Unique solvent effect on photochemistry of ortho-alkylphenacyl benzoates

Article abstract of DOI:10.1016/j.tetlet.2010.01.046

Photolysis of 2,4,6-trialkylphenacyl benzoates gives not only the corresponding indanones and benzoic acid, but also the corresponding benzocyclobutenols (CBs), which are also detected in the photolysis of mono-alkylphenacyl benzoates for the first time.

Full text of DOI:10.1016/j.tetlet.2010.01.046

Tetrahedron Letters 51 (2010) 1512–1516
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Unique solvent effect on photochemistry of ortho-alkylphenacyl benzoates
*
Bong Ser Park , Hyuk Jun Ryu
Department of Chemistry, Dongguk University, Seoul 100-715, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
Photolysis of 2,4,6-trialkylphenacyl benzoates gives not only the corresponding indanones and benzoic
acid, but also the corresponding benzocyclobutenols (CBs), which are also detected in the photolysis of
mono-alkylphenacyl benzoates for the first time. The product selectivity was heavily dependent upon
solvents and o-alkyl group. H-bonding acceptor solvents strongly favor the formation of the CB. As the
size of the o-alkyl group increases, the relative amount of the CB increases.
Received 30 November 2009
Revised 8 January 2010
Accepted 14 January 2010
Available online 18 January 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Dedicated to Professor Peter J. Wagner who
died of a cancer on August 6th 2009
Keywords:
Photochemistry
Solvent effect
Benzocyclobutenol
Hydrogen abstraction
The photochemistry of ortho-alkylphenyl ketones has received
much attention not only in mechanistic studies,1 but also in syn-
thetic application.2 The mechanism is now well established: intra-
molecular hydrogen abstraction from the triplet excited state
forms a short-lived triplet enol, which decays to ground state iso-
mers with Z and E configurations at the enol carbon. The former re-
verts to the starting ketone via fast 1,5-sigmatropic internal return,
while the latter cyclizes to form benzocyclobutenols or undergoes
catalyzed reversion to the reactant ketone. The photocyclization to
benzocyclobutenols occurs efficiently in benzene, but is inefficient
or even not detected in methanol.3 The reaction is known to be
readily quenched by acids and bases.4
It is interesting to note that the formation of benzocyclobute-
nols has not been reported in all of their studies on the photochem-
istry of
a-substituted o-alkylphenyl ketones, even though the
initial stage of the reaction seems to share a common intermediate
with photoreactions of the unsubstituted o-alkylphenyl ketones. In
our continuing efforts to find structure–reactivity relationship in
ketone photochemistry,9 we recently had a chance to investigate
photochemical behaviors of several ortho-alkylphenyl ketones hav-
ing a benzoate leaving group at the alpha position and found that
such system do indeed form benzocyclobutenols and shows an
interesting solvent effect. Here we would like to report our preli-
minary findings.
If the ortho-alkylphenyl ketones have a leaving group X such as
chloride, carboxylates, sulfonates or phosphates at alpha position
to carbonyl, indanone formation via HX elimination becomes the
major reaction route upon photolysis, which is one of the recently
developed photocages.5 The reaction mechanism seems to vary
depending on the leaving group and the reaction medium (Scheme
1). When the leaving group was Cl, Klan and Wirz showed that the
indanone was formed from the ground state E-enol in benzene but
from Z-enol in methanol.6 If the relatively poor leaving group car-
boxylates are used, the reaction occurs through the E-enol regard-
less of the solvent.7 In the case of the sulphonates, Wessig
proposed the intermediacy of 1,5-triplet biradical formed by fast
HX elimination from the initially formed 1,4-triplet biradical.8
At the beginning of this research, we were interested in steric
effects on photoinduced indanone formation from o-alkylphenacyl
benzoates. Accordingly, the trialkyl-substituted ketones, 1–3, (in
Table 1) were prepared using routine synthetic procedures: Fri-
edel–Crafts acylation of the corresponding benzene, alpha bromin-
ation, followed by substitution by benzoyloxy group.10 The purified
ketones were irradiated in benzene or methanol (typically 0.02 M)
using the Pyrex filtered UV light of 450 W Hanovia medium pres-
sure mercury arc lamp until all the starting material had disap-
peared. The resulting photoproducts were isolated by routine
purification methods using silica gel column chromatography.
Each of the isolated products was identified by analyzing its
spectroscopic data. The photolysis was also done in various sol-
vents on an NMR sample scale by attaching the degassed sample
solution by an immersion well equipped with the Hanovia medium
pressure mercury arc lamp in order to monitor the reaction more
closely.
* Corresponding author. Tel.: +82 2 2260 3223.
E-mail address: parkbs@dongguk.edu (B.S. Park).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.01.046

B. S. Park, H. J. Ryu / Tetrahedron Letters 51 (2010) 1512–1516
1513
OH
X
O
O
X
ν
h
- HX
T1
S1
X = OMs
X
OH
O
X
OH
X = Cl
MeOH
X = OCOR
MeOH
Z
E
X = Cl, OCOR
Benzene
MeO
O
O
O
+
Scheme 1.
Table 1
Product ratios (IN–CB) of photolysis of 1–3 in various solvents
R1
R1
R1
O
R1
O
OH
R3
OH
R2
OCOPh
OCOPh
ν
h
OCOPh
+
+
R2
R3
R1
R2 R1
R3
R1
R1
R1
+
Benzoic acid
1 : R = Me, R2 = R3 = H
2 : R1 = Et, R2 = Me, R = H
3 : R1 = i-Pr, R2 = R =3Me
CB-E
IN
CB-Z
1
3
Compound
Solvent
Ratio (IN:CB)
1
Benzene
MeOH
CH3CN
Acetone
DMSO
3.7:1a (78.5%:21.5%)
1:14.3 (6.5%:93.5%)
1:2.9 (25.7%:74.3%)
1:2.4 (29.1%:70.9%)
CB only
2
3
Benzene
MeOH
CH3CN
Acetone
DMSO
1:1 (50%:50%b)
CB only (E:Z = 3.4:1)
CB only (E:Z = 2.2:1)ꢀ
CB only (E:Z = 3.7:1)ꢀ
CB only (E:Z = 3.0:1)
Benzene
MeOH
1:2.8 (26.4%:73.6%)
CB only
CH3CN
Acetone
DMSO
CB only
CB only
CB only
a
Normalized to 100%. Material balances of photoproducts including benzoic acid which is omitted in the table were in the range of 90–95% based on starting materials.
E-isomer only.
b
Photolysis of 1–3 gave not only the corresponding indanones
ysis of o-alkylphenyl ketones containing a leaving group at alpha po-
sition to the carbonyl, even though they are very common in
photochemistry of the unsubstituted system. Moreover, the dra-
matic shift of product selectivity depending upon solvents is quite
intriguing, assuming that the reaction occurs through the ground
state enol intermediate as discussed in earlier reports. Wagner ob-
served that the photoinduced formation of benzocyclobutenol from
o-alkylphenyl ketones could be inhibited in methanol3 and Scaiano
later demonstrated that the reaction could efficiently be quenched
by the addition of acids and bases.4 Our results show that methanol
does not seem to give such a negative effect on the formation of
benzocyclobutenols from 1 to 3 at least in chemical yields, which
rather makes it predominate over indanone formation.
(IN) and benzoic acid, but also the benzocyclobutenols (CBs).11
The ratios of IN to CB were strongly dependent upon solvents
and ortho-alkyl groups as shown in Table 1.12 As the size of the
ortho-alkyl groups increases, the formation of CB becomes more fa-
vored over IN. Solvent effects were more dramatic; for 1, in polar
solvents CB was strongly favored over IN, while in benzene the
opposite preference was shown. For the ethyl derivative, 2, E-
and Z-benzocyclobutenols were formed in various ratios depend-
ing on the solvents. The stereochemical assignment was made by
NOE experiments of 1H NMR.
The results are very interesting considering the fact that the
benzocyclobutenol products have not been observed in the photol-

1514
B. S. Park, H. J. Ryu / Tetrahedron Letters 51 (2010) 1512–1516
Table 2
Product distribution in percentage of photolysis of 4 and 5 in various solvents
O
O
O
OH
OH
OCOPh
OCOPh
OCOPh
ν
h
CH3
R4
+
+
+
R2
R3
R2
R1
R3
R2
R3
+
OCH3
Benzoic acid
4 : R1 = Me, R2 = R3 = R4 = H
5 : R1 = Et, R2 = R4 = Me, R3 = H
IN
CB-E
CB-Z
S
Compound
Solvent
Producta (%)
IN
CB
S
4
Benzene
Benzene (with 2.0 equiv of NMI)
Benzene (with 4.0 equiv of NMI)
MeOH
CH3CN
Acetone
100
90.9
89.7
61.0
100
100
—
—
—
—
20.7
—
9.1
10.3
18.3
—
—
—
5
Benzene
92.8
89.9
81.6
27.2
70.6
63.4ꢀ
7.2c
—
—
—
0
—
—
Benzeneb (with 2.0 equiv of NMI)
10.1c
18.4c
71.8c
29.4c
36.6c
Benzene (with 4.0 equiv of NMI)
MeOH
CH3CN
Acetone
a
b
c
Normalized to 100%. Material balances of isolated products including benzoic acid which is omitted in the table were in the range of 89–95% based on starting materials.
NMI: N-methyl imidazole.
E-isomer only.
Sterically demanding 2,4,6-trialkylphenyl ketones are known to
sterically congested further. The idea is consistent with the fact
that the relative amount of CB increases in methanol and 2 and 3
give only CB in methanol.
Solvolysis product by nucleophilic solvents such as S in Table 2
was observed only in photolysis of 4 in methanol. The reaction is
known to occur by a nucleophilic attack at benzylic position of die-
nol intermediates. Absence of the solvolysis product from the trial-
kyl ketones 1–3 and from the monoethyl ketone 5 indicates that
the reaction is very sensitive to steric crowding.
Lower quantum yield in methanol (0.09) than in benzene (0.23)
in photolysis of 2,5-dimethylphenacyl benzoate has been used as a
clue to the E-dienol being the precursor of indanone formation.7
The E-dienol can revert to the starting ketone catalytically in meth-
anol, while the Z-dienol can stay longer without returning to the
starting ketone because methanol can hinder the intramolecular
1,5-H shift of the Z-dienol. If the benzoate is replaced by the better
leaving group such as Cl in photolysis of 2,5-dimethylphenacyl
benzoate, the indanone product comes from Z-dienol via hetero-
lytic cleavage of C–Cl bond in methanol, which gives the higher
quantum yield (0.76) than that in benzene (0.11).7 We, therefore,
measured the quantum yield from 1–5 using valerophenone acti-
nometer and obtained the results summarized in Table 3.
For all the ketones that we tested, the quantum yield in meth-
anol was lower than those in benzene. The result implies that
the photoreaction occurs from an intermediate to be quenched
by methanol, whose strong candidate would be the E-dienol. Then,
be more tolerable to such inhibiting effects of methanol, and benz-
ocyclobutenols have been observed in photolysis of trialkylace-
tophenones in alcoholic solvents.13 Thus, photochemistry of
mono-substituted analogues, 4 and 5, was also investigated in sev-
eral different solvents under the same reaction condition as de-
scribed above. Table 2 summarizes the results.
Again, the benzocyclobutenols were observed in photolysis of 4
and 5 in methanol, even though the relative amount was less than
those from trialkyl analogues. To our best knowledge, this is the
first time to observe the benzocyclobutenol products from photol-
ysis of ortho-alkylphenyl ketones containing a leaving group at al-
pha position to carbonyls. The CB product was not observed in
photolysis of 4 in benzene but started to appear upon addition of
a base (N-methylimidazole, NMI), and the relative amount of the
CB increased as more NMI was added. The result is very unusual
if dienol intermediates are involved in the cyclobutenol formation
since bases are known to quench dienol intermediates even more
efficiently than acids do.4 The fact that NMI alters the ratio of IN
to CB would suggest the base do more than just quench the dienols,
assuming that both products come from the E-enols. It may be
questioned if the CB really comes from the dienol intermediates.
In fact, there has been a famous argument over the intermediacy
of the dienols in photochemistry of 2,4,6-trialkylbenzophenone.14
However, in the acetophenone cases, there seems to be an agree-
ment over the involvement of dienols in the cyclobutenol forma-
tion. Forming only E-isomer of two possible CB products from 2
and 5 in benzene supports the picture of thermal conrotatory clo-
sure from the E-dienol intermediate with stereospecific manner. If
CB is formed at the diradical stage, it would have given mixtures of
two isomers for both cases.
Table 3
Quantum yieldsa in photolysis of 1–5 in benzene and methanol
Compound
Ubenzene
Umethanol
Increasing trend of CB formation in going from 1 to 3 in benzene
was not unexpected. A sterically demanding alkyl group makes the
planar dienol intermediate unstable and forces to twist the bond
connecting the hydroxyl carbon to benzene, which makes the CB
form relatively easier and stable. The steric effect will be more en-
hanced in methanol than in benzene because solvation of the die-
nol OH by hydrogen bonding will make the planar dienols
1
2
3
4
5
0.39 0.04
0.12 0.01
0.09 0.01
0.29 0.03
0.26 0.03
0.07 0.01
0.06 0.01
—
b
0.05 0.01
a
Quantum yields of disappearance of starting ketones.
Difficult to measure due to solubility problem.
b

B. S. Park, H. J. Ryu / Tetrahedron Letters 51 (2010) 1512–1516
1515
CH3
CH2
OMs
O
OMs
OH
O
O
S O
O
O
O
ν
h
H
CH2
CH3
CH2
CH2
+
MsOH
Scheme 2.
a question remains why the ratio of IN and CB varies markedly in
changing the solvent from benzene to methanol. For the trialkyl
ketones 1–3, it may be explained that solvation of OH by methanol
induces more steric demand at the E-dienol intermediate and fa-
vors formation of CB over IN just like the ortho-alkyl size effect.
However, the mono-alkyl analogues, 4 and 5, lacking the buttress-
ing effect by an extra ortho substituent requires additional factors
to be considered to explain the observed solvent effects.
to the starting material, which would result in reduced quantum
yield. However, the methanol’s quenching effect would be larger
for formation of indanone than for that of cyclobutenol by disrupt-
ing the proper geometry for releasing the benzoates with the intra-
molecular H-bonding. Moreover, the solvation of the dienol OH by
methanol increases the steric strain to force to twist the enol C@C
bond, which makes the cyclobutenol form easier and the indanone
form more difficult.
Wessig has addressed the importance of intramolecularly
hydrogen-bonded structure shown in Scheme 2 in his studies on
photoinduced indanone formation from o-alkylphenacyl
sulfonates.8
In summary, photolysis of 2,4,6-trialkylphenacyl benzoates in
benzene resulted in an efficient formation of the corresponding
indanones and benzoic acid, but in methanol it gave predomi-
nantly the corresponding benzocyclobutenols, which were also de-
tected in the photolysis of mono-alkylphenacyl benzoates for the
first time. The dramatic shift of product selectivity depending upon
the solvents can be explained by assuming that solvation of OH
group in E-dienols by methanol disrupt the intramolecular hydro-
gen bonding between the OH and carbonyl oxygen of benzoyl
group, which is a strong driving force for the release of the carbox-
ylate leaving group. Adding bases such as NMI or using other
H-bonding acceptor solvents such as acetonitrile and DMSO also
enhances the product selectivity favoring the cyclobutenol.
In his proposed reaction mechanism, the intramolecularly H-
bonded triplet biradical releases the sulfonic acid rapidly to give
the 1,5-biradical intermediate, which then forms indanone prod-
uct. It is possible that the indanone formation from 1 follows the
similar reaction pathway, even though the timing of the elimina-
tion is still an open question, vide infra. If the H-bonding facilitates
the departure of the leaving group, addition of H-bonding acceptor
molecules would inhibit such a departure. Methanol and acetoni-
trile can disrupt the intramolecular H-bond and slow down the
elimination step to make the indanone formation inefficient. It is
worthwhile to note that only CB is formed from 1 in DMSO, a
strong H-bonding acceptor.15 The fact that adding NMI to the ben-
zene solution of 4 increases the ratio of CB to IN can also be ex-
plained by the same mechanistic scenario. Interestingly, Wessig
has used the same base to optimize the yield of indanone by
neutralizing the sulfonic acids released in photolysis of o-alkylphe-
nacyl mesylates. For comparison, we have also prepared o-meth-
ylphenacyl mesylate, 2,4,6-trimethylphenacyl mesylate, o-methyl-
phenacyl chloride, and 2,4,6-trimethylphenacyl chloride, and irradi-
ated them using the same reaction condition as described above.
These ketones, however, did not give benzocyclobutenols under
any reaction conditions that we have tried, including H-bond accep-
tor solvents and the addition of NMI.
Klan has studied the mechanism of photochemistry of o-meth-
ylphenacyl sulfonates16 and reported that the reaction in methanol
proceeds from Z-dienol via heterolytic cleavage, while in benzene
the elimination occurs from the E-dienol. The reaction mechanism
parallels with those of o-alkylphenacyl chlorides and phosphates,
but is different from that of the carboxylate whose elimination oc-
curs from the E-dienol regardless of solvents. For the system hav-
ing relatively poor leaving group such as carboxylates, the
elimination step is too slow to compete with the other reaction
path such as reketonization to starting material via 1,5-H shift
from the Z-dienol. In such a system, the decay process from triplet
dienol (or biradical) to the ground state dienol would occur before
the elimination, so the retardation of the elimination step by meth-
anol or bases such as NMI would also be made at the ground state
dienol rather than at the biradical state.
Acknowledgment
This research was supported by Dongguk University (Seoul
campus).
References and notes
1. (a) Haag, R.; Wirz, J.; Wagner, P. J. Helv. Chim. Acta 1977, 60, 2595–2607; (b)
Wagner, P. J.; Sobczak, M.; Park, B. S. J. Am. Chem. Soc. 1998, 120, 2488–2489; (c)
Sobczak, M.; Wagner, P. J. Tetrahedron Lett. 1998, 39, 2523–2526; (d) Small, R.
D.; Scaiano, J. C. J. Am. Chem. Soc. 1977, 99, 7713–7714; (e) Das, P. K.; Encinas,
M. V.; Small, R. D.; Scaiano, J. C. J. Am. Chem. Soc. 1979, 101, 6965–6970; (f)
Johnston, L. J.; Scaiano, J. C. Chem. Rev. 1989, 89, 521–547.
2. (a) Sammes, P. G. Tetrahedron 1976, 32, 405–422; (b) Nicolaou, K. C.; Gray, D. L.
F. J. Am. Chem. Soc. 2004, 126, 607–612; (c) Nicolaou, K. C.; Gray, D. L. F.; Tae, J.
S. J. Am. Chem. Soc. 2004, 126, 613–627.
3. Wagner, P. J.; Subrahmanyam, D.; Park, B. S. J. Am. Chem. Soc. 1991, 113, 709–
710.
4. Scaiano, J. C.; Wintgens, V.; Netto-Ferreira, J. C. Tetrahedron Lett. 1992, 33,
5905–5908.
5. (a) Klan, P.; Zabadal, M.; Hegar, D. Org. Lett. 2000, 2, 1569–1571; (b) Klan, P.;
Pelliccioli, A. P.; Pospisil, T.; Wirz, J. Photochem. Photobiol. Sci. 2002, 1, 920–923;
(c) Bergmark, W. R.; Barnes, C.; Clark, J.; Paprian, S.; Marynowski, S. J. Org. Chem.
1985, 50, 5612–5615.
6. Pelliccioli, A. P.; Klan, P.; Zabadal, M.; Wirz, J. J. Am. Chem. Soc. 2001, 123, 7931–
7932.
7. Zabadal, M.; Pelliccioli, A. P.; Klan, P.; Wirz, J. J. Phys. Chem. A 2001, 105, 10329–
10333.
8. Wessig, P.; Glombitza, C.; Muller, G.; Teubner, J. J. Org. Chem. 2004, 69, 7582–
7591.
9. (a) Chang, D. J.; Park, B. S. Tetrahedron Lett. 2001, 42, 711–713; (b) Chang, D. J.;
Nahm, K.; Park, B. S. Tetrahedron Lett. 2002, 43, 4249–4252; (c) Cho, S.; Park, B.
S. Bull. Korean Chem. Soc. 2004, 25, 42–44; (d) Park, B. S.; Cho, S.; Chong, S.-H.
Bull. Korean Chem. Soc. 2007, 28, 1156–1158; (e) Park, B. S.; Lee, H. M. Bull.
Korean Chem. Soc. 2008, 29, 2054–2056; (f) Park, B. S.; Jeong, S. Bull. Korean
Chem. Soc. 2009, 30, 3053–3056.
Our mechanistic reasoning for the predominant formation of
cyclobutenol over indanone in methanol can be summarized as fol-
lows. We believe that both CB and IN come from E-dienol as de-
scribed above. Methanol would then decrease the efficiency of
forming both products by catalyzing reketonization of the E-dienol
10. As a representative example, spectroscopic data of 1: 1H NMR (CDCl3, 200 MHz) d
8.13 (distorted d, 2H, J = 7.4 Hz), 7.63 (distorted t, 1H, J = 7.4 Hz), 7.49
(distorted t, 2H, J = 7.4 Hz), 6.90 (s, 2H), 5.18 (s, 2H), 2.35 (s, 6H), 2.32 (s,
3H), 13C NMR (CDCl3, 50 MHz) d 203.0, 166.1, 139.6, 135.2, 134.1, 133.5, 130.0,

1516
B. S. Park, H. J. Ryu / Tetrahedron Letters 51 (2010) 1512–1516
129.4, 128.7, 128.6, 69.7, 21.2, 19.2 ppm, IR (KBr) 1725, 1716 (C@O str.) cmꢀ1
,
consumption of starting ketones, which showed that any contribution from
the secondary photoreactions of initially formed products is negligible under
our experimental conditions.
EIMS 282 (M+).
11. As
a
representative example, spectroscopic data of IN-1: 1H NMR (CDCl3,
200 MHz) d 7.10 (s, 1H), 6.94 (s, 1H), 3.06 (AA0BB0, 2H), 2.68 (AA0BB0, 2H),
2.61 (s, 3H), 2.40 (s, 3H), 13C NMR (CDCl3, 50 MHz) d 207.5, 156.6, 145.1, 138.6,
13. Kitaura, Y.; Matsuura, T. Tetrahedron 1971, 27, 1597–1606.
14. (a) Wagner, P. J.; Park, B. S. Org. Photochem. 1991, 11, 227–366; (b) Ito, Y.;
Takahashi, H.; Hasegawa, J.; Turro, N. J. Tetrahedron 2009, 65, 677–689; (c)
Nakayama, T.; Hidaka, T.; Kuramoto, T.; Hamanoue, K.; Teranishi, H.; Ito, Y.;
Matsuura, T. Chem. Lett. 1984, 1953–1956; (d) Ito, Y.; Umehara, Y.; Matsuura, T.
Chem. Lett. 1980, 939–942.
15. A referee pointed out that our experimental results were in good agreement
with the b scale (solvent hydrogen bond acceptor basicity proposed by Taft and
co-workers: Kamlet, M. J.; Abboud, J.-L. M.; Abraham, M. H.; Taft, R. W. J. Org.
Chem. 1983, 48, 2877–2887.). We appreciate for the useful comment.
16. Klan, P.; Pelliccioli, A. P.; Pospisil, T.; Wirz J. Photochem. Photobiol. Sci. 2002, 1,
920–923.
132.4, 130.4, 124.5, 37.0, 25.3, 21.9, 18.3 ppm, IR (KBr) 1698 (C@O str.) cmꢀ1
,
EIMS 160 (M+). Spectroscopic data of CB-1: 1H NMR (CDCl3, 200 MHz) d 8.11
(distorted d, 2H, J = 7.4 Hz), 7.62 (distorted t, 1H, J = 7.4 Hz), 7,48 (distorted t,
2H, J = 7.4 Hz), 6.86 (s, 1H), 6.85 (s, 1H), 4.73, 4.66 (AB quartet, 2H, J = 11.6 Hz),
3.47, 3.21 (AB quartet, 2H, J = 14.0 Hz), 2.34 (s, 3H), 2.30 (s, 3H), 2.07 (s, 1H, –
OH), 13C NMR (CDCl3, 50 MHz) d 167.0, 142.3, 141.1, 140.1, 133.3, 132.7, 130.0,
129.9, 129.6, 128.5, 121.6, 78.8, 69.9, 43.8, 22.1, 17.3 ppm, IR (KBr) 3463 (O–H
str.), 1726 (C@O str.) cmꢀ1, EIMS 282 (M+).
12. The ratios in the table were unchanged within experimental errors after
irradiating for time periods twice as long as those taken to complete