciganek, E. J. Am. Chem. Soc. 1966, 88, 2882.
(d) Zimmerman, H. E.; Givens, R. S.; Pagni, R. M. J. Am.
Chem. Soc. 1968, 90, 6096.
6. (a) Cowan, D. O.; Drisko, R. L. Elements of Organic
Photochemistry; Plenum: New York, 1976; pp 435–443.
(b) Cowan, D. O.; Drisko, R. L. J. Am. Chem. Soc. 1970, 92,
6281. (c) Cowan, D. O.; Drisko, R. L. J. Am. Chem. Soc. 1970,
92, 6286. (d) Cowan, D. O.; Koziar, J. C. J. Am. Chem. Soc.
1975, 97, 249. (e) Koziar, J. C.; Cowan, D. O. J. Am. Chem.
Soc. 1976, 98, 1001. (f) Samanta, A.; Fessenden, R. W. J. Phys.
Chem. 1989, 93, 5823. (g) Haga, N.; Takayanagi, H.;
Tokumaru, K. J. Org. Chem. 1997, 62, 3734.
7. (a) Demuth, M. Org. Photochem. 1991, 11, 37. (b) Givens,
R. S.; Oettle, W. F.; Coffin, R. L.; Carlson, R. G. J. Am. Chem.
Soc. 1971, 93, 3957. (c) Eckersley, T. J.; Parker, S. D.; Rogers,
N. A. J. Tetrahedron 1984, 19, 3749. (d) Ipaktschi, J.
Tetrahedron Lett. 1969, 26, 2153. (e) Engel, P. S.;
Acknowledgements
Authors thank the National Science Foundation (CHE-

5772
K. Pitchumani et al. / Tetrahedron 59 (2003) 5763–5772
Schexnayder, M. A. J. Am. Chem. Soc. 1972, 94, 9252.
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.;
Replogle, E. S.; Pople, J. A. Gaussian 98, Revision A.9;
Gaussian, Inc.: Pittsburgh, PA, 1998.
(f) Schexnayder, M. A.; Engel, P. S. Tetrahedron Lett. 1975,
13, 1153. (g) Engel, P. S.; Schexnayder, M. A. J. Am. Chem.
Soc. 1975, 97, 144.
8. Pitchumani, K.; Warrier, M.; Scheffer, J. R.; Ramamurthy, V.
J. Chem. Soc., Chem. Commun. 1998, 1197.
16. Shailaja, J.; Lakshminarasimhan, P. H.; Pradhan, A.; Sunoj,
R. B.; Jockusch, S.; Karthikeyan, S.; Uppili, S.;
Chandrasekhar, J.; Turro, N. J.; Ramamurthy, V. J. Phys.
Chem. A 2003, 107, 3187.
9. Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of
Photochemistry; Marcell Dekker: New York, 1988;
pp 338–341.
17. Sunoj, R. B.; Lakshminarasimhan, P.; Ramamurthy, V.;
Chandrasekhar, J. J. Comput. Chem. 2001, 22, 1598.
18. Uppili, S.; Takagi, S.; Sunoj, R. B.; Lakshminarasimhan, P.;
Chandrasekhar, J.; Ramamurthy, V. Tetrahedron Lett. 2001,
42, 2079–2083. Note the yields of products in Scheme 1 for
KX and KY have been switche due to a typographical error.
19. (a) Cristol, S. J.; Hause, N. L. J. Am. Chem. Soc. 1952, 74,
2193. (b) Cristol, S. J.; Bly, R. K. J. Am. Chem. Soc. 1960, 82,
6155. (c) Hales, N. J.; Heaney, H.; Hollinshead, J. H. Synthesis
1975, 707. (d) Ranganathan, S.; Ranganathan, D.; Mehrotra,
A. K. Synthesis 1977, 289. (e) Engel, P. S.; Schexnayder, M. A.
J. Am. Chem. Soc. 1975, 97, 144. (f) Cristol, S. J.; Noreen,
A. L. J. Org. Chem. 1976, 41, 4016. (g) Cristol, S. J.;
Schloemer, G. C.; James, D. R.; Paquette, L. A. J. Org. Chem.
1972, 37, 3852.
10. (a) McGlynn, S. P.; Azumi, T.; Kinoshita, M. Molecular
Spectroscopy of the Triplet State; Prentice-Hall: Englewood
Cliffs, NJ, 1969; pp 246–247. (b) Lower, S. K.; El-Sayed,
M. A. Chem. Rev. 1966, 66, 199.
11. Scaiano, J. C.; de Lucas, N. C.; Andraos, J.; Garcia, H. Chem.
Phys. Lett. 1995, 233, 5.
12. Thomas, K. J.; Sunoj, R. B.; Chandrasekhar, J.; Ramamurthy,
V. Langmuir 2000, 16, 4912.
13. Scheffer, J. R.; Yang, J. In CRC Handbook of Organic
Photochemistry and Photobiology; Horspool, W. H., Song,
P.-S., Eds.; CRC: Boca Raton, 1995; pp 204–221.
14. Liao, C.-C.; Yang, P.-H. In CRC Handbook of Organic
Photochemistry and Photobiology; H, H. W., Song, P.-S.,
Eds.; CRC: Boca Raton, 1995; pp 194–203.
15. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.;
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.;
Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.;
20. (a) Griffin, G.; Peters, J. J. Org. Chem. 1963, 28, 1715.
(b) Emerson, G. F.; Watts, L.; Pettit, R. J. Am. Chem. Soc.
1965, 87, 131. (c) Ciganek, E. J. Am. Chem. Soc. 1966, 88,
2882. (d) Zimmerman, H. E.; Givens, R. S.; Pagni, R. M. J. Am.
Chem. Soc. 1968, 90, 6096. (e) Schuster, D. I.; Axelrod, M.;
Auerbach, J. Tetrahedron Lett. 1963, 27, 1911. (f) Ipaktschi, J.
Tetrahedron Lett. 1969, 26, 2153. (g) Ipaktschi, J. Chem. Ber.
1972, 105, 1840. (h) Givens, R. S.; Oettle, W. F.; Coffin, R. L.;
Carlson, R. G. J. Am. Chem. Soc. 1971, 93, 3957. (i) Cristol,
S. J.; Braun, G. C.; Schloemer, G. C.; Vanden Plas, B. J. Can.
J. Chem. 1986, 64, 1081.