2-benzoylbenzoic acid: A photolabile mask for alcohols and thiols

Article abstract of DOI:10.1021/jo961638p

Photolysis of 2-benzoylbenzoate esters of primary and secondary alcohols 1 in the presence of a hydrogen donor (2-isopropanol) or an electron donor (primary amines) produces the corresponding alcohol in high yield. The fate of the benzoate is dependent on the conditions used for the photoreaction. In 2-propanol, the ketyl radical that derives from photoreduction dimerizes, to afford the benzpinacol product 3,3"-diphenylbiphthalidyl, 5. In the presence of amines the product is 3-phenylphthalide, 6, a benzhydrol derivative which is the result of simple reduction of the ketone followed by lactonization. While the photoproduct of the benzoate - 2-propanol reaction results from anticipated free radical chemistry, the amine-promoted reaction appears to result from a second, "dark", electron transfer process. We conclude that 2-benzoylbenzoic acid is an effective photolabile protecting group for primary and secondary alcohols, and preliminary studies indicate that thiols can be protected in an analogous way. Studies on the effect of benzophenone substituents and reaction solvent on the benzhydrol:benzpinacol product ratio provide mechanistic insight into the process.

Full text of DOI:10.1021/jo961638p

J . Org. Chem. 1996, 61, 9455-9461
9455
2-Ben zoylben zoic Acid : A P h otola bile Ma sk for Alcoh ols a n d
Th iols
Paul B. J ones, Michael P. Pollastri, and Ned A. Porter*
Department of Chemistry, Duke University, Durham, North Carolina 27708
Received August 23, 1996X
Photolysis of 2-benzoylbenzoate esters of primary and secondary alcohols 1 in the presence of a
hydrogen donor (2-isopropanol) or an electron donor (primary amines) produces the corresponding
alcohol in high yield. The fate of the benzoate is dependent on the conditions used for the
photoreaction. In 2-propanol, the ketyl radical that derives from photoreduction dimerizes, to afford
the benzpinacol product 3,3′-diphenylbiphthalidyl, 5. In the presence of amines the product is
3-phenylphthalide, 6, a benzhydrol derivative which is the result of simple reduction of the ketone
followed by lactonization. While the photoproduct of the benzoate-2-propanol reaction results from
anticipated free radical chemistry, the amine-promoted reaction appears to result from a second,
“dark”, electron transfer process. We conclude that 2-benzoylbenzoic acid is an effective photolabile
protecting group for primary and secondary alcohols, and preliminary studies indicate that thiols
can be protected in an analogous way. Studies on the effect of benzophenone substituents and
reaction solvent on the benzhydrol:benzpinacol product ratio provide mechanistic insight into the
process.
The use of photoremovable protecting groups has
shown a marked increase in recent years. Thus, photo-
removable protecting groups have recently seen new uses
to “cage” enzymes by protecting alcohol and amine
groups. Nucleophilic functional groups have also been
protected by photoremovable groups in recent applica-
tions in peptide and nucleic acid syntheses. The most
popular strategies for functional group protection by
photoremovable groups involve the use of the 2-nitroben-
zyl,2,3 substituted benzyloxycarbonyl,4-6 phenacyl,7 and
3,5-dimethoxybenzoinyl groups.8,9 These functional groups
have found widespread use in protecting both alcohols
and amines in organic synthesis. While the mechanism
for photodeprotection is still under active investigation
in some cases for these established photoprotecting
groups, the critical photochemical step involves either
intramolecular H transfer (2-nitrobenzyl and analogs) or
carbon-heteroatom bond fragmentation (benzyloxycar-
bonyl, phenacyl, benzoinyl). More recently developed
photoprotecting groups depend upon trans-cis isomer-
ization as the critical photochemical step.10-12 In one case
photochemical trans-cis isomerization is coupled to
ground-state lactonization and release of an enzyme
alcohol functional group, thus producing the native active
enzyme and opening the enzyme “cage”.
In order to expand the repertoire of photochemical
reactions used in functional group protection, we have
sought to develop strategies based upon the familiar
photoreduction of the phenone functional group. In
considering various possibilities, we were led to the
conclusion that esters of 2-benzoylbenzoic acid might be
effective photolabile masks for alcohol functionalities.
Indeed, there are reports in the literature of the reduction
of 2-benzoylbenzoate esters 1 through chemical,13 polara-
graphic,14,15 and photochemical16 means, which give as
product either 3-phenylphthalide, 6, or its dimer 5
(Scheme 1). These reactions also result in the production
of the alcohol from the ester precursor although previous
work has focused on the fate of the aromatic product. On
the basis of these reports, we sought to characterize the
utility of these esters as photolabile masks for the alcohol
group. This report describes the synthesis of several
esters of 2-benzoylbenzoic acid and the study of the
photoreduction of these esters using both H atom donors
and electron transfer donors as the reducing agent.
Resu lts
Primary and secondary alcohols and one thiol were
protected as the corresponding 2-benzoylbenzoate ester.
Efforts to make an ester using a tertiary alcohol (1-
methyl-1-phenylethanol) failed. The synthesis of these
esters was carried out by DCC coupling of the alcohol
(or thiol) and commercially available 2-benzoylbenzoic
acid in pyridine. Yields of isolated ester were in the
range of 50-79%.
* To whom correspondence should be addressed. Tel: (919)660-
1550. Fax: (919)660-1605. E-mail: porter@chem.duke.edu.
X Abstract published in Advance ACS Abstracts, December 1, 1996.
(1) Pillai, V. N. R. J . Am. Chem. Soc. 1980, 1-26.
(2) DeMayo, P. Adv. Org. Chem. 1960, 2, 367-426.
(3) Morrison, H. A. Chemistry of the Nitro and Nitroso Groups; J ohn
Wiley and Sons: New York, 1970; pp 185-191.
(4) Patchornik, A.; Amit, B.; Woodward, R. B. J . Am. Chem. Soc.
1970, 92, 6333-6335.
(5) Baltrop, J . A.; Schoefield, P. Tetrahedron Lett. 1962, 697-699.
(6) Chamberlin, J . W. J . Org. Chem. 1966, 31, 1658-59.
(7) Sheehan, J . C.; Umezawa, K. J . Org. Chem. 1973, 38, 3771-
3774.
(8) Sheehan, J . C.; Wilson, R. W. J . Am. Chem. Soc. 1964, 86, 5277-
5281.
Photoreduction of these compounds was carried out
using a 450 W medium pressure mercury lamp. The
solutions were degassed by bubbling argon through them
(9) Sheehan, J . C.; Wilson, R. W.; Oxford, A. W. J . Am. Chem. Soc.
1971, 93, 7222-7228.
(10) Porter, N. A.; Stoddard, B. L.; Koenigs, P.; Petratos, G. A.;
Ringe, D. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 5503.
(11) Porter, N. A.; Bruhnke, J . D. J . Am. Chem. Soc. 1989, 111, 7616.
(12) Koenigs, P.; Faust, B. C.; Porter, N. A. J . Am. Chem. Soc. 1993,
115, 9371.
(13) Risinger, G. E.; Thompson, J . A. Rec. Trav. Chim. Pay-Bas 1963,
82, 801-805.
(14) Perrin, C. L. Prog. Phys. Org. Chem. 1965, 3, 221-224.
(15) Wawzonek, S.; Laitinen, H. A.; Kwiatkowski, S. J . J . Am. Chem.
Soc. 1944, 66, 827-830.
(16) Tsujimoto, Y.; Shigemitsu, Y.; Miyamoto, T.; Odaira, Y. Bull.
Chem. Soc. J pn. 1976, 49, 1445-1446.
S0022-3263(96)01638-6 CCC: $12.00 © 1996 American Chemical Society

9456 J . Org. Chem., Vol. 61, No. 26, 1996
J ones et al.
Sch em e 1
Ta ble 1. P h otor ed u ction of Ben zoylben zoa te Ester s
w ith 2-P r op a n ol or P r im a r y Am in es
reaction of the esters of secondary alcohols required 18
h of photolysis.
yieldsb (%)
In the case of 1a , dodecanol was isolated in 90% yield
by flash chromatography, while the isolated yield of 6 in
this experiment was 74%. The reducing agent may be
an amine other than cyclohexylamine, as shown in
entries 8-10 where sec-butylamine was used. Protected
nucleoside 1e was photoreduced with sec-butylamine and
a uranium glass filter (λcutoff < 350 nm) (entry 10). The
yield of nucleoside, as determined by NMR, was 90% and
that of 3-phenylphthalide (6) was 80%. Photolysis of 1e
in the absence of the uranium glass filter gave a
significantly lower yield of alcohol, as well as unidentified
side products, but about the same yield of 6 (entry 9).
This is, presumably, due to photochemistry involving the
aromatic ring of the nucleoside which has an absorbance
at wavelengths up to 330 nm. The use of the uranium
glass filter also demonstrates that the masking group can
be removed with light of λ ) 366 nm, though a longer
photolysis time is required. The keto esters are trans-
parent at wavelengths greater than 390 nm (data not
shown).
The photoreduction of thio ester 1f using cyclohexyl-
amine as the reducing agent gave a mixture of dode-
canethiol and dodecane disulfide, which were produced
in a 3:1 ratio. In addition, both 3-phenylphthalide, 6,
and diphthalidyl 5 were produced in a 2:1 ratio. Pho-
tolysis of the thio ester in the absence of cyclohexylamine
produced significant quantities of diphthalidyl 5 and
dodecane disulfide. When the disulfide was photolyzed
in the presence of cyclohexylamine, it was slowly con-
verted to thiol. Additional reaction pathways afford a
more complex reaction mixture in the thio ester photoly-
sis than the cases involving oxo esters.
entry substrate conditionsa ROH(SH) RSSR
5
6
1
2
3
4
5
6
7
8
9
1a
1a
1b
1b
1c
1c
1d
1d
1e
1e
1f
a, 5 h
b, 5 h
a, 18 h
b, 18 h
a, 18 h
b, 18 h
a, 5 h
c, 5 h
95
95
85
85
100
20
85
90
60
90
60
0
95
0
80
tr
95
0
80
70
75
80
95
0
85
0
5
0
0
0
c, 5 h
d, 24 h
a, 5 h
10
11
0
20
>45 20
a
a: 0.1 M ketone, 1 M cyclohexylamine, 1:1 benzene:acetoni-
trile, Pyrex filter. b: 0.01 M ketone, 1:1benzene:-2-propanol, Pyrex
filter. c: 0.1 M ketone, 1 M sec-butylamine, 1:1 benzene:-2-
propanol, Pyrex filter. d: 0.1 M ketone, 1 M sec-butylamine, 1:1
b
benzene:acetonitrile, uranium filter. Yield determined by NMR.
for 20 min and were cooled during the photolysis with a
circulating water bath. A Pyrex filter was used in most
cases although photolysis through a uranium glass filter
also was effective, vide infra. Reaction times varied from
5 to 24 h. The yields listed in Table 1 were calculated
from 1H NMR of the crude reaction mixture using an
internal standard. Products of the photolysis were also
isolated and characterized or compared to authentic
samples. Photoreductions of 1a ,b by benzene/2-propanol
(entries 2 and 4, Table 1) gave good yields of alcohol and
diphthalidyl 5, obtained as a mixture of dl and meso
forms in approximately equal amounts. Benzene was
used as a cosolvent due to the low solubility of the keto
esters in neat 2-propanol. Dimer 5 precipitated from
solution during the course of the reaction.
Photoreduction of 1c gave a poor yield of cholesterol
and 5 using 2-propanol as the H donor. However, the
photoreduction of 1a -e using cyclohexylamine as the
reducing agent gives alcohol in excellent yield in all cases
examined. The byproduct of the amine-mediated pho-
toreaction is not diphthalidyl 5 but 3-phenylphthalide,
6. Spectrometric (NMR) and chromatographic (GC, TLC)
analysis indicated that the only products of the photore-
duction of keto esters 1a -e are the alcohol and 6 after
aqueous workup and the removal of volatile products in
vacuo. Prior to removal of volatile materials, GC and
TLC indicated that cyclohexanone, presumably formed
by hydrolysis of cyclohexylimine during aqueous workup,
was also formed in the reaction. The reaction of 1a was
followed by GC and found to be complete in 5 h.
Reactions of the other esters of primary alcohols studied
were also found to be complete in 5 h as well. However,
Cyclohexylamine-N-d2 was prepared by the method of
Cohen.17 Photoreduction of 1g with cyclohexylamine-N-
d2 afforded 3-phenyl phthalide-3-d, 6, in 92% yield
(Scheme 2). Photolysis of 1g in benzene:acetonitrile but
with 10 equiv of 2-propanol instead of cyclohexylamine
produces 5 instead of 6 (Scheme 2). The phthalide dimer
5 was suspended in 1:1 C6D6:CD3CN with 10 equiv of
cyclohexylamine, and this mixture was followed by NMR
for 48 h. At the end of this period no change in the NMR
was observed.
The set of benzophenones shown in Table 2 was
photoreduced using cyclohexylamine in 1:1 benzene:
acetonitrile to test the effect of substituents on the
photoreaction. The products of the reactions listed in
1
Table 2 were determined by H NMR and MS. For the
(17) Cohen, S. G.; Chao, H. M. J . Am. Chem. Soc. 1968, 90, 165-
173.

2-Benzoylbenzoic Acid as a Photolabile Mask
J . Org. Chem., Vol. 61, No. 26, 1996 9457
Sch em e 2
Sch em e 3
Ta ble 2. Cycloh exyla m in e P h otor ed u ction P r od u cts
fr om Ben zop h en on es
compd
Ar
Ar′
8
9
7a a
7ba
7c
7d
7e
2-(MeO)C6H4
2,5-(MeO)2C6H3
4-(MeO)C6H4
4-(MeO)C6H4
4-(Cl)C6H4
2-(CO2Me)C6H4
2-(CO2Me)C6H4
Ph
4-(MeO)C6H4
4-(Cl)C6H4
>95b
>95
<5
<5
90
<5
<5
>95
>95
10
7f
4-(CO2Me)C6H4
3-(CO2Me)C6H4
2-(CO2Me)C6H4
3-(NO2)C6H4
4-(NO2)C6H4
Ph
Ph
Ph
Ph
Ph
Ph
Ph
4-MePh
>95
<5
7g
1ga
7h
7i
7j
7k
70
30
>95
<5
c
<5
10
<5
>95
benzpinacol, 9k , is 1:20. In 1:1 benzene:acetonitrile, the
product ratio of 8k :9k is 1:5. In wet acetonitrile there is
slightly more benzhydrol produced, the product ratio
being 1:4, but in formamide and DMSO the ratio of 8k :
9k is 1:1 and 4:3, respectively.
>95
90
4-MePh
a
b
The product obtained was the corresponding lactone. >95
indicates only one product observed by NMR. c Complex mixture.
Discu ssion
The experiments reported here suggest that 2-benzoyl-
benzoate esters are effective protecting groups for alco-
hols. Photoreduction proves to be efficient under condi-
tions where a good H atom donor (2-propanol) or electron
donor (amine) is present. It is clear that the photore-
duction of ketones 1a -e using amine-reducing agents
produces the benzhydrol analog 6 in an apparent 2-elec-
tron reduction, instead of the expected benzpinacol 5,
produced by 1-electron reduction of the ketone. The
possibility that this is caused by a dark reaction between
5 and the amine base is ruled out by the experiment in
which no change is observed in 5 after 2 days in the
presence of cyclohexylamine under the reaction condi-
tions. Cohen reports that the photoreduction of ben-
zophenone with cyclohexylamine in benzene affords
nearly quantitative yields of benzpinacol, 15 (Scheme 3),17
and it was our expectation that 2-benzoyl benzoate esters
would have similar chemistry when photoreduced. The
Cohen mechanism for formation of pinacol following
reduction with a primary amine is shown in Scheme 3.
Following the formation of a polar exciplex between the
amine and ketone, a proton transfer occurs to yield the
radical pair 12 and 13. Hydrogen atom transfer from
13 to ground-state phenone produces a second benzhydryl
radical, and coupling of two of these radicals gives the
pinacol 15.17,20-26
F igu r e 1. Photoreduction of substituted benzophenones with
cyclohexylamine.
most part the trend holds that benzophenones with an
electron-withdrawing group ortho or para to the ketone
(7a ,b,f,g) react to give reduction to benzhydrol. Those
benzophenones that do not have such a substituent
(7c,d ,h ,j,k ) react to give pinacol coupling products.
Ketone 7g gave significant amounts of both benzhydrol
and pinacol products. While the carboxyl in this com-
pound (7g) is not in conjugation with the ketone, it does
make the ring electron deficient. 4-Nitrobenzophenone,
7i, gave a complex mixture of products, a result that may
be due to photochemistry directly involving the nitro
substituent. Two of the benzophenones studied, 7a ,b,
have the possibility of reacting via abstraction of the
δ-hydrogens on the ortho methoxy substituent.18,19 This
is not observed under the photoreduction reaction condi-
tions. When 7a or 7b is photolyzed in dry benzene and
the reaction followed by NMR, little reaction occurs. The
sample is degraded over 48 h, and the solution becomes
yellow. However, in the case of 7a , some signals appear
that are consistent with the expected spirolactone. It is
not surprising that δ-hydrogen abstraction is not ob-
served in the presence of an electron or H atom donor as
reactions of these reducing agents with excited ketones
are fast relative to the abstraction of δ-hydrogens.18b
A mechanism consistent with our observations of the
photoreactions of substituted benzophenones is presented
in Scheme 4. The 2-13 radical pair is formed after
exciplex formation and proton transfer. These steps are
in accord with the initial steps of the mechanism pro-
posed for photoreduction of benzophenone by aliphatic
amines.17,20-26 The ketyl radical center for 2 is in direct
The solvent dependence of the photoreduction of 4,4′-
dimethylbenzophenone, 7k , with cyclohexylamine was
also studied. In benzene the ratio of benzhydrol, 8k , to
(18) (a) Wagner, P. J . Acc. Chem. Res. 1989, 22 (3), 83-91. (b)
Wagner, P. J .; Park, B. S. Org. Photochem. 1991, 11, 227-366.
(19) Wagner, P. J .; Meador, M.; Park, B. S. J . Am. Soc. Chem. 1990,
112, 5199-5211.
(20) Cohen, S. G.; Baumgarten, R. J . Am. Chem. Soc. 1967, 89,
3471-3475.
(21) Cohen, S. G.; Parola, A.; Parsons, G. H. Chem. Rev. 1972, 73,
141-161.

9458 J . Org. Chem., Vol. 61, No. 26, 1996
J ones et al.
Sch em e 4
toreduction of the dipyridyl ketone gives the correspond-
ing benzhydrol almost exclusively.32 Less electron poor
ketones give pinacols and radical coupling products. The
same effect is observed in the photochemical reduction
of olefins33 in cases using triethylamine30 as the reducing
agent. Pac’s suggestion is that electron-withdrawing
groups conjugated to the ketone lower the reduction
potential of the ketyl radical to the point that it can be
reduced by the pyridinal radical. In the absence of an
electron-withdrawing group, the reduction potential of
the ketyl radical is too high for the electron transfer to
be energetically feasible. Whitten has also suggested a
double electron transfer mechanism in the photoreduc-
tion of anthraquinone with amine donors.35 The second
electron transfer results in an anthrahydroquinone/
iminium ion pair in equilibrium with an anthrahydro-
quinone/R-amino radical pair. The position of this equi-
librium is solvent dependent. The solvent dependence
of radical pair/ion pair equilibria has also been elegantly
demonstrated by Arnett.36-38
Another possibility for the formation of benzhydrol/
imine products 6 and 14 is one involving disproportion-
ation of the 2-13 radical pair. It is unclear why the
2-13 pair would disproportionate while the 12-13 pair
does not, but polar effects in the transition state may play
a role.39
If either direct charge transfer or a polar effect in
disproportionation operate, any benzophenone with an
electron-withdrawing group conjugated to the ketone
should give benzhydrol when photoreduced using an
amine-reducing agent. This is precisely what we observe
in the photoreactions of the series of benzophenones
presented in Table 2. Those phenones that have electron-
withdrawing groups substituted ortho or para to the
carbonyl give benzhydrol product, presumably from an
intermediate ion pair or polar transition state. Further
evidence for these mechanisms is found in the solvent
dependence of the photoreduction of 4,4′-dimethylben-
zophenone with cyclohexylamine. Polar solvents that
stabilize polar intermediates provide more benzhydrol,
while amine-promoted photoreduction of this benzophe-
none in nonpolar solvent gives the pinacol product.
Clearly the substituent and solvent effects on product
conjugation with the ortho carboxyl group. The electron-
withdrawing character of the carboxyl group should
increase the ability of the radical to accept an electron,
lowering its reduction potential. Electron transfer from
the easily oxidized R-amino radical 13 gives carbanion
16 and iminium ion 17. Protonation of 16 by 17 gives
the products in a manner that is consistent with the
deuterium-labeling studies shown in Scheme 2.
Electrochemical studies have shown that R-amino
radicals, such as 13, are easily and quickly oxidized to
the iminium ion.27,28 Polaragraphic data also demon-
strate that ketyl radicals are reduced easier than the
corresponding ketone.14 Pac and Peters have, in fact,
both reported electron transfer from carbon-centered
radicals in conjugation with nitrogen lone pairs to ketyl
radicals. Peters has observed the hydroxydiphenyl-
methane anion by flash photolysis in the photoreduction
of benzophenone using N-methylacridan as the reducing
agent.29 Pac has shown that the fate of the intermediate
ketyl radical is dependent on whether or not electron-
withdrawing groups are conjugated to the radical
center.30-34 As an example, the Ru(bpy)3-sensitized pho-
(31) Pac, C.; Miyauchi, Y.; Ishitani, O.; Ihama, M.; Yasuda, M.;
Sakurai, H. J . Org. Chem. 1984, 49, 26-34.
(32) Pac, C.; Takamuku, S.; Yanagida, S.; Ishitani, O. J . Org. Chem.
1987, 52, 2790-2796.
(33) Pac, C.; Miyauchi, Y.; Ihama, M.; Ishitani, O. J . Chem. Soc.,
Perkin Trans. I 1985, 1527-1531.
(34) Pac, C.; Ihama, M.; Yasuda, M.; Miyauchi, Y.; Sakurai, H. J .
Am. Chem. Soc. 1981, 103, 6495-6497.
(35) Ci, X.; Silveira da Silva, R.; Nicodem, D.; Whitten, D. G. J . Am.
Chem. Soc. 1989, 111, 1337-1343.
(36) Arnett, E. M.; Molter, K. E.; Marchot, E. C.; Donovan, W. H.;
Smith, P. J . Am. Chem. Soc. 1987, 109, 3788-3789.
(37) Arnett, E. M.; Molter, K. E.; Troughton, E. B. J . Am. Chem.
Soc. 1984, 106, 6726-6735.
(22) Inbar, S.; Linschitz, H.; Cohen, S. G. J . Am. Chem. Soc. 1981,
103, 1048-1054.
(23) Peters, K. S.; Simon, J . D. J . Am. Chem. Soc. 1981, 103, 6403-
(38) Pienta, N. J .; Kessler, R. J .; Peters, K. S.; O’Driscoll, E. D.;
Arnett, E. M.; Molter, K. E. J . Am. Chem. Soc. 1991, 113, 3773-3781.
(39) A and B are two resonance structures for the disproportionation
transitions state of radical pairs 2-13 and 12-13. In the case of 2-13
the electron-withdrawing character of Ar will increase the contribution
from structure B relative to its contribution in 12-13. The decrease
in energy of the overall transition state due to a polar contribution
may explain why 2-13 appears to disproportionate while 12-13 does
not.
6406.
(24) Peters, K. S.; Simon, J . D. J . Am. Chem. Soc. 1982, 104, 6542-
6547.
(25) Manring, L. E.; Peters, K. S. J . Am. Chem. Soc. 1985, 107,
6452-6458.
(26) (a) Shaefer, C. S.; Peters, K. S. J . Am. Chem. Soc. 1980, 102,
7567-7568. (b) Arimitsu, S.; Masuhara, H.; Mataga, N.; Tsubomura,
H. J . Phys. Chem. 1975, 79, 155-1259.
(27) Smith, P. J .; Mann, C. K. J . Org. Chem. 1969, 34, 1821-1826.
(28) Barnes, K. K.; Mann, C. K. J . Org. Chem. 1967, 32, 1474-1479.
(29) Peters, K. S.; Pang, E.; Rudzki, J . J . Am. Chem. Soc. 1982, 104,
5535-5537.
(30) Yanagida, S.; Pac, C.; Ishitani, O.; Shiragami, T.; Kabumoto,
A.; Shibata, T. J . Phys. Chem. 1990, 94, 2068-2076.

2-Benzoylbenzoic Acid as a Photolabile Mask
J . Org. Chem., Vol. 61, No. 26, 1996 9459
distribution reflect some amount of charge transfer in
the steps leading to benzhydrol.
The results obtained with thio ester 1f are more
complex than those in the cases of oxo esters 1a -e. The
reduction of the ketone undoubtedly leads to thiol and
either 5 or 6 as in the cases of the oxo esters. Addition-
ally, our experiments show that, under the reaction
conditions, photolysis of 1f in the absence of reducing
agent leads to 5 and the corresponding disulfide. The
disulfide can react under these conditions to give thiol.
All of these reactions are, most likely, occurring simul-
taneously, leading to the observed mixture of products.
The question of when lactonization actually occurs is
not addressed by any of these experiments, and it should
be noted that it is possible that lactonization could occur
at several steps in the reaction sequence. The ketyl
radical anion (analogous to 10) may lactonize to afford
an alkoxide anion and the 3-phenylphthalidyl radical.
Alternatively, the ketyl radical may cyclize to give the
alcohol and 3-phenylphthalidyl radical, or lactonization
may not occur until the benzhydrol has been formed, as
shown in Scheme 4.
F igu r e 2. Synthesis of benzoylbenzoate esters, 1a -e and thio
ester 1f.
Ta ble 3. Isola ted Yield s of Ben zoylben zoa te Ester s 1a -f
XH
product
yield (%)
n-C12H25OH
c-C12H23OH
cholesterol
geraniol
2′,3′-isopropylideneuridine
n-C12H25SH
1a
1b
1c
1d
1e
1f
76
50
67
63
79
76
Dod ecyl 2-ben zoylben zoa te (1a ): yield 2.68 g (6.8 mmol,
76%); obtained as a colorless oil via flash column chromatog-
raphy (SiO2, 9:1 hexane:EtOAc); IR (CCl4) 1740, 1680 cm-1
;
1H NMR (CDCl3, ppm) 8.09 (d, 1H, J ) 10 Hz), 7.80 (d, 2H, J
) 10 Hz), 7.60 (m, 3H), 7.45 (m, 3H), 4.01 (t, 2H, J ) 7.5 Hz),
1.2 (m, 20H), 0.85 (t, 3H, J ) 6.3 Hz); 13C NMR (CDCl3, ppm)
196.9, 165.9, 141.6, 137.0, 133.0, 132.2, 130.1, 129.5, 129.4,
128.4, 127.6, 65.7, 31.9, 29.6, 29.5, 29.4, 29.3, 29.1, 28.1, 25.8,
22.7, 14.1; GC/CIMS (CH4/NH3(g)) m/ z 395 (MH+). Anal.
Calcd for C26H34O3: C, 79.15; H, 8.69. Found: C, 79.00; H,
8.79.
We conclude that o-benzoylbenzoate esters serve as
photolabile protecting groups for alcohols. Primary and
secondary alcohols as well as thiols can be easily masked
by the formation of the corresponding 2-benzoylbenzoate
esters using very inexpensive reagents. These esters can
be converted in high yield to the alcohol/thiol and
3-phenylphthalide (6), under electron transfer conditions,
or its dimer (5), in the presence of hydrogen donors.
Cyclod od eca n yl 2-ben zoylben zoa te (1b): yield 1.77 g
(4.52 mmol, 50%); obtained as a white solid (SiO2, 7:1 hexane:
EtOAc); mp 102-103 °C; IR (Nujol) 1710, 1669 cm-1; 1H NMR
(CHCl3, ppm) 8.08 (dd, 1H, J ) 1.2, 7.7 Hz), 7.77 (m, 2H), 7.58
(m, 3H), 7.43 (d, 1H, J ) 7.7 Hz), 7.37 (m, 2H), 5.00 (m, 1H),
1.40 (m, 4H), 1.22 (m, 18H); 13C NMR (CHCl3, ppm) 196.7,
165.3, 141.4, 137.1, 133.1, 132.3, 130.3, 129.7, 129.5, 129.1,
128.5, 128.2, 127.2, 74.3, 28.2, 24.1, 24.0, 23.1, 23.0, 20.4; MS/
FAB m/ z 393.25 (MH+). Anal. Calcd for C26H32O3: C, 79.56;
H, 8.22. Found: C, 79.68; H, 8.17.
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. Melting points are uncorrected.
The following chemicals were obtained from commercial
sources and used as received: 2-benzoylbenzoic acid, 3-ben-
zoylbenzoic acid, 4-benzoylbenzoic acid, DCC, pyridine, 1-dode-
canol, cyclododecanol, cholesterol, geraniol, 2′,3′-isopropylide-
neuridine, 1-dodecanethiol, cyclohexylamine, sec-butylamine,
4-methoxybenzophenone, 4,4′-dimethoxybenzophenone, 4,4′-
dichlorobenzophenone, 3-nitrobenzophenone, 4-nitrobenzophe-
none, benzophenone, 4,4′-dimethylbenzophenone, dimethoxy-
benzene, trifluoroacetic anhydride, monomethyl phthalate, and
2-bromoanisole. Methyl 2-benzoylbenzoate,13 methyl 3-ben-
zoylbenzoate,40 methyl 4-benzoylbenzoate,35 3-phenylphtha-
lide,13 and 3,3′-diphenylbiphthalidyl41 were prepared by lit-
erature methods. Reactions were run in flame-dried glassware
under argon. Gas chromatography was performed with flame
ionization detection, coupled to a digital integrator (15 m SPB-
5, 0.20 mm i.d.). Photolyses were carried out with a 450 W
medium pressure mercury lamp. Mass spectra were obtained
using methane/ammonia (2 × 10-4 atm) for chemical ionization
mass spectra.
Gen er a l P r oced u r e for th e P r otection of Alcoh ols a n d
Th iols. To a solution of 2-benzoylbenzoic acid (2.26 g, 10
mmol), alcohol or thiol (0.9 equiv, 9 mmol), and DMAP (0.244
g, 2 mmol) in pyridine (15 mL) was added DCC (2.06 g, 10
mmol) in one portion. The reaction mixture was stirred at 25
°C under argon for 48 h. The reaction mixture was partitioned
between 1 M HCl (200 mL) and EtOAc (200 mL). The layers
were separated, and the aqueous layer was extracted with 1
× 150 mL of EtOAc. The combined organic layers were
washed with 1 × 200 mL of 1 M HCl and 1 × 100 mL of brine;
dried (MgSO4), and concentrated in vacuo. The desired
product was isolated via column chromatography (SiO2, hex-
ane/EtOAc) (Table 3).
Ch olester yl 2-ben zoylben zoa te (1c): yield 3.60 g (6.1
mmol, 67%); obtained as a white solid via flash column
chromatography (SiO2, 9:1 hexane:EtOAc); mp 119-119.5 °C;
1
IR (Nujol) 1719, 1673 cm-1; H NMR (CHCl3, ppm) 8.10 (dd,
1H, J ) 1.26, 7.74 Hz), 7.78 (m, 2H), 7.60 (m, 3H), 7.41 (m,
3H), 5.26 (m, 1H), 4.59 (m, 1H), 0.94-2.08 (m, 30H), 0.88 (m,
12H), 0.66 (s, 3H); 13C NMR (CHCl3, ppm) 196.84, 165.06,
141.47, 139.28, 137.21, 133.11, 132.32, 130.23, 129.47, 129.41,
129.34, 128.47, 127.48, 122.65, 75.54, 56.56, 56.04, 49.84,
42.23, 42.20, 39.62, 39.46, 37.17, 36.73, 36.47, 36.12, 35.74,
34.89, 31.80, 31.74, 28.18, 27.98, 26.89, 25.41, 24.22, 23.79,
22.81, 22.54, 20.92, 19.18, 19.14, 18.67, 11.82, 11.76; MS/FAB
m/ z 595.18 (MH+). Anal. Calcd for C40H54O3: C, 82.78; H,
9.15. Found: C, 82.67; H, 9.23.
Ger a n yl 2-ben zoylben zoa te (1d ): yield 2.3 g, (6.35 mmol,
63%); obtained as a colorless oil via flash column chromatog-
raphy (SiO2, 6:1 hexane:EtOAc); IR (neat) 3062, 2964, 2854,
2118, 1720, 1675, 1597, 1581 cm-1; 1H NMR (CDCl3, ppm) 8.06
(d, 1H, J ) 6.4 Hz), 7.74 (d, 2H, J ) 7.2 Hz), 7.56 (m, 3H),
7.42 (m, 3H), 5.07 (m, 2H), 4.53 (d, 2H, J ) 8.5 Hz), 1.97 (m,
4H), 1.67 (s, 3H), 1.59 (s, 3H), 1.55 (s, 3H); 13C NMR (CDCl3,
ppm) 142.6, 141.6, 133.0, 132.3, 131.8, 130.1, 129.5, 129.4,
128.3, 127.6, 123.7, 118.0, 117.4, 62.3, 39.4, 36.1, 33.7, 27.0,
26.2, 25.6, 24.8, 17.7, 16.3; MS/FAB m/ z 363.3 (MH+). Anal.
Calcd for C24H26O3: C, 79.53; H, 7.23. Found: C, 79.26; H,
7.33.
5′-((2-Ben zoyl)ben zoyl)-2′,3′-isopr opyliden eu r idin e (1e):
yield 3.5 g (7.11 mmol, 79%); obtained as a white solid via flash
column chromatography (SiO2, 1:1 EtOAc:hexane); mp 90-
93 °C; IR (Nujol) 1717, 1676, 1671, 1595, 1578 cm-1; 1H NMR
(CDCl3, ppm) 8.17 (brs, 1H), 8.03 (dd, 1H, J ) 1.2, 7.6 Hz),
7.76 (d, 2H, J ) 6.8 Hz), 7.62 (m, 3H), 7.43 (m, 3H), 7.23 (t,
1H, J ) 8.0 Hz), 5.67 (d, 1H, J ) 2.4 Hz), 5.59 (dd, 1H, J )
2.4, 8.0 Hz), 4.80 (m, 1H), 4.62 (m, 1H), 4.41 (q, 1H, J ) 8.8
Hz), 4.21 (m, 2H), 1.54 (s, 3H), 1.31 (s, 3H); 13C NMR (CDCl3,
ppm) 203.8, 194.7, 178.4, 174.8, 162.4, 149.6, 141.4, 136.7,
(40) Wagner, P. J .; Siebert, E. J . J . Am. Chem. Soc. 1981, 103, 7329-
7335.
(41) Bhatt, M. V.; Kamath, K. M.; Ravindranathan, M. J . Chem.
Soc. 1971, 3344-3347.

9460 J . Org. Chem., Vol. 61, No. 26, 1996
J ones et al.
133.4, 132.6, 130.3, 129.9, 129.5, 128.6, 128.0, 114.8, 102.5,
93.3, 84.3, 80.5, 64.5, 27.0, 25.2; MS/FAB m/ z 493.2 (MH+).
Anal. Calcd for C26H24O8N2: C, 63.41; H, 4.91; N, 5.69.
Found: C, 63.01; H, 5.30; N, 5.67.
for 20 min. The solutions were then irradiated for the
indicated time, with continuous stirring and while under a
positive pressure of argon, with a medium pressure 450 W Hg
lamp at a distance of 5 cm. In the cases of photoreduction
using amines, the reaction mixtures were partitioned between
1 M HCl and EtOAc. The aqueous layer was extracted twice
more with EtOAc, and the combined organic layers were
washed once with brine, dried over MgSO4 and concentrated
in vacuo. In cases using 2-propanol the solvent was removed
in vacuo. The product mixtures were analyzed by GC, TLC,
and NMR. The crude material was dissolved in CDCl3 and
an internal standard added (mesitylene). Product yields were
determined by integrating the appropriate product signals
relative to the NMR signal from a known amount of the added
standard. Where possible both the aromatic and aliphatic
NMR signals from the standard were used and the results
averaged. As many signals as possible from the products were
also used. The relaxation delay was set to 3 s.
Dodecyl 2-benzoylbenzoate (1.21 g, 3.07 mmol) was irradi-
ated for 6 h as a 0.1 M solution in 1:1 benzene:acetonitrile
with cyclohexylamine (3.5 mL, 30.7 mmol). The crude mixture
was worked up as indicated above, and mesitylene (0.14 mL,
1.0 mmol) was added. GC and NMR indicated only two
products which were identified as 3-phenylphthalide and
1-dodecanol by coinjection with authentic samples. From the
NMR data the yields obtained are as follows: 3-phenylphtha-
lide, 95%; 1-dodecanol, 95%. The crude mixture was purified
via flash chromatography (silica, 9:1 hexane:EtOAc eluent) to
afford 1-dodecanol (510 mg, 2.74 mmol, 90%) and 3-phe-
nylphthalide (475 mg, 2.27 mmol, 74%).
Dodecyl 2-benzoylbenzoate (0.245 g, 0.62 mmol) was irradi-
ated as a 0.01 M solution in 1:1 benzene:2-propanol. After 5
h the solvent was removed in vacuo and redissolved in CDCl3,
and mesitylene (0.025 mL, 0.179 mmol) was added. TLC and
NMR indicated only two products which were identified as
1-dodecanol and the dimer of 3-phenylphthalide. From the
NMR the yields obtained are as follows: 3-phenylphthalide
dimer, 95%; 1-dodecanol, 95%.
Cyclododecanyl 2-benzoylbenzoate (0.150 g, 0.38 mmol) was
irradiated for 18 h as a 0.1 M solution in 1:1 benzene:
acetonitrile with cyclohexylamine (0.43 mL, 3.8 mmol). The
reaction mixture was worked up as indicated above, and
mesitylene (0.015 mL, 0.108 mmol) was added. GC and NMR
indicated only two products which were identified as 3-phe-
nylphthalide and cyclododecanol by coinjection with authentic
samples. From the NMR data the yields obtained are as
follows: 3-phenylphthalide, 80%; cyclododecanol, 85%.
Cyclododecanyl 2-benzoylbenzoate (0.100 g, 0.26 mmol) was
irradiated as a 0.01 M solution in 1:1 benzene:2-propanol. After
18 h the solvent was removed in vacuo and the residue
Dod ecyl 2-ben zoylben zen eth ioa te (1f): yield 2.68 g (6.8
mmol, 76%); obtained as a colorless oil via flash chromatog-
raphy (SiO2, 9:1 hexane:EtOAc); IR (neat) 3061, 2924, 2852,
1777, 1669, 1597 cm-1; 1H NMR (CHCl3, ppm) 7.98 (dd, 1H, J
) 1.33, 7.53 Hz) 7.75 (m, 2H), 7.58 (m, 3H), 7.41 (m, 3H), 2.86
(t, 2H, J ) 7.27 Hz), 1.43 (m, 2H), 1.26 (m, 17H), 0.88 (t, 3H,
J ) 6.72 Hz); 13C NMR (CHCl3, ppm) 196.9, 192.1, 139.2, 137.2,
136.9, 132.9, 129.8, 129.4, 128.3, 128.2, 128.1, 34.9, 31.8, 29.6,
29.5, 29.4, 29.3, 29.2, 29.1, 29.0, 28.7, 22.6, 14.1; MS/FAB m/ z
411.38 (MH+). Anal. Calcd for C26H34O2S: C, 76.05; H, 8.35;
S, 7.81. Found: C, 76.10; H, 8.41; S, 7.73.
Meth yl 2-(2-Meth oxyben zoyl)ben zoa te (7a ). One-fourth
of a 200 mL solution of 2-bromoanisole (7.5 g, 40 mmol) in
THF was added to magnesium turnings (1.05 g, 42 mmol)
which had been ground and dried under high vacuum. When
the THF began to reflux the remainder of the solution was
added dropwise so that reflux was maintained. Once addition
was complete the solution was heated to reflux with a heating
mantle. Heating was maintained for 16 h. The yellow solution
was allowed to cool to ambient temperature and then cooled
further to -78 °C with a dry ice/acetone bath. The solution
remained homogeneous throughout this process. A solution
of monomethylphthaloyl chloride (from monomethyl phthalate
and thionyl chloride) in THF (0.27 M, 150 mL) was added
dropwise over 45 min. The resulting mixture was maintained
at -78 °C for 1 h and then allowed to warm to ambient
temperature with the cold bath. After 7 h the reaction mixture
was poured into 300 mL of 1 N HCl/ice and 300 mL of EtOAc.
This mixture was stirred for 15 min, and the layers were
separated. The aqueous layer was extracted once more with
100 mL of EtOAc, and the combined organic layers were
washed with 3 × 200 mL of saturated sodium bicarbonate, 1
× 200 mL of 1 N HCl, and 1 × 300 mL of brine. The organic
layer was then dried (MgSO4) and concentrated in vacuo to
yield a viscous yellow oil (11.01 g). The oil was dissolved in a
small amount of ether, and hexane was added until a precipi-
tate had formed. This solid was recrystallized from hot ethanol
to afford the desired product as white needles (3.75 g, 13.9
mmol, 35%): mp 97-98 °C; IR (Nujol) 2920, 2854, 1714, 1648
1
cm-1; H NMR (CHCl3, ppm) 3.60 (s, 3H), 3.62 (s, 3H), 6.92
(d, 1H, J ) 8.4 Hz), 7.01 (td, 1H, J ) 0.9, 7.5 Hz), 7.39 (dd,
1H, J ) 1.2, 7.3 Hz), 7.59 (m, 3H), 7.72 (dd, 1H, J ) 1.9, 7.8
Hz), 7.89 (dd, 1H, J ) 1.5, 7.1 Hz); 13C NMR (CHCl3, ppm)
197.5, 167.1, 159.2, 143.7, 134.4, 131.8, 131.4, 129.5, 127.9,
127.8, 127.6, 120.4, 112.1, 110.2, 56.0, 52.1; GC/CIMS (CH4/
NH3(g)) m/ z 270.95 (MH+). Anal. Calcd for C16H14O4: C,
71.10; H 5.22. Found: C, 71.10; H, 5.25.
redissolved in CDCl3
. Mesitylene (0.025 mL, 0.18 mmol) was
added. TLC and NMR showed that the main products were
cyclododecanol and the dimer of 3-phenylphthalide. There was
a trace of 3-phenylphthalide present as well. From the NMR
data the yields obtained are as follows: 3-phenylphthalide
dimer, 85%; cyclododecanol, 85%.
Cholesteryl 2-benzoylbenzoate (0.597 g, 1.0 mmol) was
irradiated as a 0.1 M solution in 1:1 benzene:acetonitrile with
cyclohexylamine (1.14 mL, 10 mmol). The reaction mixture
was worked up as indicated above, and mesitylene (0.045 mL,
0.324 mmol) was added. NMR indicated only two products
which were identified as 3-phenylphthalide and cholesterol by
TLC. From the NMR data the yields obtained are as follows:
3-phenylphthalide, 95%; cholesterol, 100%.
Cholesteryl 2-benzoylbenzoate (0.241 g, 0.41 mmol) was
irradiated as a 0.01 M solution in 1:1 benzene:2-propanol. After
18 h the solvent was removed in vacuo and the residue
redissolved in CDCl3. Mesitylene (0.02 mL, 0.144 mmol) was
added. TLC and NMR showed a mixture of products. Two
were identified as cholesterol and the dimer of 3-phenylphtha-
lide. From the NMR data the yields obtained are as follows:
cholesterol, 20%; 3-phenylphthalide dimer, 5%.
Geranyl 2-benzoylbenzoate (0.610 g, 1.68 mmol) was irradi-
ated for 5 h as a 0.1 M solution in 1:1 benzene:acetonitrile
with cyclohexylamine (1.92 mL, 16.8 mmol). The reaction
mixture was worked up as indicated above, and mesitylene
Meth yl 2-(2,5-Dim eth oxyben zoyl)ben zoate (7b). Monom-
ethyl phthalate (2.6 g, 14.4 mmol) was dissolved in trifluoro-
acetic anhydride (8 mL), and the solution was heated to 55 °C
for 15 min. Hydroquinone dimethyl ether (2.0 g, 14.4 mmol)
was added in one portion and the resulting solution main-
tained at 55 °C for 6 h. After cooling the solution was poured
into 100 mL of saturated sodium bicarbonate and extracted
with 2 × 50 mL of ether. The combined organic extracts were
washed with 1 × 100 mL of saturated sodium bicarbonate and
1 × 100 mL of brine, dried (MgSO4), and concentrated in vacuo
to afford a yellow oil. This oil was purified via flash chroma-
tography (SiO2, 10-35% EtOAc in hexane) to afford the desired
product as a white solid (3.36 g, 11.2 mmol, 78%): mp 85-86
1
°C; IR (Nujol) 2924, 2854, 1713, 1652 cm-1; H NMR (CHCl3,
ppm) 3.48 (s, 3H), 3.65 (s, 3H), 3.81 (s, 3H), 6.86 (d, 1H, J )
9.0 Hz), 7.05 (dd, 1H, J ) 3.2, 9.0 Hz), 7.36 (m, 2H), 7.52 (m,
2H), 7.92 (d, 1H, J ) 7.7 Hz); GC/CIMS (CH4/NH3(g)) m/ z
300.95 (MH+). Anal. Calcd for C17H16O5: C, 67.99; H, 5.37.
Found: C, 67.92; H, 5.42.
Gen er a l P r oced u r e for th e P h otoch em ica l Red u ction
of Keto Ester s 1a -f. Solutions of ketone in the desired
solvent system and any other desired reagent were prepared
and placed in a jacketed Pyrex flask, equipped with stirbar
and rubber septum. The solutions were cooled by a circulating
water bath and degassed by bubbling dry argon through them

2-Benzoylbenzoic Acid as a Photolabile Mask
J . Org. Chem., Vol. 61, No. 26, 1996 9461
(0.078 mL, 0.56 mmol) was added. GC and NMR indicated
the presence of only 3-phenylphthalide and geraniol. From
the NMR data the yields obtained are as follows: geraniol,
85%; 3-phenylphthalide, 80%.
Geranyl 2-benzoylbenzoate (0.905 g, 2.5 mmol) was irradi-
ated as a 0.1 M solution in 1:1 benzene:acetonitrile with sec-
butylamine (2.5 mL, 25 mmol). After 5 h the solvent was
removed in vacuo and the residue redissolved in CDCl3.
Mesitylene (0.071 mL, 0.51 mmol) was added to the solution.
GC and NMR indicated the presence of only 3-phenylphthalide
and geraniol. From the NMR data the yields obtained are as
follows: geraniol, 90%; 3-phenylphthalide, 70%.
5′-((2-Benzoyl)benzoyl)-2′,3′-isopropylideneuridine (0.246 g,
0.50 mmol) was irradiated for 5 h as a 0.1 M solution in 1:1
benzene:acetonitrile with sec-butylamine (0.5 mL, 5 mmol).
The reaction mixture was worked up as indicated above, and
mesitylene (0.047 mL, 0.33 mmol) was added. NMR and TLC
showed 3-phenylphthalide and 2′,3′-isopropylideneuridine to
be the major products along with a number of unidentified
minor products. From the NMR data the yields obtained are
as follows: 3-phenylphthalide, 75%; 2′,3′-isopropylideneuri-
dine, 60%.
5′-((2-Benzoyl)benzoyl)-2′,3′-isopropylideneuridine (0.246 g,
0.50 mmol) was irradiated for 18 h as a 0.1 M solution in 1:1
benzene:acetonitrile with sec-butylamine (0.5 mL, 5 mmol). A
uranium glass filter was employed to eliminate radiation with
λ < 350 nm. The reaction mixture was worked up as indicated
above, and mesitylene (0.047 mL, 0.33 mmol) was added.
NMR and TLC showed 3-phenylphthalide and 2′,3′-isopropy-
lideneuridine to be the only products. From the NMR data
the yields obtained are as follows: 3-phenylphthalide, 80%;
2′,3′-isopropylideneuridine, 90%.
Dodecyl 2-benzoylbenzenethioate (0.434 g, 1.06 mmol) was
irradiated as a 0.1 M solution in 1:1 benzene:acetonitrile with
cyclohexylamine (1.25 mL, 11 mmol). After 5.5 h a precipitate
had formed. The solvent was removed in vacuo, and the
residue was dissolved in hexane, chilled, and filtered. The
precipitate was identified as 3-phenylphthalidyl dimer (0.100
g, 0.24 mmol, 45%) by NMR. The filtrate was then worked
up as usual, and mesitylene (0.04 mL, 0.288 mmol) was added.
Four products were identified by GC and coinjection with
authentic samples. They were 3-phenylphthalide, 3-phe-
nylphthalidyl dimer, 1-dodecanethiol, and dodecane disulfide.
From the NMR data the yields obtained are as follows:
3-phenylphthalide, 20%; 1-dodecanethiol, 60%; dodecane dis-
ulfide, 20%. It was impossible to determine the amount of
3-phenylphthalidyl dimer remaining in solution.
P r ep a r a t ion a n d P h ot olysis of sa m p les for Isot op e,
Su bstitu en t, a n d Solven t Effect Stu d ies. Samples were
prepared by adding substrate, solvent, and amine to a Pyrex
test tube with screw-cap top. Argon was bubbled through the
resulting solution for 20 min, the test tube sealed, and the
sample irradiated at a distance of 5 cm from the lamp while
submerged in a water bath. At the end of the photolysis period
the solution was poured into 2 M HCl and EtOAc. The layers
were separated, and the organic layer was dried over sodium
sulfate. Solvent was removed in vacuo. The crude reaction
mixture was then analyzed by GC, TLC, and NMR to obtain
the required data. For the reactions involved in the substitu-
ent effect study, the products were not isolated.
J O961638P