Importance of Steric Effects in the Photochemical Ring Opening of Isochromenes

Full text of DOI:10.1021/jo961496z

J . Org. Chem. 1997, 62, 1183-1187
1183
Im p or ta n ce of Ster ic Effects in th e
P h otoch em ica l Rin g Op en in g of
Isoch r om en es
Richard E. Smith and Nigel G. J . Richards*
Department of Chemistry, University of Florida,
Gainesville, Florida 32611
Received August 2, 1996
The use of time-resolved X-ray crystallography1,2 is
limited by the availability of suitably “caged” substrates
that can be employed to initiate simultaneous reaction
in a sufficient number of complexes in the crystal.3
Compounds which can be transformed into appropriate
substrates by photochemical activation, usually termed
photosubstrates, have been successfully used to observe
enzyme-substrate complexes.4 The detailed character-
ization of the photochemical and photophysical behavior
of heterocycles5 represents an intriguing approach to
developing novel photocaged enzyme substrates. We
recently proposed that photoinitiated ring opening of
3-aminoisochromenes might allow generation of an amide
bond within the active site of an aspartic proteinase, via
solvent trapping of the reactive ring-opened intermediate
(eq 1).
F igu r e 1. Results of previous broadband irradiation studies
on 3-phenylisochromene (1) and 4-substituted derivatives 2
and 3. Structures 4-7 are proposed intermediates. Epoxides
8 and 9 were identified using direct methods.
reported to give products 10-12, in what was postulated
to be a two-photon process.6a Absorption of the first
photon was presumed to yield ring-opened intermediates
4-6 which subsequently underwent an intramolecular
[4 + 2] cycloaddition after activation by a second photon.
Surprisingly, while intermediate epoxides 8 and 9 were
observed and characterized, no direct evidence for the
intermediacy of 7 could be obtained in these experiments.6a
We now report, contrary to previous studies, that we
do not obtain indan 10 when 3-phenylisochromene (1) is
irradiated with either monochromatic or broadband
radiation in dry methanol. Instead, the reaction consis-
tently yields 21 through photochemically induced solvent
addition to the C3-C4 double bond (eq 2). Our observa-
tions explain the failure to isolate 7 and may suggest
some role for excited state steric interactions in control-
ling photoinduced isochromene ring-opening.
(1)
The photochemical ring opening of substituted isochro-
menes has been previously investigated using experi-
mental6 and theoretical methods.7 Thus, broadband
irradiation of 3-phenylisochromenes 1-3 (Figure 1) was
* Address correspondence to this author at Department of Chem-
istry, P.O. Box 117200, University of Florida, Gainesville, FL 32611-
7200. Phone: (352) 392-3601. Fax: (352) 846-2095. e-mail: richards@
qtp.ufl.edu.
(2)
(1) (a) Pai, E. F. Curr. Opin. Struct. Biol. 1992, 2, 821-827. (b)
Hajdu, J .; J ohnson, L. N. Biochemistry 1990, 29, 1669-1678. (c) Hajdu,
J .; Acharya, K. R.; Stuart, D. I.; J ohnson, L. N. Trends Biochem. Sci.
1987, 13, 104-109. (d) Stoddard, B. L.; Koenigs, P.; Porter, N. A.;
Petratos, K.; Petsko, G. A.; Ringe, D. Proc. Natl. Acad. Sci. U.S.A. 1991,
88, 5503-5507. (e) Stoddard, B. L.; Bruhnke, J .; Koenigs, P.; Porter,
N.; Ringe, D.; Petsko, G. Biochemistry 1990, 29, 8042-8051.
(2) (a) Singer, P. T.; Smalås, A.; Carty, R. P.; Mangel, W. F.; Sweet,
R. M. Science 1993, 259, 669-673. (b) Blow, D. Curr. Biol. 1993, 3,
204-207.
Resu lts
The synthesis of isochromenes is well documented,8
and 3-phenylisochromene 1 was efficiently prepared in
three steps from dibromide 13 in 42% overall yield
(Figure 2A).9a After quantitative formation of salt 14,
treatment with sodium benzoate gave 15 which yielded
1 via an intramolecular Wittig reaction. This route was
easily extended to give analogs 18 and 19 (Figure 2A).
Monochromatic irradiation of 1 at 325 nm, in deoxygen-
ated methanol, using a helium-cadmium CW laser, gave
acetal 21 as the only product. The photochemical reac-
tion exhibited zero-order kinetics, becoming first order
(3) Porter, N. A.; Bruhnke, J .; Koenigs, P. In Biological Applications
of Photochemical Switches; Morrison, H., Ed.; Wiley-Interscience: New
York, 1993; pp 197-241.
(4) (a) Scheidig, A.; Pai, E. F.; Schlichtling, I.; Corrie, J . E. T.; Reid,
G. P.; Wittinghofer, A.; Goody, R. S. Philos. Trans. R. Soc. Ser. A 1992,
340, 263-272. (b) Schlichtling, I.; Almo, S. C.; Rapp, G.; Wilson, K.;
Petratos, K.; Lentfer, A.; Wittinghofer, A.; Kabsch, W.; Pai, E. F.;
Petsko, G. A.; Goody, R. S. Nature (London) 1990, 345, 309-315.
(5) Photochemistry of Heterocyclic Compounds; Buchardt, O., Ed.;
Wiley: New York, 1976.
(6) (a) Padwa, A.; Au, A.; Lee, G. A.; Owens, W. J . Org Chem. 1975,
40, 1142-1149. (b) Padwa, A.; Au, A.; Owens, W. J . Chem. Soc., Chem.
Commun. 1974, 675-676.
(7) (a) Richards, N. G. J .; Cory, M. G., J r. Int. J . Quant. Chem.,
Quant. Biol. Symp. 1992, 19, 65-76. (b) Cory, M. G., J r.; Zerner, M.
C.; Richards, N. G. J . In Modeling the Hydrogen Bond, Smith, D. A.,
Ed.; ACS Symposium Series #569, American Chemical Society: Wash-
ington, DC, 1994; pp 222-234.
(8) (a) Cottet, F.; Cottier, L.; Descotes, G. Synthesis 1987, 497-498.
(b) Hug, R.; Hansen, H. J .; Schmid, H. Helv. Chim. Acta 1972, 55, 10-
15. (c) Kristinsson, H.; Mateer, R. A.; Griffin, G. W. J . Chem. Soc.,
Chem. Commun. 1966, 415-416. (d) Chatterjea, J . N. Chem. Ber. 1958,
91, 2636-2638.
(9) (a) Begasse, B.; LeCorre, M. Synthesis 1981, 197-197. (b)
Begasse, B.; Hercouet, A.; LeCorre, M. Tetrahedron Lett. 1979, 23,
2149-2150.
S0022-3263(96)01496-X CCC: $14.00 © 1997 American Chemical Society

1184 J . Org. Chem., Vol. 62, No. 4, 1997
Notes
arylisochromenes. Thus Φ was taken as approximating
the rate of solvent addition to the excited state. Plotting
log10(Φ/ΦH) against the Hammett σ values of the sub-
stituents on the phenyl ring gave a line of slope F ) -1.2
( 0.33, consistent with the involvement of a carbocation
intermediate in solvent addition.11
In our last set of experiments, we irradiated 1 in dry,
degassed methanol containing various concentrations of
benzophenone using monochromatic light. A plot of Φ°/Φ
against the concentration of benzophenone yielded a
straight line of zero slope (data not shown), where Φ°
and Φ are the primary quantum yields for the formation
of 21 in the presence, and absence, of benzophenone,
respectively. In light of this result and our computational
studies on the excited states of 1 after excitation at 325
nm,7 the photochemical reaction probably proceeds from
the singlet state.11
Discu ssion
Irradiation of 1 with monochromatic light at 325 nm,
the absorption maximum most likely to yield the ring-
opened product after electronic excitation,6,7a was antici-
pated to give 22 after solvent addition to 4 to restore the
aromatic π-system (eq 3).
F igu r e 2. (A) Preparation of isochromenes 1, 18, and 19. (i)
PPh3, toluene; (ii) RCO2Na, acetone/H2O, rt; iii) CH3CH2C-
(ONa)(CH3)CH3, toluene. (B) Synthesis of authentic 10. (i)
m-CPBA, CH2Cl2, rt; (ii) MeOH, TsOH.
(3)
when 1 was present at concentrations less than 65 µM.10
Since previous studies had reported that broadband
irradiation of 1 gave the phenylindan derivative 10, we
confirmed that 21 was also obtained as the only product
from 1 under these conditions. Finally, an authentic
sample of 10 was prepared following literature proce-
dures (Figure 2B),6a,12 although the spectroscopic proper-
ties of the material synthesized in our laboratory were
different from those reported for the product obtained in
the previous photochemical studies of 1.6a
Having established that irradiation of 1 under our
conditions did not proceed as originally reported, we next
examined the mechanism of formation of 21. After
ensuring that 21 was not obtained in the absence of light,
we determined the primary quantum yield for photoin-
duced acetal formation from 1 to be 6.9 × 10-3 molecules/
quantum, in reasonable agreement with that reported
previously.6a,11 Quantum yields for the phenyl-substi-
tuted analogs 18 and 19 were 7.5 × 10-3 and 2.9 × 10-3
molecules/quantum, respectively. For the photoaddition
of methanol to an arylisochromene, the kinetic expression
for the primary quantum yield Φ, under conditions of
saturating arylisochromene and MeOH, can be written
as Φ ) Rk′/(k′+k2) where k′ ) k1[arylisochromene]-
[MeOH] and is therefore the zero-order rate constant, k2
is the net rate constant describing all other pathways
by which the excited state can relax, and R is a propor-
tionality constant. Rearranging gives Rk′/k2 ) Φ/(1 - Φ).
In our experiments, the primary quantum yields for all
compounds were significantly smaller than unity, and
given the structural similarity of 1, 18, and 19, it is likely
that k2 and R are approximately equal for all three
Calculations had also shown that the ring-opened
intermediate 4 does not possess an appropriate absorp-
tion band at this wavelength,7a making the occurrence
of multiple photon reaction pathways, as reported for
broadband irradiation, unlikely. The formation of 21 in
our irradiation experiments was therefore unanticipated
given the previous work on the photochemistry of isochro-
menes 1-3.6a In order to rule out the possibility that
we were forming, but failing to isolate, 10 in our experi-
ments, we prepared an authentic sample starting from
2-phenylindene using literature procedures.6a,12 Com-
parison of the spectroscopic properties of 10 prepared in
our laboratory, however, with those reported in the
original broadband irradiation studies revealed a number
of differences. First, a broad absorbance centered at 3325
cm-1 in the IR spectrum of authentic 10, presumably
associated with the hydroxyl group, was not reported for
the photoproduct formed by broadband irradiation of 1
in the original study.13 The IR data for the photoproduct
obtained in the original work are strikingly similar to
that observed for 1 (see Experimental Section). We have
observed that addition of 95% EtOH (a recrystallization
solvent in the earlier study) to 21 precipitates 1 as a
white solid. Second, in the original characterization of
the putative photoproduct, the resonance arising from the
C1 protons was assigned a chemical shift of δ 4.93.6a In
contrast, the cognate protons in 11 were reported to
possess a chemical shift of δ 3.38,6a which is a striking
difference given the structural similarity of 10 and 11.
1
(10) Bunce, N. J . In CRC Handbook of Organic Photochemistry;
Scaiano, J . C., Ed.; CRC Press: Boca Raton, FL, 1989; Vol. I, pp 241-
256.
Examination of the H-NMR spectrum of authentic 10,
however, shows that a resonance at δ 3.58 can be
(11) Wagner, P. J . In CRC Handbook of Organic Photochemistry;
Scaiano, J . C., Ed.; CRC Press: Boca Raton, FL, 1989; Vol. II, pp 261-
269.
(13) In the original experiments,6a the IR data for the compound
produced on broadband irradiation of 1 was reported as IR (KBr) ν
(12) Braun, J .; Manz, G. Chem. Ber. 1929, 62, 1059-1065.
1618, 1492, 1451, 1389, 1264, 1208, 1063 cm-1
.

Notes
J . Org. Chem., Vol. 62, No. 4, 1997 1185
assigned to the C1 protons, in agreement with the
assignments previously reported for 11 and 12.6a,14
Furthermore, on addition of D2O, the resonance at δ 1.53
was lost due to exchange of the hydroxyl proton and the
δ 3.38 doublet collapsed to a singlet indicating that the
hydroxyl group was attached to C-3 in 10. These
observations are further support for the proposal that the
structure of the photoproduct originally obtained from
broadband irradiation of 1 should be reassigned to that
of 21. On the other hand, the authors of the original
study reported that mixing their photoproduct with an
authentic sample of 10 (prepared by the independent
route employed in our studies) did not depress the
melting point. In light of the differences in spectroscopic
data reported here, it is possible that the workup and
purification conditions used in the original study might
have transformed 21, the actual photoproduct, and
authentic 10 into a third compound via acid-catalyzed
rearrangement. The validity of such a hypothesis re-
mains to be investigated.
There is no doubt that compounds 2 and 3 undergo
photoconversion to 11 and 12, respectively, via the
postulated ring-opening pathway (Figure 1). For ex-
ample, epoxides 8 and 9 have been isolated and demon-
strated to give 11 and 12, respectively, upon further
broadband irradiation in methanol.6a We note, however,
that the intermediacy of epoxide 7 was only inferred in
previous irradiation experiments involving 3-phenyliso-
chromene (1) on the basis of the photoproduct identifica-
tion.
In further experiments, we obtained an approximate
linear free energy relationship for the addition of metha-
nol to 1, under conditions of monochromatic irradiation,
using a series of analogs of 1.15 While the number of
derivatives possessing absorption maxima close to 325
nm was limited, the F-value obtained from our experi-
ments was consistent with the intermediacy of a transi-
tion state possessing cationic character. This raised the
possibility that acetal 21 was being produced by acid-
catalyzed solvent addition due to the photochemical
oxidation of methanol.16 We determined, however, that
3-phenylisochromene (1) is recovered unchanged after
being stirred at room temperature for up to 18 h in 0.2
M HCO2H/MeOH. It is therefore likely that 21 is formed
by photochemical addition of methanol to the C3-C4
double bond of 117 and that the reaction does not involve
ring-opened compound 4 as an intermediate. While it is
possible that addition of methanol to the carbonyl group
of 4 followed by 6-endo-trig ring closure could yield 21,
the second step corresponds to a disfavored cyclization.18
In addition, we have been unable to detect the formation
of adduct 22 under our standard reaction conditions.
Photochemical excitation of cyclic olefins, at least for
cyclohexenes19a,b and larger ring systems,17 gives products
from the singlet excited state via double-bond isomer-
ization. In protic solvents, protonation of this highly
strained species gives a carbocation which can then react
with the solvent. It is clear from our experiments that
the formation of 21 proceeds via such a mechanism,
raising questions concerning the molecular reasons un-
derlying our observation that the photochemical reaction
of 1 differs from that of 2 and 3. In our computational
investigations of the ring opening of isochromenes 1-3,7a
we determined that the absorption at 323 nm corresponds
to a π-π* transition principally associated with the C3-
C4 bond. As a result, there is a localized distortion of
the molecular geometry in the excited state, analogous
to that observed for ethylene upon π-π* electronic
excitation. Our calculations also indicate that the cyclic
system must become substantially distorted before the
σ* antibonding MO associated with the C1-O2 bond
becomes the dominant contribution in the S1 excited
state.7a Given the fusion of the cyclohexene and benzene
rings, however, molecular distortion is likely to be
difficult, consistent with the low quantum yield for the
conversion of 1 into 21, unless steric interactions between
substituents, other than hydrogen, at C3 and C4 can be
relieved by the twisting motion. Hence, in 1, as the C4
substituent is a hydrogen, it is possible that the amount
of C3-C4 double-bond twisting might be lower compared
to that in the excited states of 2 and 3. With the ring-
opening pathway disfavored, the excited state would then
return to the ground state by the usual mechanisms or
undergo subsequent reaction with methanol. Alterna-
tively, the additional sp2 centers in 1 might prevent cis-
trans isomerization and give rise to a zwitterionic
intermediate via electronic polarization,20 unless steric
interactions between the C3 and C4 substituents, other
than hydrogen, could be relieved. Trapping of the
carbocation formed by protonation of the zwitterion would
then yield 21. While this pathway is less plausible, the
effect of fused aromatic rings on the isomerization of
cyclohexenes does not appear to have been investigated
by direct spectroscopic measurement. In any case, such
models suggest that the steric interactions between the
C3 and C4 substituents are an important factor in
modulating the photoinitated ring opening of 3-aryliso-
chromenes. The general applicability of this observation
to a number of related heterocyclic compounds is under
investigation.
Exp er im en ta l Section
Melting points were recorded using a Fisher-J ohns melting
point apparatus and are uncorrected. Chemical shifts are
reported in ppm (δ) downfield of tetramethylsilane as an internal
reference (δ 0.0). Splitting patterns are abbreviated as follows:
s, singlet, d, doublet, t, triplet, q, quartet, and, m, multiplet.
Combustion analyses for C, H, and N were determined in the
Microanalysis Facility in the Department of Chemistry, Uni-
versity of Florida. Ammonia or isobutane was used in the CI
measurements. Analytical thin layer chromatography (TLC)
(14) We note that, in light of our results, Dr. Padwa has reexamined
the original spectroscopic data for the product of the original photo-
chemical studies on 1. It appears that the NMR assignments given in
the earlier paper were misreported and a correction has been published
concerning this point. Padwa, A.; Au, A.; Lee, G. A.; Owens, W. J .
Org. Chem. 1996, 61, 9072.
(15) Yasuda, M.; Matsuzaki, Y.; Shima, K.; Pac. C. J . Chem. Soc.,
Perkin Trans. 2 1988, 745-751.
(16) (a) Meinwald, J .; Chapman, R. A. J . Am. Chem. Soc. 1968, 90,
3218-3226. (b) Cristol, S. J .; Lee, G. A.; Noreen, A. L. Tetrahedron
Lett. 1971, 4175-4179.
(17) (a) Kropp, P. J .; Reardon, E. J ., J r.; Gaibel, Z. L. F.; Williard,
K. F.; Hattaway, J . H., J r. J . Am. Chem. Soc. 1973, 95, 7058-7067.
(b) Hixson, S. S. Tetrahedron Lett. 1973, 277-280. (c) Miyamoto, N.;
Kawanisi, M.; Nozaki, H. Tetrahedron Lett. 1971, 2565-2566.
(18) (a) Baldwin, J . E. J . Chem. Soc., Chem. Commun. 1976, 734-
736. (b) Baldwin, J . E. J . Chem. Soc., Chem. Commun. 1976, 736-
738.
(19) (a) Bonneau, R.; J oussot-Dubien, J .; Salem, L.; Yarwood, A. J .
J . Am. Chem. Soc. 1976, 98, 4330-4331. (b) Saltiel, J .; Marchand, G.
R. J . Am. Chem. Soc. 1991, 113, 2702-2708. (c) Marshall, J . A. Science
1970, 170, 137-141. (d) Kropp, P. J . In CRC Handbook of Organic
Photochemistry and Photobiology; Horspool, W. M., Song, P.-S., Eds.;
CRC Press, Boca Raton, FL, 1995; pp 105-114. (e) Kropp, P. J . Pure
Appl. Chem. 1970, 24, 585-598.
(20) (a) Bonacic-Koutecky, V.; Bruckmann, P.; Hiberty, P.; Koutecky,
J .; Leforestier, C.; Salem, L. Angew. Chem., Int. Ed. Engl. 1975, 14,
575-576.

1186 J . Org. Chem., Vol. 62, No. 4, 1997
Notes
was performed on silica gel 60F-245 plates, unless otherwise
stated. Flash chromatography was performed by standard
methods21 on Davisil grade 633 type 60A silica gel (200-425
mesh). Laser (monochromatic) irradiation experiments were
performed using a Liconix He/Cd CW laser at 325 nm, nominally
rated at 10 mW. Laser power determinations employed a Phir
Model 10A thermopile power meter. A 450 W Hanovia mercury-
vapor arc lamp, fitted with a Pyrex water-cooling jacket, was
used in all broadband irradiation experiments. Diethyl ether
was distilled from sodium metal/benzophenone ketyl. Toluene
and CH2Cl2 were distilled from CaH2 before use while MeOH
was purified by distillation from magnesium. m-Chloroperben-
zoic acid (m-CPBA) was purified by being washed with NaHCO3
before being dried in vacuo. All other compounds were pur-
chased from Aldrich and used without further purification.
Reactions requiring anhydrous conditions were performed under
a positive pressure of N2.
Gen er a l P r oced u r e for t h e P r ep a r a t ion of 3-Ar yliso-
ch r om en es. Solutions of the appropriate carboxylate salts (1
equiv) in either water or 1:1 aqueous acetone were added to a
slurry of the known triphenylphosphonium salt 14 (1 equiv)
(prepared by reaction of triphenylphosphine and dibromide 13)9
and stirred at rt for 8-12 h. After removal of acetone in vacuo,
an equal volume of water was added and the resulting solution
was extracted with CH2Cl2. The organic extracts were then
dried (Na2SO4), and removal of the solvent under reduced
pressure gave the esters 15-17 in isolated yields of greater than
94%. Conversion of these esters into the corresponding iso-
chromenes was accomplished by the dropwise addition of 1 equiv
of sodium amylate (0.11 M solution in anhydrous toluene) to a
stirred slurry of the appropriate ester in boiling toluene (0.1 M
concentration). After being refluxed for a further 4-8 h, the
solution was allowed to cool and the solvent was removed under
reduced pressure to give a solid residue that was washed with
95% aqueous EtOH. Purification was then effected by recrys-
talization.
use in the subsequent cyclization reaction. Treatment of 17 (6.02
g, 10 mmol) with sodium amylate in toluene and subsequent
workup gave a solid residue. Recrystallization from aqueous
EtOH yielded isochromene 19 as white plates: 816 mg, 32%;
mp 109-110 °C; IR (KBr) ν 1614, 1484, 1449, 1402, 1273, 1091,
1055, 1002 cm-1; 1H NMR (CDCl3, 300 MHz) 5.20 (2 H, s), 6.43
(1 H, s), 7.00-7.40 (8 H, m); 13C NMR (CDCl3, 75.4 MHz) δ 69.0
(t), 101.5 (d), 123.6 (d), 123.8 (d), 126.3 (d), 126.7 (d), 128.0 (d),
128.3 (d), 128.5 (d), 131.7 (s), 132.8 (s), 134.6 (s), 152.9 (s); UV
(MeOH) λ (ꢀ) 328 (18 300), 240 nm (17 100); MS (FAB/TD) m/ e
(relative intensity) 244 ((M + 2)+, 29), 242 (M+, 94), 155 (22),
89 (15); exact mass calcd for M+ C15H11ClO requires 242.0498,
found 242.0473 (FAB/TD). Anal. Calcd for C15H11ClO: C, 74.23;
H, 4.57. Found: C, 74.62; H, 4.81.
3-Hyd r oxy-2-m eth oxy-2-p h en ylin d a n (10). A solution of
phenylindene 20 (300 mg, 1.56 mmol)12 and freshly purified
m-CPBA (673 mg, 3.90 mmol) in CH2Cl2 (20 mL) was stirred at
rt for 14 h. The reaction mixture was then washed sequentially
with 10% aqueous NaHCO3 (20 mL) and water (20 mL) before
the organic phase was dried (MgSO4). Removal of the solvent
under reduced pressure yielded 2,3-epoxyphenylindan (7) as a
1
colorless oil: 242 mg, 74%; H NMR (CDCl3, 300 MHz) 3.41 (1
H, d, J ) 18.0 Hz), 3.57 (1 H, d, J ) 18.0 Hz), 4.33 (1 H, s),
7.20-7.70 (9 H, m). The epoxide 7 (233 mg, 1.12 mmol) was
then dissolved in MeOH (5 mL) together with TsOH (4 mg, 0.02
mmol) and the solution stirred at rt for 1 h. After removal of
the solvent under reduced pressure, flash chromatography
(alumina; eluant 50% EtOAc/petroleum ether; Rf ) 0.8) yielded
10 as a pale yellow oil: 113 mg, 42%; IR (CHCl3) ν 3330, 1518,
1496, 1407, 1248, 1233, 1202, 1170, 1028 cm-1; 1H NMR (CDCl3,
300 MHz) 1.53 (1 H, exchangeable d, J ) 8.0 Hz), 3.10 (3 H, s),
3.33 (1 H, d, J ) 16.0 Hz), 3.58 (1 H, d, J ) 16.0 Hz), 5.19 (1 H,
d, J ) 8.0 Hz), 7.10-7.60 (9 H, m); MS (FAB/NBA) m/ e (relative
intensity) 240 (M+, 5), 239 (M+ - H, 9), 223 (M+ - OH), 184
(24), 137 (100); exact mass calcd for M+ - OH C16H15O requires
223.1123, found 223.1229 (FAB/NBA).
Br oa d ba n d Ir r a d ia tion of 3-P h en ylisoch r om en e (1). A
solution of 1 (25 mg, 0.12 mmol) in anhydrous MeOH (75 mL),
purged with dry N2, was irradiated at rt in a sealed Pyrex flask
for 22 h, using a mercury-vapor arc lamp fitted with a Pyrex
water-cooling jacket. Removal of the solvent under reduced
pressure gave 3-methoxy-3-phenylbenzo-2H-pyran (21) as a
yellow oil: 28 mg, 97%; IR (CHCl3) ν 1151, 1105, 1047 cm-1; 1H
NMR (CDCl3, 300 MHz) δ 3.01 (1 H, d, J ) 17.5 Hz), 3.11 (3 H,
s), 3.16 (1 H, d, J ) 17.5 Hz), 4.94 (1 H, d, J ) 19.0 Hz), 4.97 (1
H, d, J ) 19.0 Hz), 7.03-7.75 (9 H, m); 13C NMR (CDCl3, 75.4
MHz) δ 40.8 (t), 49.6 (q), 62.8 (t), 99.6 (s), 124.1 (d), 125.6 (d),
126.0 (d), 127.1 (d), 127.3 (d), 127.5 (d), 131.9 (d), 134.0 (s), 135.8
(s), 141.8 (s); MS (FAB MAGIC/TFA) m/ e (relative intensity)
279 (MK+, 47), 223 (12), 103 (67), 85 (100); exact mass calcd for
M+ C16H16O2K requires 279.0787, found 279.0915 (FAB MAGIC/
TFA).
Mon och r om a t ic Ir r a d ia t ion of 3-P h en ylisoch r om en e
(1). A solution of 1 (4.3 mg, 21 µmol) in dry MeOH (3 mL),
purged thoroughly with dry N2, was irradiated using a laser
source at 325 nm for 86 h in a sealed quartz cell. Removal of
the solvent under reduced pressure yielded a 3:2 mixture of
starting material 1 and the acetal 21, upon 1H NMR analysis.
No other compounds were evident in the NMR spectrum of the
crude reaction products or by TLC analysis.
3-P h en ylisoch r om en e (1). Sodium benzoate (1.59 g, 11
mmol) was reacted with triphenylphosphonium salt 14 (5.25 g,
10 mmol), according to the general procedure, to give ester 15
(5.56 g, 98%) of sufficient purity for use in the subsequent
cyclization reaction. Treatment of 15 (1.42 g, 2.5 mmol) with
sodium amylate in toluene and subsequent workup gave a solid
residue. Recrystallization from aqueous EtOH produced iso-
chromene 1 as a white solid: 226 mg, 41%; mp 122-124 °C [lit.6a
mp 122-124 °C]; IR (KBr) ν 1618, 1489, 1451, 1388, 1275, 1264,
1203, 1075, 1063 cm-1 1H NMR (CDCl3, 300 MHz) 5.23 (2 H,
;
s), 6.46 (1 H, s), 7.00-7.75 (9 H, m); 13C NMR (CDCl3, 75.4 MHz)
δ 69.0 (t), 101.1 (d), 123.4 (d), 123.7 (d), 125.0 (d), 126.4 (d), 128.0
(s), 128.2 (d), 128.3 (d), 128.8 (d), 132.0 (s), 134.3 (s), 154.1 (s);
UV (MeOH) λ (ꢀ) 323 (15 600), 235 nm (15 200); MS (FAB/NBA)
m/ e (relative intensity) 208 (M+, 100), 136 (68), 91 (48); exact
mass calcd for M+ C15H12O requires 208.0994, found 208.0888
(FAB/NBA).
3-(4′-Meth oxyp h en yl)isoch r om en e (18). Sodium p-meth-
oxybenzoate (2.61 g, 15 mmol) was reacted with triphenylphos-
phonium salt 14 (7.88 g, 15 mmol), according to the general
procedure, to give ester 16 (8.6 g, 96%) of sufficient purity for
use in the subsequent cyclization reaction. Treatment of 16 (2.99
g, 5 mmol) with sodium amylate in toluene and subsequent
workup gave a solid residue. Recrystallization from aqueous
EtOH yielded isochromene 18 as white plates: 480 mg, 38%;
mp 126-128 °C; IR (KBr) ν 1602, 1502, 1443, 1249, 1172, 1114,
1055, 1026 cm-1; 1H NMR (CDCl3, 300 MHz) 3.83 (3 H, s), 5.20
(2 H, s), 6.43 (1 H, s), 6.85-7.28 (8 H, m); 13C NMR (CDCl3,
75.4 MHz) δ 55.3 (q), 69.0 (t), 99.5 (d), 113.8 (d), 123.1 (d), 123.7
(d), 126.0 (d), 126.6 (d), 127.0 (s), 127.8 (d), 128.1 (d), 132.3 (s),
154.1 (s), 160.3 (s); UV (MeOH) λ (ꢀ) 330 (22 300), 240 nm
(14 600); MS (FAB/TD) m/ e (relative intensity) 239 (MH+, 32),
238 (M+, 100), 237 (15), 209 (8), 135 (15), 105 (10), 91 (8); exact
mass calcd for M+ C16H14O2 requires 238.0994, found 238.0974
(FAB/TD).
P r im a r y Qu a n tu m Yield Deter m in a tion s. Primary quan-
tum yields for the conversion of 1, 18, and 19 into their
respective methanol adducts were obtained by monitoring the
decrease in their associated UV absorption spectra at 325, 330,
and 328 nm, respectively. Samples containing a single iso-
chromene (3 mL aliquots) were subject to monochromatic
irradiation at 325 nm, and the absorbance was measured at 10
min intervals. The photon dose rate I (nmol photons/s) was
calculated from
I ) (λ/hcfmA)R ) 1.135R
3-(4′-Ch lor op h en yl)isoch r om en e (19). Sodium p-chlo-
robenzoate (3.57 g, 20 mmol) was reacted with triphenylphos-
phonium salt 14 (10.53 g, 20 mmol), according to the general
procedure, to give ester 17 (11.42 g, 95%) of sufficient purity for
where h is Planck’s constant, λ is the wavelength of the photons,
c is the velocity of light, A is Avogadro’s constant, fm is the power
meter calibration factor (W mV-1), and R is the mean of ten
power meter determinations (mV). Primary quantum yields (Φ)
were obtained from the linear portion of the plot describing the
decrease in concentration of isochromene versus photon dose.10
(21) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923-
2925.

Notes
Separate experiments to determine the effects of varying
J . Org. Chem., Vol. 62, No. 4, 1997 1187
tion in clarifying many details of the original experi-
ments and for making a number of useful comments on
the studies reported herein. We also acknowledge
J ames Winefordner and Ben Smith for access to laser
facilities and Kirk Schanze for help with the broadband
irradiation experiments.
concentrations of benzophenone upon the primary quantum
yields for the conversion of 1 to 21 were carried out in dry, N2-
purged MeOH. Thus, the concentration of benzophenone was
varied from 0.15 to 1.5 mM, while that of 1 was fixed at 0.15
mM. A linear calibration curve, constructed using a 1.5 mM
solution of benzophenone in MeOH, was employed to correct the
number of photons absorbed by 1 in solutions containing
benzophenone, as the latter compound has a small absorbance,
relative to that of 1 at the irradiation wavelength.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra for 1, 18, 19, 20, and 21 and 1H spectra for 10 (14
pages). This material is contained in libraries on microfiche,
immediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
Ack n ow led gm en t. This work was supported by a
Research Development Award from the Division of
Sponsored Research of the University of Florida. Par-
tial funding was also provided by Procter & Gamble,
Cincinatti. We thank Dr. Albert Padwa for his coopera-
J O961496Z