Efficient photocyclization of o-alkylbenzaldehydes in the solid state: Direct observation of E-xylylenols en route to benzocyclobutenols

Article abstract of DOI:10.1021/jo015718r

The photocyclization to benzocyclobutenols of o-alkyl aromatic aldehydes that are predestined for γ-hydrogen abstraction is found to occur efficiently in the solid state; in contrast, solution-phase photolysis is known to afford a mixture of several produ

Full text of DOI:10.1021/jo015718r

J . Org. Chem. 2001, 66, 7013-7019
7013
Efficien t P h otocycliza tion of o-Alk ylben za ld eh yd es in th e Solid
Sta te: Dir ect Obser va tion of E-Xylylen ols en Rou te to
Ben zocyclobu ten ols
J . Narasimha Moorthy,*,† Prasenjit Mal,† R. Natarajan,† and P. Venugopalan*,‡
Department of Chemistry, Indian Institute of Technology, Kanpur 208 016, India, and
Department of Chemistry, Panjab University, Chandigarh 160014, India
moorthy@iitk.ac.in
Received April 27, 2001
The photocyclization to benzocyclobutenols of o-alkyl aromatic aldehydes that are predestined for
γ-hydrogen abstraction is found to occur efficiently in the solid state; in contrast, solution-phase
photolysis is known to afford a mixture of several products. It is shown that mesitaldehyde, which
is a liquid, also undergoes efficient cyclization when subjected to photolysis as a solid inclusion
complex. The marginal energy differences in the relative energies of the E-enols and the
corresponding cyclobutenols in the case of cyano-substituted mesitaldehydes has permitted direct
observation, for the first time, of the E-enols en route to benzocyclobutenols. The AM1 calculations
suggest that the cyano-substitution causes intrinsic stabilization of the E-enols relative to the
corresponding cyclobutenols, while the bromo groups do the opposite. The lack of observation of
the red color in bromo- and formyl-substituted aldehydes is attributed to rapid cyclization of the
E-enols to the their respective cyclobutenols even at low temperatures.
In tr od u ction
synthetically useful benzocyclobutenol (CB) is only a
minor product (10%). Surprisingly, the solid-state photo-
reactivity of o-alkylaromatic aldehydes has been explored
only little if any.18-20 The limited study has nonetheless
revealed fascinating photochromism based on enolization.
The photoinduced color change has been attributed to the
formation of o-xylylenol, which has been shown by
Wagner and co-workers to be the precursor of the
cyclobutenol.4 The latter, according to Ito and co-work-
ers,2,3 is obtained directly from the triplet-excited biradi-
cal () triplet enol).21 Be this mechanistic dichotomy as
it may, the formation of cyclobutenol has been found to
occur in a poor yield (ca. 10%) in a thoroughly examined
case of solid-state photolysis of an o-alkylaromatic alde-
hyde.19 Our recent observation of facile formation of
cyclobutenol in the solid-state photolysis of p-anisalde-
hydes20 prompted us to explore the solid-state photo-
behavior of o-alkylaromatic aldehydes that are predes-
tined for γ-hydrogen abstraction. Furthermore, a detailed
study deemed worthwhile in view of the formation of a
complex mixture of products in the solution-state pho-
tolysis and intriguing photochromism in the solid state.
Herein we report the efficient photocyclization of crystal-
The solution as well as solid-state photochemistry of
o-alkylphenyl ketones has been well established;1-9 cy-
clization to synthetically important benzocyclobutenols10-15
is, in general, the major pathway of photolysis. In a
striking contrast, the solution-state photochemistry of
o-alkylaryl aldehydes is frustrated by the formation of
myriad of products. For example, photolysis of o-methyl-
benzaldehyde has been reported to afford as many as
seven photoproducts (Scheme 1) depending on the condi-
tions of photolysis.16,17 It should be noted that the
* Corresponding author. Tel: 91-512-597438, Fax: 91-512-597436.
† Department of Chemistry, Indian Institute of Technology, Kanpur
208 016, India.
‡ Department of Chemistry, Panjab University, Chandigarh 160 014,
India.
(1) Sammes, P. G. Tetrahedron 1976, 32, 4050.
(2) Ito, Y.; Giri, B. P.; Nakasuji, M.; Hagiwara, T.; Matsuura, T. J .
Am. Chem. Soc. 1983, 105, 1117.
(3) Ito, Y.; Nishimura, H.; Umehara, Y.; Yamada, Y.; Tone, M.;
Matsuura, T. J . Am. Chem. Soc. 1983, 105, 1590.
(4) Wagner, P. J .; Subrahmanyam, D.; Park, B.-S. J . Am. Chem.
Soc. 1991, 113, 709.
(5) Sobczack, M.; Wagner, P. J . Tetrahedron Lett. 1998, 2523.
(6) Kawata, E.; Saito, M.; Yoshioka, M. J . Chem. Soc., Perkin Trans.
1 2000, 1015.
(7) Ito, Y.; Matsuura, T.; Fukuyamu, K. Tetrahedron Lett. 1988, 29,
3087.
(8) Ito, Y.; Yasui, S.; Yanauchi, J .; Ohba, S.; Kano, G. J . Phys. Chem.
A. 1998, 102, 5415.
(9) Ito, Y.; Kano, G.; Nakamura, N. J . Org. Chem. 1998, 63, 5643.
(10) Yang, H.; Choy, W. J . Org. Chem. 1988, 53, 5796.
(11) Macdonald, D. I.; Durst, T. J . Org. Chem. 1988, 33, 3663.
(12) Zhang, X.; Foote, C. S. J . Org. Chem. 1994, 59, 5235.
(13) Fitzgerald, J . J .; Pagano, A. R.; Sakoda, V.; Olofson, R. A. J .
Org. Chem. 1994, 59, 4121.
(16) Findlay, D. M.; Tchir, M. F. J . Chem. Soc., Chem. Commun.
1974, 514.
(17) Arnold, B. J .; Mellows, S. M.; Sammes, P. G.; Wallace, T. W. J .
Chem. Soc., Perkin Trans. 1 1974, 401.
(18) Kumar, V. A.; Venkatesan, K. J . Chem. Soc., Perkin Trans. 2
1991, 829.
(19) Sarkar, T. K.; Ghosh, S. K.; Moorthy, J . N.; Fang, J .-M.; Nandi,
S. K.; Sathyamurthy, N.; Chakraborthy, D. Tetrahedron Lett. 2000,
41, 6909.
(14) Olofson, R. A.; Michael, F. E.; Fitzgerald, J . J . Tetrahedron Lett.
1994, 39191.
(15) Henteman, M. F.; Allen, J . G.; Danishefsky, S. J . Angew. Chem.,
Int. Ed. 2000, 39, 11.
(20) Moorthy, J . N.; Mal, P.; Natarajan, R.; Venugopalan, P. Org.
Lett. 2001, 3, 1579.
(21) Das, P. K.; Encinas, M. V.; Small, R. D., J r.; Scaiano, J . C. J .
Am. Chem. Soc. 1979, 101, 6965.
10.1021/jo015718r CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/15/2001

7014 J . Org. Chem., Vol. 66, No. 21, 2001
Moorthy et al.
Sch em e 1
hexadiyne-1,6-diol, was synthesized following the re-
ported procedure.23
Solu tion -Sta te P h otolysis. The solution-state photo-
chemistry of the parent mesitaldehyde 1a has been
known for a long time;24 accordingly, the photolysis of
ca. 0.03 M solution of 1a in 2-propanol yields the
cyclobutenol and the mesitoic acid in 33 and 14% yields,
respectively (eq 1). We have similarly tried to establish
line mesitaldehydes 1-3 and present evidence for direct
observation of (E)-o-xylylenol on the way to cyclobutenol
formation.
Sch em e 2
the solution-state photochemistry of the aldehydes
1-3. As representative cases, the benzene solutions (ca.
5 mM) of the aldehydes 1c, 1d , 2a -c, and 3 were
subjected to irradiation in a Rayonet reactor (λ ) 350
nm) under oxygen-excluded conditions. After irradiation
for 24 h, the only isolable product in the case of 1c was
found to be the phthalide 4 in 25% yield (eq 2);25 the
Resu lts
Syn th esis of th e Ald eh yd es 1-3. The mesitaldehyde
1b was synthesized by the reduction of 4-bromo-2-
cyanomesitylene using DIBAL (Scheme 2). The mesit-
aldehydes 1c and 2a were available together from the
reduction of 2,4-dicyanomesitylene in the ratio of 1:1.08
(77% yield), respectively. The dibromomesitaldehyde 1d
was prepared by bromination of mesitaldehyde 1a using
NBS in TFA-H2SO4, a recently reported procedure.22 The
reduction of tricyanomesitylene using DIBAL afforded
the aldehydes 1e and 2c (86%) in a ratio of 1:1.25.
Similarly, the reduction of 2,4-dicyano-6-bromomesity-
lene yielded the aldehydes 1f and 2b (78%, 1:1.48). The
trimethyltrimesaldehyde 3 was prepared by reducing
tricyanomesitylene using large excess of DIBAL. The
lattice inclusion host diol, viz., 1,1,6,6-tetraphenyl-2,4-
remainder was a polar polymeric material. The TLC
analysis of the photolysate from 1d indicated the forma-
tion of several products and the cyclobutenol could not
be readily identified. A similar scenario was found with
(23) Toda, F.; Tokumaru, Y. Chem. Lett. 1990, 987.
(24) Matsuura, T.; Kitaura, Y. Tetrahedron 1969, 25, 4487.
(25) The phthalide has been shown to derive from trapping of the
photoenol by the adventitious oxygen present during photolysis, see:
Wasserman, H. H.; Mariano, P. S.; Keehn, P. M. J . Org. Chem. 1971,
36, 1765.
(22) Duan, J .; Zhang, L. H.; Dolbier, W. J ., J r. Synlett. 1999, 1245.

Solid-State Photocyclization of o-Alkylbenzaldehydes
J . Org. Chem., Vol. 66, No. 21, 2001 7015
Ta ble 1. Resu lts of Solid -Sta te P h otocycliza tion of
Ald eh yd es 1-3a
entry
substrate
temp (°C)
conv (%)b
yield (%)c
1
2
1a
1b
30
0e
40
33
40
91
100
86
42
100
75
18
72
100d
92
82
30e
30
3
4
5
6
7
1c
1d
68
90
30
30e
-80e
30
100d
100d
90
1e
8
9
30e
-80e
30
100d
100d
89
10
11
12
13
14
1f
2a
2b
2c
3
30
30
30
30
92
66
74
79
56f
21f
25f
24f
F igu r e 1. UV-vis absorption spectra of 1e before (dotted line)
and after (solid line) irradiation (λ g 300 nm).
Two regioisomeric cyclobutenols are in principle pos-
sible for the aldehydes 1b, 1c, 1f, and 2a -c depending
on the orientation of the formyl group in the crystal
lattice (eq 3). In all of these cases, mixtures of the two
possible regioisomeric cyclobutenols were isolated. No
effort was made to separate the isomers and quantify
their relative ratios.
a
Unless otherwise mentioned, all the photolyses were conducted
in a Rayonet reactor (λ ) 350 nm) for 24 h by dispersing the
powdered samples in sealed test tubes purged with a stream of
nitrogen gas. The samples were stirred periodically to ensure
uniform exposure. Based on recovered starting material. c Iso-
b
lated yields based on conversion, unless otherwise mentioned.
From 1H NMR analysis, error ( 10%. e High-pressure Hg-lamp
d
(λ g 300 nm), duration ) 2 h. f The remainder was an intractable
polymeric material.
the photolysis of 2a and 3. When 2b was photolyzed until
its disappearance (TLC analysis), a highly intractable
polymeric material was obtained. However, the cyclo-
butenol (2b-CB) was isolated in 53% yield at 30%
conversion. Similarly, 2c afforded the cyclobutenol 2c-
CB in 21% yield at 80% conversion. Clearly, the solution-
state photochemistry of o-alkylbenzaldehydes leads to a
complex mixture of several products.
Solid State P h otoch em istr y of th e In clu sion Com -
p lex betw een Mesita ld eh yd e 1a a n d th e Host Diol.
As the parent mesiatldehyde 1a is a liquid, we formed
its crystalline inclusion complex with the well-known host
diol.26 The stoichiometry of the complex was established
to be 1:2 (host:guest) from 1H NMR analysis and un-
equivocally from X-ray crystal structure determination
(vide infra). The photolysis of the gently ground crystals
of the inclusion complex, dispersed in a Pyrex container
purged with N2 gas for 15 min, was conducted in a
Rayonet reactor (λ ) 350 nm). From 1H NMR analysis
of the irradiated crystals, the cyclobutenol was estab-
lished to be the sole product at 40% conversion (entry 1,
Table 1).
All the cyano-substituted aldehydes 1c, 1e, and 1f
turned orange-red color upon exposure to UV-irradiation.
The color persisted as long as the samples were exposed,
but disappeared immediately (ca. 2 min at 25-30 °C)
when protected from light at room temperature. A similar
behavior was exhibited by the isophthaldehydes 2a and
2c. No color was observed in the case of 1a -diol complex,
1b, 1d , and 2b. The trimesaldehyde 3 gave rise to yellow
color. Figure 1 shows the absorption spectra typically
recorded for the case of 1e before and after exposure to
UV-radiation from a high-pressure Hg-lamp. On the basis
of the documented kinetic behavior of Z- and E-enols,1,27,28
we attribute the absorption in the visible region between
400 and 600 nm to the E-enol (eq 4);29 the Z-enols are
known to be considerably unstable and undergo rapid
reversion to the aldehyde.1,27
Solid -Sta te P h otolysis of th e Ald eh yd es 1b-f, 2,
a n d 3. The solid-state photolysis of all of these aldehydes
was carried out in the same manner as described above.
In preparative photolyses, ca. 100 mg of the solid sample
was irradiated and subjected to column chromatography.
The results of solid-state photolyses are given in Table
1. The aldehydes 1b-f clearly afforded the cyclobutenols
in good yields (entries 2-10). A notable difference in the
photobehavior was observed when isophthalaldehydes 2
and trimesaldehyde 3 were subjected to solid-state pho-
tolysis. In all these cases, in addition to the cyclobutenol,
a highly polar polymeric material was invariably formed
in significant amounts. As may be seen from the results
in Table 1, the yields of cyclobutenols from 2b, 2c, and 3
(entries 12-14) are significantly low, suggesting that the
formation of intractable material increases with formyl
substitution.
The two aldehydes 1d and 1e, differing only in terms
of their substituents, viz., bromo and cyano, exhibit quite
contrasting behavior in that the latter turns red, while
the former remains colorless upon irradiation. We have
simultaneously irradiated these two aldehydes at low
(-80 °C) and at room (30 °C) temperatures for 2 h using
the high-pressure Hg source (λ > 300 nm). Both 1d and
1e were found to exhibit notable temperature depend-
(27) Haag, R.; Wirz, J .; Wagner, P. J . Helv. Chim. Acta 1977, 60,
2595.
(28) Gebicki, J .; Krantz, A. J . Chem. Soc., Perkin Trans. 2 1984,
1623.
(29) For UV-vis absorption spectrum of the E-enol of o-methyl-
benzaldehydes, see: refs 20 and 28.
(26) Tanaka, K.; Toda, F. Chem. Rev. 2000, 100, 1025.

7016 J . Org. Chem., Vol. 66, No. 21, 2001
Moorthy et al.
F igu r e 2. The crystal packing diagram of the complex of
mesitaldehyde 1a with the host diol. The hydrogen bonding
between the guest carbonyl oxygen and the host diol are shown
with a broken line.
F igu r e 3. The molecular packing diagram of 1e. The C-H‚‚‚O
bonds are shown with a bold broken line and those of C-H‚‚‚N
with a light broken line.
ence; while the yield of cyclobutenol was higher (entries
5 and 8) at room temperature (ca. 30 °C), it considerably
reduced at low temperature (entries 6 and 9).30 Notice-
ably, the temperature effect is more pronounced in the
case of 1e.
X-r a y Cr ysta l Str u ctu r e An a lyses a n d Str u ctu r e-
Rea ctivity Cor r ela tion s. To establish (i) the complex
formation between the lattice inclusion host diol and
mesitaldehyde 1a and (ii) to correlate the solid-state
photoreactivity of mesitaldehydes with their crystal
structures, we sought to examine the X-ray crystal
structures of the complex of 1a with the diol, 1d , and
1e. The aldehydes 1d and 1e were considered in par-
ticular, since both these aldehydes undergo efficient
cyclization with only 1e undergoing color change to red
during irradiation. We surmised that the apparent dif-
ferences in their photobehavior could be rationalized from
the crystal packing analyses. While the crystals of 1a -
diol complex and 1e could be grown readily, our efforts
to obtain crystals of 1d suitable for X-ray studies were
not successful.
Figure 2 shows the perspective view of molecular
packing diagram of the complex between 1a and the diol.
It can be readily discerned that the hydroxy groups at
the termini of the linear chain of the diol act as anchors
to bind the two guest mesitaldehyde molecules through
strong O-H‚‚‚O hydrogen bonds (dO‚‚‚H ) 1.96 Å, θO-H‚‚‚O
) 169.7°). Figure 3 shows the molecular packing diagram
of 1e. One observes that the weaker C-H‚‚‚O and
C-H‚‚‚N intermolecular interactions operate in concert
in controlling the crystal packing. The carbonyl oxygen
participates in C-H‚‚‚O intermolecular hydrogen bond-
ing with two hydrogens of the methyl groups of different
neighbors. Interestingly, both nitrogen atoms of the CtN
groups hydrogen bond to two different hydrogens of the
neighboring molecules.
F igu r e 4. Geometrical parameters for γ-hydrogen abstraction.
Ta ble 2. Sch effer ’s Geom etr ica l P a r a m eter s for
γ-Hyd r ogen Abstr a ction
compound
d (Å)
θ (deg)
∆0 (deg)
τo (deg)
1a -diol
1e
2.55
2.43
94.4
112.6
97.8
94.8
11.6
10.3
From their extensive studies on photochemical γ-hy-
drogen abstraction of ketones in solid state, Scheffer and
co-workers have set forth certain geometrical parameters,
viz., d, ∆o, θ, and τo, which quantitatively describe the
feasibility of H-abstraction.31 The abstraction is said to
be at its best when d ) 2.3-3.0 Å, ∆o ) 90-120°, θ )
180°, and τo ) 0° (Figure 4). The values calculated from
X-ray structural parameters for abstraction of closest
methyl hydrogens for 1a -diol complex and for 1e are
shown in Table 2. These values deviate significantly from
the ideal ones, which are seldom realized. Nonetheless,
they fall in the range generally observed for a variety of
substrates that undergo Norrish Type II reaction in the
solid-state.31 Thus, the γ-hydrogen abstraction, the pri-
mary event, is facile in both cases to yield the triplet-
excited 1,4-biradical/enol.
AM1 Ca lcu la tion s. Semiempirical AM1 calculations
have been shown to yield results in good agreement with
the experiment for the o-xylylenol to benzocyclobutenol
transformation.32 We have carried out these calculations
(QCMP 137, MOPAC/PC)33 for the aldehydes 1a , 1d , 1e,
and 3 to assess the relative stabilities of cyclobutenol and
E-enol in these cases. In Scheme 3 are shown the
(30) Although the photolyses were carried out at identical conditions,
a comparison of the temperature-dependent conversions of solid-state
reactions is subject to a variety of factors such as homogeneous
dispersion of the sample, uniform light intensity, etc. Since a similar
trend in the yields was observed in three independent experiments,
we believe that the reaction is indeed temperature-dependent.
(31) Ihmels, H.; Scheffer, J . R. Tetrahedron 1999, 55, 885.
(32) Coll, G.; Costa, A.; Deya, P. M.; Flexas, F.; Rotger, C.; Saa, J .
M. J . Org. Chem. 1992, 57, 6222.
(33) Dewar, M. J . S.; Zeobish, E. G.; Healy, E. F.; Stewart, J . J . P.
J . Am. Chem. Soc. 1985, 107, 3092.

Solid-State Photocyclization of o-Alkylbenzaldehydes
J . Org. Chem., Vol. 66, No. 21, 2001 7017
Sch em e 3. En th a lp ies of Rea ction
volves abstraction of a γ-hydrogen by the triplet-excited
carbonyl to afford 1,4-radical, which is considered equiva-
lent to the triplet-excited enol.21 This species may decay
either to the E-enol or Z-enol. The latter is known to be
short-lived due to a faster reverse hydrogen transfer
process (thermal 1,5-sigmatropic shift) that regenerates
the precursor aldehyde.1,27 On the other hand, the E-enols
are longer-lived and all the bimolecular trapping chem-
istry is in agreement with its stereochemsitry.1 Wagner
and co-workers have convincingly shown that the benzo-
cyclobutenols indeed derive from thermal conrotatory
closure of the E-enols.4 In contrary, Ito and co-workers
believe that the cyclobutenols need not necessarily obtain
from E-enols, but can derive from direct collapse of the
1,4-triplet biradical/enol.2,3 Indeed, Coll et al. have also
found no evidence for involvement of enols in the forma-
tion of cyclobutenols from methoxy-substituted benz-
aldehydes.32 Thus, the mechanistic dichotomy with re-
gard to the formation of cyclobutenols continues to be
intriguing.
enthalpies of reaction. The AM1 results predict that the
dicyano-substitution (1e) stabilizes the E-enol relative to
the cyclobutenol (∆Hf ) 1.16 kcal/mol vs 1.99 kcal/mol
for mesitaldehyde 1a ), while the dibromo-substitution
(1d ) causes the opposite effect (∆Hf ) 3.79 kcal/mol). The
latter is true even with diformyl substitution (3, ∆Hf )
2.79 kcal/mol).
Discu ssion
The Scheffer’s geometrical parameters for 1a -diol
complex and for 1e suggest that the primary γ-hydrogen
abstraction in their crystal lattices is facile to afford the
respective triplet 1,4-biradicals, whose structures have
been shown from theoretical calculations to be twisted
at one end.34 The observation of red-color, attributed to
E-enols, in all the cyano-substituted aldehydes 1c, 1e,
1f, and 2c, and efficient formation of cyclobutenols clearly
suggest that pathway b (Scheme 4) is operative in the
cyclizations. Indeed, the temperature dependence of the
cyclization is clearly borne out from the comparison of
the results of photolysis for 1d and 1e at low and high
temperatures under otherwise identical conditions of
irradiation. The yield of cyclobutenol is considerably low
for 1e at low temperature (entry 9, Table 1), but
significantly high at room temperature (entry 8). A
similar but less pronounced effect is observed for 1d
(entries 5 and 6). Therefore, the lack of observation of
E-enol (red color) cannot be taken to imply that path b
(Scheme 4) is not operative in this aldehyde, i.e., 1d , and
other aldehydes 1a ,b, 2a ,b, and 3 that do not turn red
upon irradiation. After all, why the similar class of
compounds differing only in terms of substituents should
follow two different pathways cannot be easily reconciled.
As noted at the outset, the solution-state photolysis of
o-alkylaryl aldehydes leads to a complex mixture of
products with a highly intractable polymeric material
often formed in significant yields.16,17,24,32 The reported
results of photolysis for 1a 24 and those for 1c and 1d
(present study) in benzene solutions (5 mM, λ ) 350 nm)
further emphasize the complexity with solution phase
photolysis. In contrast, the results of solid-state photoly-
sis of aldehydes 1 are remarkable as can be seen from
results given in Table 1. While the yields of benzocyclo-
butenols are good to excellent for aldehydes 1, one begins
to observe a complex mixture of products with consider-
able material loss leading to low yields of cyclobutenols
in the case of aldehydes 2 and 3. Nonetheless, the results
for aldehydes 1 suggest that the solid-state photolysis is
a viable alternative to access benzocyclobutenols.
The results of photolysis of inclusion complex of 1a
with the host diol are significant in that the utility of
lattice inclusion compounds for isolating the liquid
samples as crystalline materials and regulating the
photochemistry of included guest molecules is further
demonstrated. Indeed, the inclusion complex formation
with a chiral host is an excellent strategy for asymmetric
synthesis and optical resolution.26
We shall now consider the mechanism with which the
benzocyclobutenols may be formed in the crystal lattice.
Scheme 4 is an exemplary of events that occur following
photoexcitation of the aldehydes. The mechanism in-
To answer the question as to why some aldehydes
exhibit red color due to E-enols and some others not, we
undertook AM1 theoretical calculations. The AM1 en-
thalpies of the reaction in Scheme 3 for the aldehydes
Sch em e 4

7018 J . Org. Chem., Vol. 66, No. 21, 2001
Moorthy et al.
conducted on silica gel (particle size: 60-120 µm; Acme
Synthetic Chemicals, India).
1a ,d ,e and 3 show that cyano-substitution causes intrin-
sic stabilization of the E-enols relative to the correspond-
ing cyclobutenols, while the bromo and formyl groups do
the opposite. Whether or not these gas-phase results are
applicable to the condensed phase is questionable. None-
theless, it is reasonable to assume that a similar trend
in the relative energies of the enols and cyclobutenols
operates in the solid-state such that the exothermicity
associated with cyclization is higher for all aldehydes
without cyano substituents.35 The rapid cyclization thus
may preclude the E-enols from being observed even at
low temperatures. The observation of red color in all the
cyano-substituted aldehydes can similarly be reasoned
based on a marginal difference in the energies of E-enols
and the corresponding cyclobutenols in the crystal lattice.
A closer inspection of the molecular packing diagram for
1e reveals that the hydroxyl group of the E-enol (formed
in the crystal lattice) can find itself in the close proximity
of cyano groups, such that that the E-enol may be
stabilized through O-H‚‚‚N intermolecular hydrogen
bonds involving the nitrogen atoms of the cyano group.
Presumably, a composite of the effects due to crystal
lattice stabilization and cyano-substitution (of enols)
cause the energy difference between the cyclobutenols
and the corresponding enols marginal. Otherwise, the red
color due to E-enols would not be observed.
In conclusion, it is shown that the synthetically im-
portant benzocyclobutenols are efficiently formed in the
solid-state photolysis of mesitaldehydes that are predes-
tined for γ-hydrogen abstraction. Thus, the solid-state
photolysis is a convenient way to access benzocyclo-
butenols in contrast to the solution-phase photolysis of
o-alkylbenzaldehydes, which leads to notoriously complex
mixtures. The strategy based on the formation of inclu-
sion complex for photolysis in the solid state of liquid
samples is further demonstrated using the diol host. The
marginal energy differences in the relative energies of
the E-enols and the respective cyclobutenols in the case
of cyano-substituted mesitaldehydes has permitted direct
observation, for the first time, of the E-enols en route to
benzocyclobutenols. The AM1 calculations suggest that
the cyano-substitution causes intrinsic stabilization of the
E-enols relative to the corresponding cyclobutenols, while
the bromo groups do the opposite. The lack of observation
of the red color in bromo- and formyl-substituted alde-
hydes is attributed to rapid cyclization of the E-enols to
their respective cyclobutenols even at low temperatures.
Syn th eses. 2-Bromomesitylene, 2,4-dibromomesitylene, 2,4,6-
tribromomesitylene were prepared according to the reported
procedures.36 The cyanomesitylenes, viz., 2-cyano-4-bromo-
mesitylene, 6-bromo-2,4-dicyanomesitylene, 2,4-dicyanomesi-
tylene, and 2,4,6-tricyanomesitylene were synthesized from the
corresponding bromomesitylenes following the reported pro-
cedure using CuCN in DMF.37 The host diol was available from
the known literature procedure.23
The aldehyde 1b was synthesized by reducing 4-bromo-2-
cyanomesitylene using DIBAL in CH2Cl2. The mesitaldehydes
1c and 2a were available together in 1:1.08 ratio from the
reduction of 2,4-dicyanomesitylene using DIBAL in 77% total
yield. The aldehydes 1e and 2c were obtained from the
reduction of tricyanomesitylene (86%, 1e:2c ) 1:1.25). Simi-
larly, the reduction of 6-bromo-2,4-dicyanomesitylene with
DIBAL yielded 1f and 2b (1:1.48) in an overall yield of 78%.
The aldehyde 1d was prepared by bromination of mesitalde-
hyde according to the recently reported procedure using NBS
in TFA-H2SO4.22 The trimethyltrimesaldehyde 3 was prepared
from reduction of 2,4,6-tricyanomesitylene using DIBAL.
Syn th esis of 1d . To a solution of mesitaldehyde 1a (0.74
g, 6 mmol) in trifluoroacetic acid (2.6 mL) was added 5.4 mL
of concd H2SO4. To this reaction mixture was added 1.5 g of
N-bromosuccinimide, and the reaction mixture was stirred at
room temperature for 18 h. After this period, the contents were
poured crushed ice and the solution was made alkaline with
2 N NaOH. The organic matter was extracted with dichloro-
methane (30 mL × 2). The combined extracts were washed
with water, dried over anhyd Na2SO4, filtered, and removed
in vacuo. The residue was subjected to silica gel column
chromatography (10% EtOAc-petroleum ether) to obtain 1.35
g (88%) of 1d as colorless crystals,38 mp 160 °C (dec); IR (KBr)
cm-1 2923, 1698, 1444, 1378, 1557, 1156, 1104; 1H NMR
(CDCl3, 400 MHz) δ 2.49 (s, 6H), 2.67 (s, 3H), 10.40 (s, 1H);
13C NMR (CDCl3, 100 MHz) δ 20.3, 26.8, 125.0, 127.6, 137.0,
137.6, 194.0.
Gen er a l P r oced u r e for t h e R ed u ct ion of Cya n o-
Su bstitu ted Mesita d eh yd es. A 20 mL solution of dicyano-
mesitaldehyde (2.5 mmol) in dichloromethane was taken in a
two-necked round-bottom flask under a nitrogen gas atmo-
sphere and cooled to 0 °C in an ice bath. After 10 min, 6.0 mL
(6.0 mmol) of DIBAL in toluene was slowly introduced into
the reaction flask. The reaction mixture was gradually allowed
to attain the room temperature. After stirring for 12 h, it was
quenched with dil HCl, and the contents were heated at reflux
for 30 min. Subsequently, the reaction mixture was extracted
with dichloromethane (25 mL × 3), dried over anhyd Na2SO4,
and filtered and solvent removed in vacuo. The crude product
was submitted to silica gel column chromatography.
1b: Colorless crystalline powder; mp 81-83 °C; IR (KBr)
cm-1 2924, 1743, 1584, 1455, 1376; 1H NMR (CDCl3, 400 MHz)
δ 2.35 (s, 3H), 2.42 (s, 3H), 2.61 (s, 3H), 6.91 (s, 1H), 10.44 (s,
1H); 13C NMR (CDCl3, 100 MHz) δ 19.6, 20.1, 24.7, 127.2,
129.9, 132.2, 139.2, 140.3, 143.7, 193.2. Anal. Calcd for C10H11
-
Exp er im en ta l Section
BrO (mol wt 227.10) C, 51.78; H, 4.34. Found: C, 51.31; H,
4.97.
Gen er a l Asp ects. UV-vis absorption spectra were run on
Shimadzu spectrophotometer (UV 160 A) and Infrared spectra
on Bruker Vector 22 FT-IR spectrophotometer. NMR spectra
were run on J EOL 400 MHz NMR Spectrophotometer with
deuterated solvents as internal standards. The elemental
analyses were performed on Thermoquest CE (EA 1110)
instrument. The melting points were determined with Perfit
Melting Point Apparatus (Perfit India) and are uncorrected.
The solvents were used as received and purified by standard
procedures where necessary. The commercial chemicals from
Lancaster (UK), S.D. Fine Chemicals (India), DIBAL (Lan-
caster) were used as purchased. Column chromatography was
1c: 86%, colorless powder, mp 102-104 °C; IR (KBr) cm-1
2963, 2924, 2218, 1688, 1592, 1552, 1434, 1379, 1281, 1070;
1H NMR (CDCl3, 400 MHz) δ 2.48 (s, 3H), 2.54 (s, 3H), 2.74
(s, 3H), 7.00 (s, 1H), 10.48 (s, 1H); 13C NMR (CDCl3, 100 MHz)
δ 18.3, 20.8, 21.2, 113.7, 116.5, 131.1, 144.8, 145.4, 146.8, 191.8.
1e: Colorless powder, mp 124-126 °C; IR (KBr) cm-1 2923,
1
1697, 1560, 1446, 1383, 1261; H NMR (CDCl3, 400 MHz) δ
2.75 (s, 3H), 2.78 (s, 6H), 10.46 (s, 1H); 13C NMR (CDCl3, 100
MHz) δ 18.9, 29.7, 115.0, 148.0, 190.7.
1f: Colorless solid, mp 165 °C (dec); IR (KBr pellet) cm-1
2924, 2222, 1697, 1559, 1210, 1022; 1H NMR (CDCl3, 400 MHz)
(34) Wagner, P. J .; Sobczak, M.; Park, B.-S. J . Am. Chem. Soc. 1998,
120, 2488.
(35) It is relevant to note that 2,6-dialkyl substitution is known to
enhance photocyclization of phenyl alkyl ketones, see: Kitaura, Y.;
Matsuura, T. Tetrahedron 1971, 27, 1597.
(36) Varma, P. S.; Subrahmanian, T. S. J . Ind. Chem. Soc. 1936,
13, 192.
(37) Weis, C. D. J . Org. Chem. 1962, 27, 2964.
(38) Banfi, S.; Montanari, F.; Quici, S. Recl. Trav. Chim. Pays-Bas
1990, 109, 117.

Solid-State Photocyclization of o-Alkylbenzaldehydes
J . Org. Chem., Vol. 66, No. 21, 2001 7019
δ 2.63 (s, 3H), 2.64 (s, 3H), 2.66 (s, 3H), 10.45 (s, 1H); 13C NMR
(CDCl3, 100 MHz) δ 18.2, 20.6, 24.0, 114.8, 116.2, 128.0, 133.6,
142.0, 144.3, 146.2, 192.2.
2a : Colorless solid, mp 81-82 °C; IR (KBr) cm-1 2922, 1697,
1680, 1581, 1377, 1091; 1H NMR (CDCl3, 400 MHz) δ 2.50 (s,
6H), 2.72 (s, 3H), 6.93 (s, 1H), 10.55 (s, 1H); 13C NMR (CDCl3,
100 MHz), δ 20.9, 15.3, 132.5, 133.1, 143.0, 145.1, 193.4. Anal.
Calcd for C11H12O2 (mol wt 176.22) C, 74.98; H, 6.86. Found:
C, 74.63; H, 6.71.
2b: Colorless powder, mp 150 °C (dec); IR (KBr) cm-1 2924,
2853,1695, 1555, 1442, 1261, 1204, 1067; 1H NMR (CDCl3, 400
MHz) δ 2.56 (s, 6H), 2.49 (s, 3H), 10.48 (s, 1H); 13C NMR
(CDCl3, 100 MHz) δ 15.5, 20.9, 128.8, 129.6, 135.1, 138.1,
142.7, 194.0. Anal. Calcd for C11H11BrO2 (mol wt 255.11) C,
51.79; H, 4.34. Found: C, 52.22; H, 4.55.
2c: Colorless powder, mp 155-160 °C (dec); IR (KBr) cm-1
2923, 2853, 2223, 1695, 1557, 1382, 1072; 1H NMR (CDCl3,
400 MHz) δ 2.68 (s, 3H), 2.72 (s, 6H), 10.52 (s, 2H); 13C NMR
(CDCl3, 100 MHz) δ 16.0, 18.9, 118.0, 133.9, 145.0, 147.0,
192.4.
1d -CB: IR (KBr) cm-1 3286, 2923, 1439, 1140; 1H NMR
(CDCl3, 400 MHz) δ 2.23 (s, 3H), 2.51 (s, 3H), 2.81 (d, 1H, J 1
) 14.4 Hz), 3.40 (dd, 1H, J 1 ) 14.7 Hz, J 2 ) 4.4 Hz), 5.16 (d,
1H, J ) 3.7 Hz); 13C NMR (CDCl3, 100 MHz) δ 18.2, 24.1, 41.9,
68.7, 116.5, 127.0, 133.2, 138.2, 141.2, 144.6.
1e-CB: IR (KBr) cm-1 3469, 2924, 2228, 1604, 1382, 1131;
1H NMR (CDCl3, 400 MHz) δ 2.46 (s, 3H), 2.67 (s, 3H), 3.12
(dd, 1H, J 1) 15.9 Hz, J 2 ) 1.9 Hz), 3.67 (dd, 1H, J 1 ) 15.9 Hz,
J 2 ) 4.4 Hz), 5.29 (d, 1H); 13C NMR (CDCl3, 100 MHz) δ 16.2,
20.3, 42.1, 69.4, 107.3, 114.1, 114.5, 115.6, 142.3, 145.2, 147.7,
150.4.
1f-CB: IR (KBr) cm-1 3415, 2923, 2223, 1597, 1418, 1380;
1H NMR (CDCl3, 400 MHz) δ 2.36 (s, 3H), 2.54 (s, 3H), 3.48
(dd, 1H, J 1 ) 15.9 Hz, J 2 ) 4.4 Hz), 2.93 (dd, 1H, J 1 ) 15.9
Hz, J 2 ) 2.0 Hz), 5.22 (d, 1H, d ) 22.5 Hz); 13C NMR (CDCl3,
100 MHz) δ 15.5, 21.6, 42.5, 68.2, 114.0, 116.8, 127.1, 136.5,
143.0, 144.8, 147.8.
2a -CB: IR (neat film) cm-1 3400, 2925, 1690, 1614, 1265,
1
1127; H NMR (CDCl3, 400 MHz) δ 2.24 (s, 3H), 2.54 (s, 3H),
3.08 (d, 1H, J ) 15.4 Hz), 3.72 (dd, 1H, J 1 ) 15.4 Hz, J 2 ) 3.2
Hz), 5.26 (d, 1H), 10.19 (s, 1H); 13C NMR (CDCl3, 100 MHz) δ
17.0, 18.1, 41.9, 70.7, 116.3, 127.8, 131.9, 139.8, 144.5, 146.0,
190.2.
2b-CB: IR (KBr) cm-1 3376, 1680, 1601, 1375, 1262, 1217,
1173, 1123; 1H NMR (CDCl3, 400 MHz) δ 2.32 (s, 3H), 2.67 (s,
3H), 3.06 (dd, 1H, J 1) 15.5 Hz, J 2 ) 2.0 Hz), 3.68 (dd, 1H, J 1
) 15.6 Hz, J 2 ) 4.4 Hz), 5.29 (d, 1H), 10.19 (s, 1H); 13C NMR
(CDCl3, 100 MHz) δ 19.1, 19.5, 42.1, 71.2, 118.6, 129.6, 128.8,
140.6, 141.3, 145.2, 190.1.
2c-CB: IR (KBr) cm-1 3450, 2924, 2222, 1686, 1603, 1379,
1261, 1131; 1H NMR (CDCl3, 400 MHz) δ 2.45 (s, 3H), 2.79 (s,
3H), 3.19 (dd, 1H, J 1) 16.1 Hz, J 2 ) 4.4 Hz), 3.78 (dd, 1H, J 1
) 16.2 Hz, J 2 ) 4.4 Hz), 5.31 (d, 1H), 10.22 (s, 1H); 13C NMR
(CDCl3, 100 MHz) δ 16.4, 17.9, 42.7, 70.5, 113.4, 116.3, 128.5,
143.2, 145.3, 145.9, 150.1, 188.9.
3-CB: IR (KBr) cm-1 3406, 2924, 1672, 1377, 1261, 1098;
1H NMR (CDCl3, 400 MHz) δ 2.48 (s, 3H), 2.81 (s, 3H), 3.15
(dd, 1H, J 1 ) 16.1 Hz, J 2 ) 2.0 Hz), 3.75 (dd, 1H, J 1) 16.1 Hz,
J 2 ) 2.7 Hz), 5.33 (d, 1H, J ) 2.7 Hz), 10.31 (s, 1H), 10.55 (s,
1H).
3: Colorless solid, mp 170 °C (dec); IR (KBr) cm-1 2885, 1689,
1557, 1418, 1387, 1069; 1H NMR (CDCl3, 400 MHz) δ 2.58 (s,
9H), 10.56 (s, 3H); 13C NMR (CDCl3, 100 MHz) δ 16.2, 134.9,
143.1, 194.2. Anal. Calcd for C12H12O3 (mol wt 204.23) C, 70.57;
H, 5.92. Found: C, 70.94; H, 5.73.
In clu sion Com p lex F or m a tion betw een Mesita ld e-
h yd e 1a a n d th e Host Diol. The crystals of the inclusion
complex between 1a and diol were easily grown by slow
evaporation of their solution (2.5:1 equiv) in CH2Cl2-C6H6 (1:
1) at room temperature. The stoichiometry of the complex was
established to be 1:2 (host:guest) from integrations of the
characterstic signals of the host and guest in the 1H NMR
spectrum.
Solu tion -Sta te P h otolysis. The benzene solutions (100
mL, ca. 5 mM) of aldehydes 1c,d , 2a -c, and 3 were purged
with a stream of N2 gas for 15 min and irradiated in the
Rayonet reactor (λ ) 350 nm) for 24 h. The photolysates were
periodically monitored by TLC. In all the cases, excepting 1c,
the TLC analysis indicated the formation of several products
and the cyclobutenols were not readily identifiable. In the case
of 1c, the formation of a single product was clearly observed.
Silica gel chromatography (10% EtOAc/petroleum ether) of the
photolysate afforded a colorless crystalline product in 25%
yield, which was subsequently characterized from 1H and 13C
NMR analyses as the phthalide 4.25 1H NMR (CDCl3, 400 MHz)
δ 2.62 (s, 3H), 2.70 (s, 3H), 5.33 (s, 2H), 7.24 (s, 1H); 13C NMR
(CDCl3, 100 MHz) δ 17.6, 20.7, 16.8, 104.6, 114.5, 122.5, 132.9,
144.8, 148.3, 150.9, 169.2.
P r ep a r a tive Solid -Sta te P h otolysis. In a typical reaction,
ca. 100 mg of the gently ground aldehyde was dispersed in a
Pyrex container and purged with a stream of nitrogen gas for
15-10 min. The solid sample was subsequently irradiated,
under a nitrogen gas atmosphere, in a Rayonet reactor fitted
with 350 nm lamps for 24 h. After this period, the irradiated
mixture was subjected to silica gel column chromatography.
The combined fractions from the chromatography correspond-
ing to the cyclobutenol were stripped off the solvent in vacuo
at room temperature and characterized.
The benzocyclobutenol from the solid-state photolysis of the
complex of mesitaldehydes 1a and diol host was established
by comparison of its spectral data with those documented in
the literature.24 All of the cyclobutenols exhibited similar
spectral characteristics.
1b-CB: IR (KBr) cm-1 3301, 2920, 1462, 1202, 1169; 1H
NMR (CDCl3, 400 MHz) δ 2.13 (s, 3H), 2.26 (s, 3H), 2.81 (d, J
) 14 Hz, 1H), 3.41 (dd, J 1 )14 Hz, J 2 ) 4.4 Hz, 1H), 5.10 (d,
J ) 4.4 Hz, 1H), 6.82 (s, 1H); 13C NMR (CDCl3, 100 MHz) δ
16.0, 22.5, 42.0, 68.5, 115.7, 123.5, 131.3, 138.7, 142.5, 144.4.
1c-CB: IR (KBr) cm-1 3434, 2924, 2224, 1610, 1457, 1401,
1296, 1219; 1H NMR (CDCl3, 400 MHz) δ 2.38 (s, 3H), 2.45 (s,
3H), 2.94 (d, J ) 15.2 Hz, 1H), 3.51 (dd, 1H, J 1 )15.2 Hz, J 2
) 4.7 Hz), 5.23 (d, 1H, 4.4 Hz), 6.86 (s, 1H); 13C NMR (CDCl3,
100 MHz) δ 15.5, 21.5, 42.2, 69.9, 111.8, 117.4, 122.7, 130.1,
137.1, 144.2, 146.9.
Tem p er a tu r e-Dep en d en t Solid -Sta te P h otolysis. For
examining the temperature dependence on the yields of the
photoreactions, two portions of the same sample (1d or 1e)
were dispersed in NMR tubes and purged with a N2 gas for
15 min. The irradiation of each at a particular temperature
was conducted by focusing the Pyrex-filtered radiation from
high-pressure Hg-lamp on to the sample held in a quartz
Dewar Vessel equipped with a window (Applied Photophysics
Limited). The bath temperature in the Dewar was externally
controlled. During low temperature photolysis, the condensate
that formed on the window of the Dewar flask was constantly
removed by blowing hot air from a dryer. The whole setup was
maintained undisturbed during the photolysis at two different
temperatures, except for changing the NMR tubes containing
the samples. In three independent experiments, a similar
trend in the yields, i.e., the yield at low temperature (-80 °C)
was significantly lower than that at room temperature (30 °C),
was observed.
Ack n ow led gm en t. We thank Department of Sci-
ence and Technology (DST), New Delhi, for financial
support. P.M. gratefully acknowledges the J unior Re-
search Fellowship (J RF) from Council of Scientific and
Industrial Research, New Delhi.
Su p p or tin g In for m a tion Ava ila ble: The crystallographic
data for 1a -diol complex and 1e, and summary output files
of AM1 calculations. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O015718R