Thermolysis and Photosensitized Oxygenation of Tetrasubstituted Cyclopropenes

Article abstract of DOI:10.1021/jo991803b

Bicyclic cyclopropenes 14a, 14b, and 26 were prepared by various synthetic routes. Polymer rose Bengal (p-RB) photosensitized oxygenation of bicyclooctenes 14a,b in CDCl₃ proceeded sluggishly (variable O₂ uptake of ca. 0.35-0.75 equiv in 8 h) and was accompanied by sensitizer bleaching. Preparative gas chromatography of the complex product mixtures from 14a and 14b yielded both dienes (Z- and E-29, 30, and 31) and enones (E- and Z-12, 32, 34). By contrast, p-RB photosensitized oxidation of bicyclononene 26 in CDCl₃ proceeded somewhat more rapidly (O₂ uptake of ca. 1 equiv in 2.5 h) yielding enones (20, 42-45) exclusively upon GC separation. The diene products, observed in the case of 14, result from the thermolysis of the remaining unreacted cyclopropenes, while the enones are the oxygenation products. The oxygenation was slowed by radical inhibitors, but not by ¹O₂ quenchers; nor were any oxidation products observed when these cyclopropenes were reacted with triphenylphosphine ozonide, a chemical ¹O₂ source. The data indicates that a photosensitizer-initiated free radical autoxidative process is involved. Likely intermediates in this oxygenation are epoxide 27 or 37 and hydroperoxide 28 or 38, for the bicyclooctene (14) and bicyclononene (26) systems, respectively. The absence of ¹O₂ product in these cyclopropene systems, in contradistinction to their higher homologues, may be attributable to either the relatively long C_α-H_allylic distance in alkylcyclopropenes, which places the abstractable allylic hydrogen "out of reach", or their relatively high IP. Either, or both, of these factors may have slowed the rate of the singlet oxygenation of the cyclopropenes to a point where free radical processes compete favorably. In the course of this study, we also explored the singlet oxygenation (DABCO inhibited) of enones 12a,b and 20. These generated, respectively, a mixture of peroxides identified as α-keto hydroperoxides 51/54 and hemiperketals 52/55 (the cyclic form of β-keto hydroperoxides 53/56). Phosphine reduction of these peroxides yields the corresponding alcohols 33/43 and 32/42.

Full text of DOI:10.1021/jo991803b

J . Org. Chem. 2000, 65, 1807-1817
1807
Th er m olysis a n d P h otosen sitized Oxygen a tion of Tetr a su bstitu ted
Cyclop r op en es
Aryeh A. Frimer,* Michal Afri, Simeon D. Baumel, Pessia Gilinsky-Sharon,
Zilpa Rosenthal, and Hugo E. Gottlieb
The Ethel and David Resnick Chair in Active Oxygen Chemistry, The Department of Chemistry,
Bar-Ilan University, Ramat Gan 52900, Israel
Received November 22, 1999
Bicyclic cyclopropenes 14a , 14b, and 26 were prepared by various synthetic routes. Polymer rose
Bengal (p-RB) photosensitized oxygenation of bicyclooctenes 14a ,b in CDCl3 proceeded sluggishly
(variable O2 uptake of ca. 0.35-0.75 equiv in 8 h) and was accompanied by sensitizer bleaching.
Preparative gas chromatography of the complex product mixtures from 14a and 14b yielded both
dienes (Z- and E-29, 30, and 31) and enones (E- and Z-12, 32, 34). By contrast, p-RB photosensitized
oxidation of bicyclononene 26 in CDCl3 proceeded somewhat more rapidly (O2 uptake of ca. 1 equiv
in 2.5 h) yielding enones (20, 42-45) exclusively upon GC separation. The diene products, observed
in the case of 14, result from the thermolysis of the remaining unreacted cyclopropenes, while the
enones are the oxygenation products. The oxygenation was slowed by radical inhibitors, but not by
1O2 quenchers; nor were any oxidation products observed when these cyclopropenes were reacted
with triphenylphosphine ozonide, a chemical 1O2 source. The data indicates that a photosensitizer-
initiated free radical autoxidative process is involved. Likely intermediates in this oxygenation
are epoxide 27 or 37 and hydroperoxide 28 or 38, for the bicyclooctene (14) and bicyclononene (26)
systems, respectively. The absence of 1O2 product in these cyclopropene systems, in contradistinction
to their higher homologues, may be attributable to either the relatively long CR-Hallylic distance in
alkylcyclopropenes, which places the abstractable allylic hydrogen “out of reach”, or their relatively
high IP. Either, or both, of these factors may have slowed the rate of the singlet oxygenation of the
cyclopropenes to a point where free radical processes compete favorably. In the course of this study,
we also explored the singlet oxygenation (DABCO inhibited) of enones 12a ,b and 20. These
generated, respectively, a mixture of peroxides identified as R-keto hydroperoxides 51/54 and
hemiperketals 52/55 (the cyclic form of â-keto hydroperoxides 53/56). Phosphine reduction of these
peroxides yields the corresponding alcohols 33/43 and 32/42.
In tr od u ction
Cyclopropene is a petite storehouse of ca. 50 kcal of
strain energy,4 which clearly suggests itself as an in-
triguing candidate for singlet oxygenation. Indeed, more
than 20 years ago, Griffin and co-workers5 carried out
the first in a series of studies on the 1O2 reaction of
alkylated and arylated derivatives of 1,2-diarylcyclopro-
penes 1a -d (eq 1). In each case, the oxidative cleavage
products observed could be rationalized by invoking
dioxetanes 25 or endoperoxides 31d,e,3 as the initial
intermediates. Indeed, in the case of vinyl aromatics,
dioxetanes and endoperoxides are generally observed as
Over the past two decades, we have been exploring the
effect of strain on the rate, mode, and direction of singlet
oxygen (1O2) reactions.1 In particular, we have focused
on various small ring olefin systems in which ring-strain
decreases or develops as we proceed toward product.
These studies have suggested that, in the transition state
leading to product, 1O2 is essentially insensitive to strain
considerations.1a-1c This conclusion is consistent with
prior evidence that singlet oxygen reactions have very
small activation energies (0.5-8 kcal/mol)2 and that the
product-determining transition state is reactant-like and
occurs quite early.3
(4) Baird, N. C.; Dewar, M. J . S. J . Am. Chem. Soc. 1967, 89, 3966-
3970.
(5) (a) Politzer, I. R.; Griffin, G. W. Tetrahedron Lett. 1973, 4775-
4778. (b) Griffin, G. W.; Politzer, I. R.; Ishikawa, K.; Turro, N. J .; Chow,
M.-F. Tetrahedron Lett. 1977, 1287-1290. (c) Vaz, C.; Griffin, G.;
Christensen, S.; Lankin, D. Heterocycles 1981, 15, 1643-1658. (d) See
also the introduction to Griffin, G. W.; Kirschenheuter, G. P.; Vaz, C.;
Umrigar, P. P.; Lankin, D.; Christensen, S. Tetrahedron 1985, 41,
2069-2080.
(6) (a) Denny, R. W.; Nickon, A. Org. React. 1973, 20, 133-336. (b)
Gollnick, K.; Kuhn, H. J . In Singlet Oxygen; Wasserman, A. A., Murray,
R. W., Eds.; Academic Press: New York, 1979; pp 287-427; see
especially pp 318-327.
(7) (a) Schenk, G. O.; Koch, E. Z. Electrochemistry 1960, 64, 170.
(b) Rosental, I. In Singlet O2 - Volume I: Physical-Chemical Aspects;
Frimer, A. A., Ed.; Chemical Rubber Company: Boca Raton, FL, 1985;
pp 13-38.
(8) Closs, G. L.; Closs, L. E.; Boll, W. A. J . Am. Chem. Soc. 1963,
85, 3796-3800.
(9) Friedrich, L. E.; Leckonby, R. Y.; Stout, D. M.; Lam, Y.-S. P. J .
Org. Chem. 1978, 43, 604-610.
* To whom correspondence should be addressed. Phone: 972-3-
5318610, fax: 972-3-5351250, e-mail: frimea@mail.biu.ac.il.
(1) (a) Frimer, A. A.; Roth, D.; Sprecher, M. Tetrahedron Lett. 1977,
1927-1930. (b) Frimer, A. A.; Farkash, T.; Sprecher, M. J . Org. Chem.
1979, 44, 989-995. (c) Frimer, A. A.; Roth, D. J . Org. Chem. 1979, 44,
3882-3887. (d) Frimer, A. A.; Antebi, A. J . Org. Chem. 1980, 45, 2334-
2340. (e) Frimer, A. A. Isr. J . Chem. 1981, 21, 194-202. (f) Frimer, A.
A. J . Photochem. 1984, 25, 211-226. (g) Frimer, A. A. In The Chemistry
of Peroxides; Patai, S., Ed.; Wiley: New York, 1983; pp 201-234. (h)
Frimer, A. A.; Stephenson, L. M. In Singlet O2 - Volume II: Reaction
Modes and Products - Part I; Frimer, A. A., Ed.; Chemical Rubber
Company, Boca Raton, FL, 1985; pp 67-91. (i) Frimer, A. A.; Weiss,
J . J . Org. Chem. 1993, 58, 3660-3667. (j) Frimer, A. A.; Weiss, J .;
Gottlieb, H. E.; Wolk, J . L. J . Org. Chem. 1994, 59, 780-792.
(2) (a) Koch, E. Tetrahedron 1968, 24, 6295-6318. (b) Ashford, R.
D.; Ogryzlo, E. A. J . Am. Chem. Soc. 1975, 97, 3604-3607.
(3) Frimer, A. A. Chem. Rev. 1979, 79, 359-387.
10.1021/jo991803b CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/24/2000

1808 J . Org. Chem., Vol. 65, No. 6, 2000
Frimer et al.
Sch em e 1. Syn th etic Sch em e for th e P r ep a r a tion of Cyclop r op en es 14a (R ) R′ ) CH3) a n d 14b (R ) CH3;
R ) H)
Sch em e 2. Syn th etic Sch em e for th e P r ep a r a tion
of Alk ylid en ecycloh ep ta n on es 12a (R ) R′ ) CH3)
a n d 12b (R ) CH3; R′ ) H), a n d
the sole primary singlet oxygenation products, even if
allylic hydrogens are available for a possible ene reaction
mode.1e,g,3,6
Isop r op ylid en ecycloocta n on e 20 (R ) R′ ) H)
1
In an attempt to focus in on the O2 ene reaction in
cyclopropenyl systems, Frimer and Antebi1d carried out
the photosensitized oxygenation of cycloolefins 4a -c (eq
2). A plethora of products were formed which were
rationalized in terms of secondary rearrangements of
initially formed allylic hydroperoxide 5 and cyclopropene
epoxide 6. Although the former is the expected ene
were unsuccessful in fishing out reaction products. We
decided, therefore, to prepare bulkier systems, namely
bicyclooctenes 14a and 14b and bicyclononene 26. The
former were initially prepared from cycloheptanone (7)
by a modification of the Friedrich procedure (Scheme 1).9
We found, however, that this approach to the crucial
precursors, alkylidenecycloheptanones 12a and 12b, was
both tedious and afforded low yields. Hence, we turned
to the far more convenient silyl enol ether crossed aldol
method,10 outlined in Scheme 2. For the purpose of
oxidation product identification in the corresponding
bicyclononene system (vide infra), we also utilized Scheme
2 to synthesize isopropylidenecyclooctanone 20. The
preparation of bicyclononene 26 was effected via the
cyclopropanation of cyclooctyne (eq 3).11
1
product, the authors demonstrate that O2 is not involved.
The oxygenation is rather a free-radical (“type I”)7
autoxidative process, presumably photosensitizer-initi-
ated via the abstraction of the labile C-3 ring hydrogen.
The above results lead us to the conclusion that our
search for a singlet oxygen ene process should be directed
toward tetraalkylated cyclopropenes, in which only exo-
cyclic allylic hydrogens are available. The preparation
and photosensitized oxygenation of several such systems
is reported below. In brief, our results demonstrate that,
while the expected oxygenation is indeed observed, the
process is once again free-radical in nature and does not
1
involve O2.
Resu lts a n d Discu ssion
A. Syn t h esis a n d P h ot ooxid a t ion of Cyclop r o-
p en es 14a , 14b, a n d 26. In initial studies, we prepared
the simplest tetrasubstituted cyclopropene, tetrameth-
ylcyclopropene.8 Upon photosensitized oxidation, a slug-
gish oxygen uptake was indeed observed; however, we
(10) (a) Fleming, I.; Paterson, I. Synthesis 1979, 736-738. (b)
Mukaiyama, T.; Narasaki, K. Org. Synth. 1987, 65, 6-11.
(11) (a) Brandsma, L.; Verkruijsig, H. D. Synthesis 1978, 290. (b)
Wittig, G.; Hutchison, J . J . J ustus Leibigs Ann. Chem. 1970, 741, 79-
88.

Thermolysis and Oxygenation of Substituted Cyclopropenes
J . Org. Chem., Vol. 65, No. 6, 2000 1809
Ta ble 1. Distr ibu tion of Dien e P r od u cts in th e
P h otosen sitized Oxygen a tion of Cyclop r op en es 14a a n d
14b
Sch em e 3. P r od u ct F or m a tion in th e
P h otosen sitized Oxygen a tion of Tetr a su bstitu ted
Cyclop r op en es 14a , 14b, a n d 26
diene product distributiona (%)
substrate 14
Z-29
E-29
30
31
a
b
36
25
28
25
36
25
-
25
a
The values are the average of at least three runs with a spread
of e2%.
Our working hypothesis was that the diene products
obtained result from GC-induced thermolysis of unre-
acted cyclopropene, while the enones are rearranged
oxygenation products. For the purpose of clarity, we will
deal separately with the identification and mechanism
of formation of these two classes of compounds.
B. F or m a tion of th e Dien e P r od u cts. The relative
distribution of diene products obtained from the photo-
oxidation of cyclopropenes 14a and 14b are given in Table
1. The products in each case were identified based on
their spectral data, with NOE assisting in the E and Z
assignments in the case of 29a and 29b, and 31b
(E-isomer exclusively). In addition, compounds 30a , 30b,
and 31b were independently synthesized in three steps:
Friedel Crafts acylation of cycloheptene9 generates â-chlo-
roketones 46;13 base-catalyzed elimination of HCl from
the latter yields enones 47; a Wittig reaction produces
the desired dienes (eq 4). (While 31b isolated from the
photooxidation of 14b was exclusively the E-isomer, the
NMR of 31b prepared by the Wittig reaction revealed
the presence of ca. 20% of the corresponding Z-isomer
31c; see Experimental Section and eq 12 below).
Polymer Rose Bengal (p-RB) photosensitized oxidation
of bicyclooctenes 14a ,b in CDCl3 proceeded sluggishly
(variable O2 uptake of ca. 0.35-0.75 equiv in 8 h) and
was accompanied by sensitizer bleaching.12 As sum-
marized in Scheme 3, preparative gas chromatography
of the complex product mixtures from 14a and 14b
yielded both dienes (Z- and E-29, 30, and 31) and enones
(E- and Z-12, 32, 34). By contrast, p-RB photosensitized
oxidation of bicyclononene 26 in CDCl3 proceeded some-
what more rapidly (O2 uptake of ca. 1 equiv in 2.5 h)
yielding enones (20, 42-45) exclusively upon GC separa-
tion.
For the purpose of our later discussion, it should be
noted that triphenylphosphine is commonly added at the
end of 1O2 reactions to reduce any hydroperoxides formed
to the more stable alcohols. In addition, the Ph3P reduc-
tion of hydroperoxides is typically quite exothermic.
Indeed, instances where release of heat are not observed
indicate either that hydroperoxides are either not formed
at all, or are so labile that they rearrange to nonperoxidic
products prior to Ph3P treatment.1b,d,f,g In the present
study, the addition of Ph3P generated no heat and had
little if any effect on the product distribution. Hence,
workup of the present reaction mixtures proceeded
without prior addition of Ph3P.
As noted above, we assumed that the dienes result
from GC-induced thermolysis/rearrangement of unre-
acted cyclopropene. Indeed, such rearrangements are well
precedented.14 An excellent case in point is the report of
(12) (a) Sensitizer bleaching is often the first indication of the
involvement of free radical processes. See, for example, refs 1d and 1i.
Indeed, we had chosen p-RB because it is generally not bleached
photochemically12b and is compatible with nonpolar solvents.12c (b)
Graf, G.; Braun, A. M.; Faure, J . J . Org. Chimia 1980, 34, 234. (c)
Paczkowska, B.; Paczkowska, J .; Neckers, D. C. Photochem. Photobiol.
1985, 42, 603-604.
(13) March, J . Advanced Organic Chemistry, 3rd ed.; Wiley: Inter-
science: New York; 1985; p 536. The large J 1, ) 9.8 Hz in 46a
2
indicates that H1 and H2 are quasi-trans a,a to each other (i.e., that
the chlorine and acyl group are in a quasi trans e,e relationship); see
Experimental Section.

1810 J . Org. Chem., Vol. 65, No. 6, 2000
Frimer et al.
Ta ble 2. Oxid a tion P r od u ct Distr ibu tion in th e P h otosen sitized Oxygen a tion of Cyclop r op en es 14a , 14b, a n d 26 u n d er
Va r iou s Con d ition s
rxn ratea
(mL/min/mmol)
olefin
rxn cond
oxygenation product distribution (%)b
14
a
b
P-RB/CDCl3
RT
RT
E-12
5
Z-12
5
32
8
69
33
-
-
34
61
31
35
21
-
36
-
-
0.04c
0.02d
-
-
26
P-RB/CH3CN
RT
20
17
10
13
87
84
42
27
38
38
-
43
38
36
31
-
44
5
1
-
-
5
45
13
15
18
13
5
0.160e
0.270
0.015
0.022
0.003
RT; DABCO
-40 °C
RT; DTBPf
RT; dark
3
3
a
b
The reaction rate is based on the rate of oxygen uptake (mL/min) per mmol of substrate. The values are the average of at least three
runs with a spread of e2%. c Approximately 0.75 equiv of O2 were absorbed in 8 h. Approximately 0.35 equiv of O2 were absorbed in 8
d
h. e Approximately 1 equiv of O2 was absorbed in 2.5 h. f 2,6-Di-tert-butylphenol.
Sch em e 4. P r od u ct F or m a tion in th e Sin glet
Oxygen a tion of En on es 12 a n d 20
Closs and co-workers who note that 7,7-dimethylbicyclo-
heptene 48 (a lower homologue of 26) readily rearranges
to 49 and 50 (eq 5; lower homologues of 39 and 40).15
This hypothesis explains why the relative total amount
of dienes isolated by GC from the photooxygenation of
14a and 14b, as compared to the total amount of enones
observed, varied from photooxidation to photooxidation;
it was simply proportional to the amount of unreacted
substrate remaining (as approximated by the amount of
oxygen uptake). Also consistent with this hypothesis is
the fact that no dienes (e.g., 39, 40, or 41) are observed
in the case of bicyclononene 26, where the photooxidation
proceeded essentially to completion.
To check the validity of this hypothesis, cyclopropenes
14a and 14b were injected into the GC. When the
injector, detector, and column temperatures were main-
tained below 110 °C, the starting olefins were obtained
essentially unchanged. However, above these conditions
rearrangement is observed. In the case of 14a , with the
GC injector and detector set for 190 °C and the column
at 120 °C, the products E- and Z-29a and 30a were
obtained essentially in the ratio of Table 1. Similarly,
heating a neat sample of 14a to 170 °C for 0.5 h gave
the same results. Surprisingly, in the case of 14b, GC
thermolysis as above yielded E- and Z-29b exclusively,
in a 1:1 ratio; heating a neat sample of 14b to 170 °C for
0.5 h yielded equivalent amounts of 30b and 31b. The
source of this discrepancy is not yet clear to us; we
suspect the involvement of secondary rearrangements,
but we have not pursued the question further.
propenes 14a , 14b, and 26 is summarized in Table 2. For
the purpose of elucidating the mechanism of these
photooxidations (vide infra), the photosensitized oxygen-
ation of 26 was repeated under various conditions, and
these results are also given in Table 2. The products in
each case were identified based on their spectral data,
with NOE assisting in the E and Z assignments for 12a
and 12b. In addition, most of these products could be
independently synthesized as outlined in Scheme 4. Thus,
singlet oxygenation (DABCO inhibited) of enones17 12a ,
12b, and 20 generated, in each case, a mixture of
peroxides identified as R-keto hydroperoxides 51/54 and
hemiperketals 52/55 (the cyclic form of â-keto hydroper-
oxides 53/56). Phosphine reduction of these peroxides
yields the corresponding alcohols 33/43 and 32/42.
Several comments are appropriate at this juncture.
Generally speaking, in the above-mentioned singlet
oxygenation of enones 12, we irradiated the E/ Z-mixture.
In one instance we did react pure (from GC) E-12a and
pure Z-12a and obtained from each essentially the same
product distribution of peroxides 51a and 52a . As shown
We note in closing that molecular modeling studies16
suggest that Z-29a is 0.3 kcal more stable than the
E-analogue, while Z- and E-29b are essentially degener-
ate. These data are, of course, consistent with the results
of Table 1.
1
in Scheme 5, since O2 can attack the double bond from
above or below, both E- and Z-isomers generate the same
enantiomeric mixture of R-hydroperoxy ketones 51/51′
and â-hydroperoxy ketones 53/53′ as the primary prod-
ucts. As already noted above, the latter cyclize to hemi-
perketals 52, a process which generates another chiral
carbon and, hence, two diastereomers. Indeed, the NMR
C. F or m a tion of th e En on e P r od u cts. The product
distribution in the photosensitized oxygenation of cyclo-
(14) (a) Closs, G. L. In Advances in Alicyclic Chemistry; Hart H.,
Karabtsos, G. J ., Eds.; Academic Press: New York, 1966; Vol. I, pp
86-94. (b) Halton, B.; Banwell, M. G. In Chemistry of the Cyclopropyl
Group; Patai, S., Rappaport, Z., Eds.; Wiley: 1980; pp 1272-1281.
(15) Closs, G. L.; Bo¨ll, W. A.; Heyn, H.; Dev, V. J . Am. Chem. Soc.
1968, 90, 173-178, especially p 174.
(17) (a) Frimer, A. A. In The Chemistry of Enones - Part 2; Patai,
S., Rappoport, Z., Eds.; Wiley: New York, 1989; pp 781-921. (b)
Ensley, H. In Advances in Oxygenated Processes; Baumstark, A., Ed.;
J AI Press Inc.: Greenwich, CT, 1990; Volume 2, pp 181-201. (c)
Elemis, Y.; Foote, C. S. J . Am. Chem. Soc. 1992, 114, 6044-6050.
(16) Autodesk HyperChem 3, AM1 semiempirical calculations.

Thermolysis and Oxygenation of Substituted Cyclopropenes
J . Org. Chem., Vol. 65, No. 6, 2000 1811
Sch em e 5. Ster eoch em ica l Asp ects of th e Sin glet
Oxygen a tion of En on es 12
molecule of substrate, which generally results in hydro-
gen atom abstraction from the substrate, and initiates
free radical autoxidative processes (equations 6 and 7).
“Type II”, on the other hand, is characterized by initial
interaction between the sensitizer triplet and a molecule
of oxygen. The traditional “type II”20 involves transfer
of the sensitizer’s excitation energy to molecular oxygen
generating singlet molecular oxygen (1O2), which is the
oxygenation species (eqs 8 and 9).
Type I:
Sens3 + RH f •SensH + R•
R• + 3O2 f ROO• f oxygenation products (7)
(6)
Type II:
Sens3 + 3O2 f Sens1 + 1O2
RH + 1O2 f oxygenation
(8)
(9)
There is ample evidence in this study to rule out a “type
II” oxygenation and demonstrate that a nonsinglet
oxygen/free radical mechanism predominates in the
photosensitized oxidation of tetraalkylatedcyclopro-
penes: (1) The 1O2-quencher DABCO1g,21 did not slow the
rate or course of the reaction. (2) On the other hand,
addition of the free-radical inhibitor 2,6-di-tert-butyl-
phenol1g,22 had a dramatic damping effect on the rate,
with a concomitant change in the product distribution.
(3) When the reaction temperature was lowered from 25
°C to -40 °C, the rate of oxygen uptake dropped by a
factor of 10. Such an effect is typical of free radical
autoxidations. However, because of their low activation
energies, the rate of singlet oxygenations are well-known
to be insensitive to reaction temperature.1g (4) When 26
was exposed to 1O2 chemically generated from triphenyl
phosphite ozonide,1d,23 no reaction products were ob-
served.
Having determined that these cyclopropenes undergo
photosensitized oxidation via a “type I” mechanism, we
now turn to the details of product formation. The mech-
anism which we believe best fits the available data
involves free radical autoxidative generation of two initial
products (see Scheme 3): cyclopropyl epoxides 27/37s
via short chain polyperoxidation,24 and alkylidenecyclo-
propyl hydroperoxides 28/38sfrom a competing autoxi-
dative hydroperoxidation process.22 As we shall outline
spectral data for hemiperketals 52 reveals the presence
of two diastereomeric formssin a ratio of 2:1 for 52a and
1.2:1 for 52b. NOE studies indicate that the major isomer
in each case has the methyl and hydroxyl groups cis to
each other (cis-52; see Scheme 5). Studying Scheme 5
makes it clear that while 53 cyclizes to trans-52 and cis-
52, its enantiomer 53′ cyclizes to trans-52′ and cis-52′.
While these are actually four different diastereomers, the
“primed” set are merely mirror images of the unprimed
analogues and hence indistinguishable by NMR in non-
chiral solvents.
Second, we have suggested above that hemiperketals
52/55 result from the cyclization of the initial ene product
â-keto hydroperoxides 53/56. We should note, however,
that in the case of 52a , we have found no spectral
evidence for the presence of the open form 53a in the
product mixture. Indeed, we see no carbonyl absorption
in the FTIR of 52a . Furthermore, while thiourea readily
reduces the open hydroperoxide 51a , it does not do so
for cyclic peroxide 52a . In the case of 52b, however, we
do see a moderate IR absorption at 1701 cm-1 (conjugated
carbonyl) suggesting the presence of a low equilibrium
concentration of 53b.
Finally, we note that the hydrogens R to the carbonyl
of several of the above enones (20, 34a , 35a , 42, 44, and
45) appear in the 1H NMR as an “inverted quintet”
(abbreviated as inv quint) with peaks in a ratio of ca.
5:1:4:1:5. Such a pattern is typical of an AA′XX′ sys-
tems.18
Turning now to the question of mechanism, the reader
is reminded that there are two major classes of photo-
sensitized oxidative processes, appropriately called “type
I” and “type II”.19 The former is a photoinitiated autoxi-
dation in which only ground-state triplet molecular
oxygen (3O2) is involved. A “type I” process is character-
ized by the direct reaction of the triplet sensitizer with a
(20) J efford, C. W.; Boschung, A. F. Helv. Chim. Acta 1977, 60,
2673-2685.
(21) (a) Ouannes, C.; Wilson, T. J . Am. Chem. Soc. 1968, 90, 6527-
6528. (b) Davidson, R. S.; Trethewey, K. R. J . Am. Chem. Soc. 1976,
98, 4008-4009. (c) Davidson, R. S.; Trethewey, K. R. J . Chem. Soc.,
Perkin Trans. 2 1977, 178-182. (d) Monroe, B. M. J . Phys. Chem. 1977,
81, 1861-1864. (e) Deneke, C. F.; Krinsky, N. I. Photochem. Photobiol.
1977, 25, 299-304.
(22) Cf. Foote, C. S. In Free Radicals in Biology; Pryor, W. A., Ed.;
Academic Press: New York, 1976; Vol. II, pp 85, 101.
(23) Cf. Murray, R. W.; Kaplan, M. L. J . Am. Chem. Soc. 1969, 91,
5358-5364.
(24) (a) Mayo, F. R. J . Am. Chem. Soc. 1958, 80, 2465-2480. (b)
Mayo, F. R.; Miller, A. A.; Russell, G. A. J . Am. Chem. Soc. 1958, 80,
2500-2507. (c) Van Sickle, D. E.; Mayo, F. R.; Gould, E. S.; Arluck, R.
M. J . Am. Chem. Soc. 1967, 89, 977-984. (d) Mayo, F. R. Acc. Chem.
Res. 1968, 7, 193-201. (e) Howard, J . A. In Free Radicals; Kochi, J .
K., Ed.; Wiley: New York, 1973; Vol. II, pp 3-62; see especially p 25ff.
(f) Filippova, T. V.; Blyumberg E. A. Russian Chem. Rev. 1982, 51,
582-591. (g) This short-chain polyperoxidation occurs only very slowly
at room temperature to ultimately yield 20; see the last entry in Table
2.
(18) Gunther, H. NMR Spectroscopy; Wiley: Chichester: 1980; pp
181ff.
(19) (a) Schenk, G. O.; Koch, E. Z. Electrochemistry 1960, 64, 170.
(b) Rosenthal, I. In Singlet O2 - Volume I: Physical-Chemical Aspects;
Frimer, A. A., Ed.; Chemical Rubber Company, Boca Raton, FL, 1985;
pp 13-38. (c) For a more complete discussion, see ref 1i.

1812 J . Org. Chem., Vol. 65, No. 6, 2000
Frimer et al.
Sch em e 7. P r op osed Mech a n ism of Dien on e
F or m a tion in th e P h otosen sitized Oxygen a tion of
Tetr a su bstitu ted Cyclop r op en es 14a , 14b, a n d 26
Sch em e 6. P r op osed Mech a n ism of En on e
F or m a tion in th e P h otosen sitized Oxygen a tion of
Tetr a su bstitu ted Cyclop r op en es 14a , 14b, a n d 26
Turning now to hydroperoxides 28/38, one common
secondary reaction of peroxides in general, and hydro-
peroxides (R′ ) H) in particular, is the Kornblum-
DeLaMare reaction (eq 10).1f,1h,3,17a A homologous dehy-
dration reaction would be expected for cyclopropyl hy-
droperoxides, yielding â,λ-enones (eq 11).
below, it is the well-precedented secondary transforma-
tions of each of these initial products which yields the
plethora of final products observed.
Regarding epoxides 27/37, the rearrangement of cy-
clopropyl epoxides to the corresponding enones is well
documented.25 Indeed, the MCPBA epoxidation of 14a
and 14b is reported to yield E- and Z-12a and 12b,
respectively;9 we verified that the same transformation
occurs in the case of 26 as well. We suggest that, under
the free radical reaction conditions prevailing in this
system, enones 12 and 20 are further autoxidized,17a
presumably via allylic hydrogen abstraction, ultimately
yielding hydroxyenones 32/42 and 33/43 (Scheme 6).
These hydroxyenones are clearly not generated via a
singlet oxygen mechanism (cf. Scheme 4), since the data
As shown in Scheme 7, similar transformations in the
case of alkylidenecyclopropyl hydroperoxides 28/38 could
easily lead to any of the remaining products, dienones
34-36/44-45, depending on which hydrogens are lost.
D. Absen ce of 1O2 P r od u ct. We have thus demon-
strated that the photooxidation of tetrasubstituted cy-
clopropenes 14a , 14b, and 26 occurs via a “type I” free-
radical process. The question remains, however, as to
why no1O2 ene reaction is observed. Two possible ap-
proaches are available. We have previously suggested1b,1e
that the interatomic distance between the R-olefinic
carbon and the γ-allylic hydrogen, the distance which the
1
of Table 2 indicate that the addition of the O2 quencher
DABCO has only a small effect if any on the product
distribution. Table 2 further reveals that, in the presence
of the radical chain-breaker, 2,6-di-tert-butylphenol, the
rate of oxygen uptake is strongly inhibited; nevertheless,
of the small amount of reaction that does occur, the short
chain epoxide-derived enone product 20 predominates
and is not further oxidized. This is consistent with the
mechanism of Scheme 6.
1
attacking O2 must span in the ene reaction irrespective
of mechanism, may be a pivotal consideration in deter-
mining whether the hydrogen is abstractable. Too great
a distance may well place legitimate candidates “out of
reach”.
Using bond lengths and angles from literature data,
we have calculated the CR-Hallylic distance for isobutylene,1e
methylenecyclobutane,1e methylcyclobutene,26 methylene-
cyclopropane,1e and methylcyclopropene.27 It should be
noted that in these calculations, the preferred 1O2 ene
reaction orientation6 was used whenever possible, i.e., the
(25) See the following leading papers and the references therein:
(a) Maynard, G. D.; Paquette, L. A. J . Org. Chem. 1991, 56, 5480-
5482. (b) Friedrich, L. E.; Leckonby, R. A.; Stout, D. M. J . Org. Chem.
1980, 45, 3198-3202. (c) Crandall, J . K.; Conover, W. W., II. J . Org.
Chem. 1978, 43, 1323-1327. (d) refs 1d and 9.
(26) Based on the electron diffraction data of Shand and co-
workers,26a the CR-Hallylic distance is calculated as 3.013 Å. MINDO 3
calculations by Dewar et al.26b give a substantially higher value of 3.164
Å. In Scheme 8, we have given the average value of 3.088 Å. (a) Shand,
W., J r.; Schomaker, V.; Fischer, J . F. J . Am. Chem. Soc. 1944, 66, 636-
640. See also: Huang, Y. S.; Beaudet, R. A. J . Chem. Phys. 1970, 52,
935-940. (b) Bingham, R. C.; Dewar, M. J . S.; Lo, D. H. J . Am. Chem.
Soc. 1975, 97, 1294-1301.
(27) The microwave spectral data of Kemp and Flygare27a yields a
CR-Hallylic distance of 3.241 Å. MINDO 3 calculations by Dewar et al.26b
give almost the same value of 3.238 Å. IMOA calculations of Eckert-
Makasic and Makasi27b yield a substantially lower value of 3.229 Å.
In Scheme 8, we have cited the experimental value of 3.241 Å. (a)
Kemp, M. K.; Flygare, W. H. J . Am. Chem. Soc. 1969, 91, 3163-3167.
See also: Allen, F. A. Tetrahedron 1982, 38, 645-655. (b) Eckert-
Makasic, M.; Makasic, Z. B. J . Mol. Struct., THEOCHEM 1982, 86,
325-340.

Thermolysis and Oxygenation of Substituted Cyclopropenes
J . Org. Chem., Vol. 65, No. 6, 2000 1813
Sch em e 8. Ca lcu la ted Cr-Ha llylic Dista n ces for
singlet oxygenation has slowed to a point where free
radical processes compete favorably.31 Future work will
explore the effect of IP lowering substituents on the
course of this reaction.
Isobu tylen e a n d Va r iou s Cycloolefin s
Exp er im en ta l Section
NMR spectra were obtained on 600 (designated as NMR*),
300 and 200 Fourier transform spectrometers, using TMS as
the internal standard. Assignments (see Supporting Informa-
tion) were facilitated by correlating proton and carbon chemical
shifts through analysis of residual couplings in off-resonance
decoupled spectra. The carbon numbering of the various
compounds used in the nomenclature and spectral assignments
is shown in Schemes 1 and 3. EI and CI mass spectra were
measured at 70 eV, unless otherwise indicated. Analytical thin-
layer chromatography (TLC) was performed using Merck silica
gel microcards. Preparative runs (PTLC) were carried out on
Merck silica gel F254 precoated plates, and the products were
extracted from the silica by stirring overnight in a solution of
10% CH3OH in CHCl3. The retention times given are for the
analytical runs. Column chromatography was performed on
Merck silica (230-400 mesh). For preparative gas chroma-
tography (GC), a thermal conductivity detector was used, and
peak areas were determined by triangulation.
Ta ble 3. Ion iza tion P oten tia l a n d Rela tive 1O2
Rea ctivity of Sm a ll Rin g Cycloa lk en es
Compounds 8-14,9 22-26,11 4332 and 47a -c33 are all known.
P r ep a r a tion of E- a n d Z-2-(3′-Meth yl-2′-bu tylid en e)-
cycloh ep ta n on e (12a ), E- a n d Z-2-(2′-Bu tylid en e)cyclo-
h ep t a n on e (12b ), a n d 2-Isop r op ylid en ecyclooct a n on e
(20). (a ) F r ied r ich P r oced u r e. The compounds 12a and 12b
were prepared as previously described by Friedrich and co-
workers9 (Scheme 1), scaled up 5-fold. Two modifications are
worth mentioning. The conversion of vinyl chloride 9 to ketal
10 was accomplished as follows: A three-necked round-bottom
flask, fitted with a reflux condenser topped with a nitrogen
inlet, was charged under with ethylene glycol (15 g; 242 mmol),
p-dioxane (18 mL), and freshly cut sodium pieces (0.72 g; 36.3
mmol). The mixture was heated under reflux (oil bath tem-
perature up to 165-170 °C) for 4 h, at which time vinyl
chloride 9 (a or b; 20 mmol) was added dropwise to the
refluxing solution. Heating and stirring under nitrogen were
continued for another 40 h; workup and distillation, as
described by Friedrich,9 gave the desired product 10 (ca. 11-
15 mmol; 55-75% yield). We also note that the dehydration-
deprotection of â-hydroxyketals 11 generating enones 12 is
accompanied by a 50% yield of cycloheptanone, presumably
formed in a retroaldol process from the â-hydroxyketone
intermediate 16 (see next paragraph). For enones 12, the Z to
E ratio is 1:2 in the case of 12a , and 1:1 in the case of 12b.
We succeeded in purifying and separating the Z- and E-
isomers of 12a and 12b from each other either by GC34a [at
145 °C, and with a flow rate of 75 mL/min (retention times:
27 and 30.5; 29.5 and 32 min. respectively)] or by silica column
chromatography (TLC retention times given in data below).
The previously published9 NMR spectral data for compounds
8-12 were obtained on a 60 MHz instrument and the corre-
sponding 13C data are lacking. In addition, CW, 2D-COSY, and
NOE experiments were performed to assist in atom and
configuration assignments. Hence, the improved and missing
data are recorded below.
abstracted allylic hydrogen was aligned in a plane
perpendicular to the plane of the double bond. The
necessary bond rotations were accomplished using Gauss-
ian 76.28 As shown in Scheme 8, the reactive cycloolefins
all have a CR-Hallylic distance below 3.09 Å, while for
those which are unreactive this value is above 3.24 Å.
As suggested above, it may be these crucial 0.15 Å which,
in the latter case, place the abstractable allylic hydrogen
“out of reach”.
An additional, perhaps more convincing, rationale is
1
available to explain the lack of O2 reactivity in these
cyclopropene systems. Several authors have found a good
general correlation between olefin HOMO energies (or
1
ionization potentials) and the relative rates of O2 ene
reaction,6,29 i.e., the higher E(olefin-HO) (or the lower the
IP) the greater the rate. As shown in Table 3, the
ionization potential (IP) of cyclopropene is substantially
higher than that of its homologues.30 We suggest, there-
fore, that in the case of cyclopropene, the rate of the
(28) Binkley, J . S.; Whiteside, R. A.; Hariharan, B. C.; Seeger, R.;
Pople, J . A.; Hehre, W. J .; Newton, M. D., Gaussian Inc., Pittsburgh,
PA, 1976.
(29) (a) Kearns, D. R. J . Am. Chem. Soc. 1969, 91, 6554-6563. (b)
Paquette, L. A.; Liotta, D. C.; Baker A. D. Tetrahedron Lett. 1976,
2681-2684. (c) Adam, W.; Carballiera, N.; Cheng, C.-C.; Sakanishi,
K.; Gleiter, R. J . Org. Chem. 1979, 44, 851-853. (d) van den Heuvel,
C. J . M.; Verhoeven, J . W.; de Boer, Th. Recl. Trav. Chim. Pays-Bas
1980, 99, 280-284. (e) Hurst, J . R.; Schuster J . Am. Chem. Soc. 1982,
104, 6854-6856. (f) Reference 6b, Table 2, pp 291-295 and 298ff. (g)
Monroe, B. M. In Singlet O2 - Volume I: Physical-Chemical Aspects;
Frimer, A. A., Ed.; Chemical Rubber Company: Boca Raton, FL, 1985;
pp 177-224; see especially pp 201-206. (h) Reference 1h, p 71. (i) We
note, however, that the IP/rate correlation is far from perfect,29e-h
primarily because ionization potential is by no means the sole
determinant of reaction rate.
(b) Silyl En ol Eth er Cr ossed Ald ol Meth od . As outlined
in Scheme 2, the title compounds could be prepared from silyl
enol ethers 15 and 18, which were prepared in turn via the
procedure of Fleming and Paterson.10a The TiCl4-mediated
condensation of the silyl ethers with the appropriate ketone
in dichloromethane, according to the procedure of Mukaiyama
and Narasaka,10b yielded the desired â-hydroxyketones 16a ,
(30) (a) Bischof, P.; Heilbronner, E. Helv. Chim. Acta 1970, 53,
1677-1682. (b) Wiberg, K. B.; Ellison, G. B.; Wendoloski, J . J .; Brundle,
C. R.; Kuebler, N. A. J . Am. Chem. Soc. 1976, 98, 7179-7182. (c)
Turner, D. W. In Molecular Photoelectron Microscopy; Wiley-Inter-
science: New York, 1970. (d) Aue, D. H.; Mishishnek, M. J .; Shelham-
mer, D. F. Tetrahedron Lett. 1973, 4799-4802.
(32) Maignan, C.; Rouessac, F. Bull. Chim. Soc. Fr. 1974, 2035-
2039.
(33) Freerksen, R. W.; Selikson, S. J .; Wroble, R. R.; Kyler, K. S.;
Watt, D. S. J . Org. Chem. 1983, 48, 4087-4096.
(34) (a) 13 ft × 1/4 in. copper column packed with 20% Carbowax
20M Chromosorb WAW. (b) 13 ft × 1/4 in. copper column packed with
7% SE-30 on Chromosorb WAW.
(31) Cf. J efford, C. W.; Rimbault, C. G. Tetrahedron Lett. 1981, 22,
91-94.

1814 J . Org. Chem., Vol. 65, No. 6, 2000
Frimer et al.
16b, and 19. It should be noted that both 16a and 16b were
obtained as a mixture of diastereomers and partially contami-
nated by cycloheptanone and enones 12a or 12b (presumably
formed via retroaldol and dehydration, respectively; see below).
Attempts to purify 16a and 16b by distillation merely in-
creased the amount of contamination. As a result, 1H NMR
data could not be assigned, though 13C NMR and HRMS data
could be readily extracted. Acid (H2SO4)-catalyzed dehydration
of the latter in methanol, following the method of Friedrich
and co-workers,9 yielded the corresponding enones 12a (Z:E
) 1:1), 12b (Z:E ) 1:1) and 20 in 50% yield. The residue was
predominantly the corresponding cycloheptanone or cyclo-
octanone, presumably regenerated via a retroaldol process.
Azeotropic dehydration of 19 in refluxing benzene (using
p-TsOH and a Dean-Stark trap) generated 20 in 75% yield,
E-12b: Rf (10% ethyl acetate in hexane) 0.35; 1H NMR
(CDCl3) δ 2.59-2.44 (m, 2H), 2.38-2.25 (m, 2H), 2.19 (2q, J 3′,4′
) 7 Hz, 2H), 1.77 (s, 3H), 1.76-1.59 (m, 6H), 1.02 (t, J 3′,4′ ) 7
Hz, 3H); 13C NMR (CDCl3) δ 209.50, 142.85, 136.31, 43.91,
30.67, 29.22, 28.38, 27.73, 24.56, 19.47, 12.20; HRMS calcd
(C11H18O, M+) 166.1358, obsd 166.1351.
15: 1H NMR* (CDCl3) δ 4.98 (t, J 2,3 ) 6 Hz, 1H), 2.19 (m,
2H), 1.96 (m, 2H), 1.65 (m, 2H), 1.53 (m, 2H), 1.49 (m, 2H),
0.14 (s, 9H); 13C NMR* (CDCl3) δ 155.93, 108.55, 35.45, 31.51,
27.74, 25.29, 25.17, 0.16; MS (CI, 70 ev) m/z 185 (MH+, 1.83%),
113 (MH+ - HSiMe3, 49.49%), 95 (MH+ - HOSiMe3, 14.58%);
HRMS (CI) calcd (C10H21OSi, MH+) 185.1361, obsd 185.1360.
16a : Mixture of diastereomers: MS (CI, 70 ev) m/z 199
(MH+, 14%), 181 (MH+ - H2O, 100%), 113 (MH+ - C5H10O,
89%); HRMS calcd (C12H23O2, M+) 199.1698, obsd 199.1670.
Major diastereomer: 13C NMR* (CDCl3) δ 219.35, 75.77, 55.81,
44.27, 35.08, 28.25, 27.94, 23.87, 23.57, 19.87, 16.94, 15.81.
Minor diastereomer: 13C NMR* (CDCl3) δ 218.86, 75.04, 55.66,
43.32, 33.07, 27.87, 27.72, 24.36, 23.63, 19.59, 17.57, 16.23.
16b: Mixture of diastereomers: MS (CI, 70 ev) m/z 185
(MH+, 15%), 167 (MH+ - H2O, 100%), 113 (MH+ - C4H8O,
38%); HRMS (CI) calcd (C11H21O2, MH+) 185.1541, obsd
185.1520. Major diastereomer: 13C NMR* (CDCl3) δ 218.79,
73.57, 57.99, 44.09, 32.95, 28.31, 28.20, 25.61, 23.81, 22.96,
7.55. Minor diastereomer: 13C NMR* (CDCl3) δ 218.20, 73.38,
57.42, 44.03, 31.34, 28.26, 28.11, 25.15 (C3), 24.25 (C6), 23.78
(C1′), 7.40 (C4′).
1
again accompanied by cyclooctanone In the H NMR below of
20, H1′ are the hydrogens on the isopropylidenyl methyl which
is cis to the carbonyl, while H3′ are those that are on the trans
methyl.
(c) Cyclop r op en e Ep oxid a tion P r oced u r e. Enone 20
was also synthesized via the MCPBA epoxidation of cyclopro-
pene 26 in CCl4. GC of the product mixture gave the desired
product along with an as yet unidentified isomer. The MCPBA
epoxidation of 14a and 14b is reported in the literature to yield
E- and Z-12a and 12b, respectively.9
8: 1H NMR (CDCl3) δ 5.94 (t, J 2,3 ) 6.5 Hz, 1H), 2.57-2.52
(m, 2H), 2.11 (m, 2H), 1.71-1.57 (m, 8H); 13C NMR (CDCl3) δ
136.16, 128.77, 38.38, 30.72, 27.79, 26.71, 25.93.
9a : 1H NMR (CDCl3) δ 6.17 (dd, J 1,7a ) 6 Hz, J 1,7b ) 7 Hz,
18: 1H NMR* (CDCl3) δ 4.70 (t, J 2,3 ) 8 Hz, 1H), 2.14 (m,
2H), 1.97 (m, 2H), 1.55 (m, 2H), 1.49 (m, 2H), 1.46 (m, 4H),
0.16 (s, 9H); 13C NMR* (CDCl3) δ 153.05, 105.45, 31.05, 30.98,
27.83, 26.40, 26.36, 25.52, 0.45; MS (EI, 70 ev) m/z 198 (M+,
26.19%), 183 (M+ - CH3, 36.00%), 115 (M+ - CHCOSiMe3,
29.16%); HRMS calcd (C11H22OSi, M+) 198.1439, obsd 198.1400.
1H), 3.77 (dd, J 3,4a ) 4 Hz, J 3,4b ) 6, 1H), 2.90 (sept, J 2′,3′
)
J 2′,3′′ ) 7 Hz, 1H), 2.22-1.32 (m, 8H), 1.14 (d, J 2′,3′ ) 7, 3H),
1.10 (d, J 2′,3′′ ) 7 Hz, 3H); 13C NMR (CDCl3) δ 211.99, 132.29,
132.20, 59.34, 38.96, 27.90, 26.78, 26.11, 25.96, 18.83, 18.46.
9b:1H NMR (CDCl3) δ 6.16 (t, J 1,7 ) 6.6 Hz, 1H), 3.54 (dd,
J 3,4a ) 3.5 Hz, J 3,4b ) 6.5 Hz, 1H), 2.62 and 2.60 (ABq of q,
J gem ) 18.5 Hz, J 2′,3′ ) 7 Hz, 2H), 1.85-1.42 (m, 8H), 1.08 (t,
J 2′,3′ ) 7 Hz, 3H); 13C NMR (CDCl3) δ 208.07, 132.44, 132.06,
60.93, 34.28, 27.98, 26.97, 26.08, 25.96, 7.83.
19: 1H NMR* (CDCl3) δ 3.52 (bs, 1H, OH), 2.55 (dd, J 2,3a
)
4 Hz, J 2,3b ) 11 Hz, 1H, H2), 2.12 (m, 2H), 1.62-1.25 (m, 10H),
0.93 (s, 3H), 0.92 (s, 3H); 13C NMR* (CDCl3) δ 222.61, 71.13,
56.38, 45.39, 29.04, 28.36, 27.91, 26.73, 24.69, 24.64, 22.83;
MS (CI, 70 ev) m/z 185 (MH+, 100%), 167 (MH+ - H2O, 56%);
HRMS (CI) calcd (C11H21O2, MH+) 185.1541, obsd 185.1560.
20: 1H NMR (CDCl3) δ 2.53 (inv quint, J ) 3 Hz, 2H), 2.50-
2.45 (m, 2H), 1.84 (s, 3H), 1.79 (s, 3H), 1.83-1.75 (m, 2H),
1.60-1.51 (m, 4H), 1.47-1.42 (m, 2H); 13C NMR (CDCl3) δ
212.48, 136.66, 136.06, 49.51, 29.05, 27.61, 27.22, 26.65, 25.72,
10a : 1H NMR (CDCl3) δ 3.87 (s, 4H), 3.18 (dd, J 6,7a ) 4 Hz,
J 6,7b ) 10 Hz, 1H), 2.80 (sept, J 2′,3′ ) J 2′,3′′ ) 7 Hz, 1H), 2.10-
1.30 (m, 10H),1.04 (d, J 2′,3′ ) J 2′,3′′ ) 7 Hz, 6H); 13C NMR
(CDCl3) δ 214.79, 112.76, 64.51, 63.94, 57.05, 41.96, 37.48,
28.27, 26.63, 26.51, 22.47, 18.36, 17.58.
22.70, 21.23; IR (CDCl3) 1667 (s, CdO), 1600 (m, CdC) cm-1
;
10b: 1H NMR (CDCl3) δ 3.89-3.84 (m, 4H), 2.99 (dd, J 6,7a
) 10 Hz, J 6,7b ) 4 Hz, 1H), 2.62 and 2.40 (ABq of q, J gem ) 18
Hz, J vic ) 7 Hz, 2H), 1.88-1.50 (m, 10H), 1.02 (t, J 2′,3′ ) 7 Hz,
3H); 13C NMR (CDCl3) δ 211.62, 112.65, 64.58, 63.86, 59.20,
37.43, 36.87, 26.88, 26.85, 26.12, 22.39, 7.82.
MS (EI, 70 eV) m/z 166 (M+, 10.85%), 95 (M+ - CO-CH3-
CCH3-H, 16.48%), 82 (COCC(CH3)2, 34.96%), 67 (COCCCH3,
42.92%); MS (CI, 70 ev) m/z 167 (MH+, 100%), 151 (MH+
-
CH3, 87.00%); HRMS (CI) calcd (C11H18O, MH+) 167.1435, obsd
167.1430.
11a : 1H NMR (CDCl3) δ 4.10-3.72 (m, 4H), 2.80 (sept, J 2′,3′
) J 2′,3′′ ) 7 Hz, 1H), 2.50-1.30 (m, 11H), 1.1 (s, 3H), 0.93 (d,
J 2′,3′ ) 7 Hz, 3H), 0.82 (d, J 2′,3′′ ) 7 Hz, 3H).
P r ep a r a t ion of ∆1,7-8-Isop r op ylb icyb lo[5.1.0]oct a n e
(14a ) a n d ∆1,7-8-Eth ylbicyblo[5.1.0]octa n e (14b). The title
compounds were prepared from alkylidenecycloheptanones 12a
and 12b, respectively, as previously described by Friedrich and
co-workers9 (see Scheme 1), with the following minor modifica-
tions. After the preparation of tosylhydrazones 13 in methanol,
the solvent was rotary evaporated at room temperature and
the colored residue recrystallized from CH2Cl2/pentane. The
white crystals were washed with methanol chilled to -70 °C.
Cyclopropanes 14a and 14b (retention times: ca. 16 min) were
separated by GC34b at 90 °C (injector and detector at 100 °C)
with a flow rate of 200 mL/min. Compounds 13 and 14 are
known;9 the substantially improved and missing spectral data
are recorded below.
11b: 1H NMR (CDCl3) δ 4.13-3.81 (m, 4H), 2.99 (dd, J 6,7a
) 10 Hz, J 6,7b ) 4 Hz, 1H), 2.45 (ABq of q, J gem ) 18 Hz, J vic
2
′,3′ ) 7 Hz, 2H), 1.89-1.26 (m, 10H), 1.16 (s, 3H), 0.89 (t, J 2′.3′
) 7 Hz, 3H); 13C NMR (CDCl3) δ 112.66, 74.20, 63.70, 58.20,
44.86, 31.58, 28.86, 28.55, 25.94, 25.08, 24.01, 7.90.
Z-12a : Rf (5% ethyl acetate in hexane) 0.20; 1H NMR
(CDCl3) δ 2.89 (sept, J 3′,4′ ) 7 Hz, 1H), 2.51-2.38 (m, 2H),
2.27-2.19 (m, 2H), 1.75-1.63 (m, H4, 6H), 1.62 (s, 3H), 0.96
(d, J 3′,4′ ) 7 Hz, 6H); 13C NMR (CDCl3) δ 210.84, 143.20, 136.63,
43.72, 31.42, 30.28, 28.79, 28.15, 24.35, 20.95, 11.63; HRMS
calcd (C12H20O, M+) 180.1514, obsd 180.1509.
E-12a : Rf (5% ethyl acetate in hexane) 0.27; 1H NMR
(CDCl3) δ 2.85 (sept, J 3′,4′ ) 7 Hz, 1H), 2.55-2.41 (m, 2H),
2.37-2.19 (m, 2H), 1.73-1.60 (m, 6H), 1.59 (s, 3H), 1.01 (d,
J 3′,4′ ) 7 Hz, 6H); 13C NMR (CDCl3) δ 210.94, 144.01, 135.82,
43.74, 30.46, 29.81, 29.16, 27.99, 24.33, 20.41, 13.37; HRMS
calcd (C12H20O, M+) 180.1514, obsd 180.1511.
Z-12b: Rf (10% ethyl acetate in hexane) 0.27; 1H NMR
(CDCl3) δ 2.57-2.44 (m, 2H), 2.38-2.28 (m, 2H), 2.12 (2q, J 3′,4′
) 7 Hz, 2H), 1.85 (s, 3H), 1.77-1.58 (m, 6H), 1.05 (t, J 3′,4′ ) 7
Hz, 3H); 13C NMR (CDCl3) δ 209.50, 142.36, 136.87, 43.81,
30.67, 28.97, 28.68, 28.51, 24.45, 18.03, 13.31; HRMS calcd
(C11H18O, M+) 166.1358, obsd 166.1355.
13a , Ma jor isom er : 1H NMR (CDCl3) δ 7.92-7.81 (m, 2H),
7.43-7.26 (m, 2H), 5.55 (s, 1H), 2.89 (sept, J 3′,4′ ) J 3′,4′′ ) 7
Hz, 1H), 2.45 (s, 3H), 2.25-1.35 (m, 10H), 1.53 (s, 3H), 0.96
(d, J 3′,4′ ) J 3′,4′′ ) 7 Hz, 6H); Min or isom er : 1H NMR (CDCl3)
δ 7.92-7.81 (m, 2H), 7.43-7.26 (m, 2H), 5.55 (s, 1H), 2.89
(sept, J 3′,4′ ) J 3′,4′′ ) 7 Hz, 1H), 2.45 (s, 3H), 2.25-1.35 (m,
10H), 1.13 (s, 3H), 0.69 (6H, d, J 3′,4′ ) J 3′,4′′ ) 7 Hz); Z-E
m ixtu r e: 13C NMR (CDCl3) δ 162.39, 143.73, 137.80, 137.48,
135.85, 128.09, 126.72, 36.22-25.83, 21.55, 18.43, 12.39.
13b, Ma jor isom er : 1H NMR (CDCl3) δ 7.88-7.82 (m, 2H),
7.30-7.27 (m, 3H), 2.46 (s, 3H), 2.25-1.40 (m, 10H), 2.03 (q,
J 3′,4′ ) 7 Hz, 1H), 1.63 (q, J 3′,4′ ) 7 Hz, 1H), 1.62 (s, 3H), 0.97

Thermolysis and Oxygenation of Substituted Cyclopropenes
J . Org. Chem., Vol. 65, No. 6, 2000 1815
(t, J 3′,4′ ) 7 Hz, 3H); Min or isom er : 1H NMR (CDCl3) δ 7.88-
7.82 (m, 2H), 7.30-7.27 (m, 3H), 2.46 (s, 3H), 2.25-1.40 (m,
10H), 2.03 (q, J 3′,4′ ) 7 Hz, 1H), 1.63 (q, J 3′,4′ ) 7 Hz, 1H), 1.23
(s, 3H), 0.65 (t, J 3′,4′ ) 7 Hz, 3H); Z-E m ixtu r e: 13C NMR
(CDCl3) δ 162.39, 143.73, 137.80, 137.48, 135.85, 128.09,
126.72, 36.22-25.83, 21.55, 18.43, 12.39.
When 14a was injected into the GC,34b with the injector and
detector set for 190 °C and the column at 120 °C, the products
E- and Z-29a and 30a were obtained essentially in the ratio
of Table 1. Similarly, heating a neat sample of 14a to 170 °C
for 0.5 h gave the same results. When 14b was injected into
the GC34b as above, the products E- and Z-29b were obtained
exclusively, in a 1:1 ratio; heating a neat sample of 14b to 170
°C for 0.5 h yielded equivalent amounts of 30b and 31b.
29a (E/Z m ixtu r e): MS (CI, methane, 70 eV) m/z 181 (MH+
+ CH4, 72.6%), 165 (MH+, 47.9%), 149 (MH+ - CH4, 36.3%),
109 (MH+ - C4H8, 51.3%), 95 (MH+ - C5H10, 100%); HRMS
exact mass calcd for C12H20 (M+) 164.1565, found 164.1644.
Z-29a : 1H NMR (CDCl3) δ 6.39 (d, J 1,2 ) 11 Hz, 1H), 5.61 (dt,
J 1,2 ) 11 Hz, J 1,7 ) 5 Hz, 1H), 2.92 (sept, J 3′,4′′ ) 7 Hz, 1H),
2.38 (t, J 4,5 ) 6 Hz, 2H), 2.18 (m, 2H), 1.69-1.60 (m, 4H), 1.58
(s, 3H), 0.94 (d, J 3′,4′ ) 7 Hz, 6H); 13C NMR (CDCl3) δ 137.22,
131.19, 129.79, 129.60, 30.90, 28.52, 28.46, 27.00, 26.08, 20.54,
11.30. E-29a : 1H NMR (CDCl3) δ 6.30 (d, J 1,2 ) 11 Hz, 1H),
5.64 (dt, J 1,2 ) 11 Hz, J 1,7 ) 5 Hz, 1H), 2.94 (sept, J 3′,4′ ) 7 Hz,
1H), 2.42 (t, J 4,5 ) 6 Hz, 2H), 2.18 (m, 2H), 1.69-1.60 (m, 4H),
1.58 (s, 3H), 0.96 (d, J 3′,4′ ) 7 Hz, 6H); 13C NMR (CDCl3) δ
137.22, 132.35, 129.79, 129.60, 30.26, 29.90, 29.51, 27.96,
26.25, 21.05, 12.20.
14a : 1H NMR (CDCl3) δ 2.50-2.18 (m, 4H), 1.93-1.43 (m,
7H), 1.10 (s, 3H), 0.75 (d, J 1′,2′ ) 6.7 Hz, 6H); 13C NMR (CDCl3)
δ 130.14, 36.40, 31.98, 29.71, 28.18, 26.69, 22.71, 21.09.
14b: 1H NMR (CDCl3) δ 2.41-2.22 (m, 4H), 1.87-1.77 (m,
4H), 1.61-1.43 (m, 2H), 1.56 (q, J 1′,2′ ) 7.5 Hz, 2H), 1.14 (s,
3H), 0.69 (t, J 1′,2′ ) 7.5 Hz, 3H); 13C NMR (CDCl3) δ 131.41,
32.83, 31.99, 29.71, 28.26, 26.64, 24.42, 12.04.
Gen er a l P h otooxid a tion P r oced u r e. The previously
described35 photooxidation apparatus was flushed with oxygen
and charged with solvent (1-5 mL) containing substrate (0.2-
8.0 mmol) as well as a spatula tipful of polymer-based Rose
Bengal (p-RB, Dye Tel Inc., POB 23, Perrysburg, Ohio) to serve
as photosensitizer. The photooxidation (λ > 360 nm) was
allowed to proceed until an 1 equiv of O2 had been absorbed
or until oxygen uptake had essentially ceased. At times, it was
necessary to stop the photooxidation and add new p-RB,
because the intense red color of the sensitizer had faded to a
pale pink. At the conclusion of the photooxidation, 1.1 equiv
of triphenylphosphine was added to the reaction mixture to
reduce any hydroperoxides present to the corresponding al-
cohols. The product solution was filtered through a cotton plug
to remove the polymer-based sensitizer, and concentrated prior
to gas or column chromatography. Where indicated, fractions
29b (E/Z m ixtu r e): MS (CI, methane, 70 eV) m/z 151 (MH+,
100%), 135 (M+ - CH3, 6.8%), 109 (M+ - C3H5, 21.3%), 95 (M+
- C4H7, 67.2%), 81 (M+ - C5H9, 41.3%); HRMS exact mass
calcd for C11H18 (M+) 150.1216, found 150.1408. E-29b: 1H
NMR* (CDCl3) δ 6.30 (d, J 1,2 ) 11 Hz, 1H), 5.63 (dt, J 1,2 ) 11
Hz, J 1,7 ) 5.5 Hz, 1H), 2.39 (q, J ) 5.5 Hz, 2H), 2.19 (m, 2H),
1
proved to be a mixture of isomers; the H and 13C NMR data
2.09 (m, 2H), 1.71 (s, 3H), 1.70-1.60 (m, 4H), 0.99 (t, J 3′,4′ )
was extracted from this mixture. The products were identified
by their spectral data and, where possible, by independent
synthesis (vide infra). Once all the products had been identi-
fied, the product distribution was determined based on the
spectrum of the crude reaction mixture. DABCO (10-5 to 10-3
M)21b,c and DTBP (0.4 M) were added to the photooxidation
7.5 Hz, 3H); 13C NMR (CDCl3) δ 133.97, 131.72, 130.55, 129.82,
30.01, 28.31, 27.83, 27.13, 26.50, 18.08, 13.26. Z-29b: 1H NMR*
(CDCl3) δ 6.34 (d, J 1,2 ) 11 Hz, 1H), 5.61 (dt, J 1,2 ) 11 Hz, J 1,7
) 5.5 Hz, 1H), 2.38 (q, J ) 5.5 Hz, 2H), 2.19 (m, 2H), 2.10 (q,
J 3′,4′ ) 7.5 Hz, 2H), 1.71 (s, 3H), 1.70-1.60 (m, 4H), 0.96 (t,
J 3′,4′ ) 7.5 Hz, 3H); 13C NMR (CDCl3) δ 134.00, 131.01, 130.46,
129.72, 30.59, 28.31, 27.62, 26.90, 26.31, 17.42, 12.66.
1
solution to determine, respectively, whether O2 or free radical
30a : 1H NMR (CDCl3) δ 5.85 (t, J 2,3 ) 7 Hz, 1H), 4.84 (d,
J 1′,1′′ ) 1.5 Hz, 1H), 4.70 (t, J 1′′,3′ ) J 1′,1′′ ) 1.5 Hz, 1H), 2.51
(m, 1H), 2.26 (m, 2H), 2.16 (m, 2H), 1.76 (m, 2H), 1.51 (m,
4H), 1.04 (d, J 3′,4′ ) 7 Hz, 6H); 13C NMR (CDCl3) δ 158.84,
146.10, 127.76, 105 0.86 (C1′), 98.94, 32.70, 31.36, 30.32, 26.67,
processes, respectively, were involved.
P h otooxid a tion of Cyclop r op en es 14a , 14b, a n d 26. The
title cyclopropenes were dissolved in solvent (CHCl3 for 14a
and 14b, CH3CN for 20) and irradiated, according to the above
general oxidation procedure. The addition of Ph3P generated
no heat and had little if any effect on the product distribution;
hence, workup of the reaction mixtures proceeded without
prior addition of Ph3P. The products were separated by
preparative GC33a (oven: 140 °C; injector and detector: 200
°C; flow: 75 mL/min.). In the case of bicyclooctenes 14a and
14b, both dienes and enones were obtained. The relative total
amount of dienes isolated by GC, as compared to the total
amount of enones observed, varied from photooxidation to
photooxidation; it was simply proportional, however, to the
amount of unreacted substrate remaining (as approximated
by the amount of oxygen uptake). Product distribution within
each group is shown in Tables 1 and 2. Dienes 30a , 30b, and
31b were independently synthesized as described below.
Compound 43 is known;32 the substantially improved and
missing spectral data are recorded below. Compound 43 was
also independently synthesized via the Grignard addition of
isopropenylmagnesium bromide32 to cyloocta-1,2-dione.36 At-
tempted dehydration of 43 with thionyl chloride and pyridine37
yielded only a few percent of 44. The carbon numbering in the
data below is as shown in Scheme 3. Z/ E assignments were
based on NOE experiments. In 34a and 34b, no 13C NMR
carbonyl absorption was observed.
26.90, 22.33; MS (EI, 70 eV) m/z 164 (M+, 100%), 149 (M+
-
CH3, 20.3%), 135 (M+ - C2H5, 5.6%), 121 (M+ - C3H7, 58.18%),
107 (M+ - C4H9, 38.96%), 93 (M+ - C5H11, 19.07%); HRMS
(CI) calcd (C12H21, MH+) 165.1600, obsd 165.1643.
30b: 1H NMR* (CDCl3) δ 5.95 (t, J 2,3 ) 7 Hz, 1H), 4.92 (d,
J 1′,1′′ ) 1 Hz, 1H), 4.78 (dt, J 1′,1′′ ) 1 Hz, 1H, J 1′′,3′ ) 0.5 Hz),
2.23 (m, 4H), 2.19 (m, 2H), 1.76 (m, 2H), 1.49 (m, 2H), 1.47
(m, 2H), 1.04 (t, J 3′,4′ ) 7.5 Hz, 3H); 13C NMR (CDCl3) δ 152.10,
144.74, 128.14, 107.87, 32.62, 30.00, 28.42, 26.98, 26.78, 26.51,
13.42; MS (EI, 70 eV) m/z 151 (MH+, 35.86%), 150 (M+, 8.11%),
135 (M+ - CH3, 6.56%), 121 (M+ - C2H5, 9.78%), 111 (M+
-
C3H7, 16.56%), 109 (M+ - C3H5, 19.51%), 97 (M+ - C4H9,
14.12%), 95 (M+ - C4H7, 53.00%); HRMS (CI) calcd (C11H19
,
MH+) 151.1430, obsd 151.1486.
31b: 1H NMR* (CDCl3) δ 5.84 (t, J 2,3 ) 7 Hz, 1H), 5.54 (q,
J 3′,4′ ) 6 Hz, 1H), 2.32 (m, 2H), 2.18 (m, 2H), 1.75 (m, 2H),
1.74 (s, 3H), 1.68 (d, J 3′,4′ ) 6 Hz, 3H), 1.47 (m, 2H) 1.46 (m,
2H); 13C NMR (CDCl3) δ 147.26, 137.70, 126.36, 118.60, 32.77,
29.48, 28.45, 27.01, 26.71, 14.12, 14.09; MS (EI, 70 eV) m/z
151 (MH+, 35.86%), 150 (M+, 8.11%), 135 (M+ - CH3, 6.56%),
121 (M+ - C2H5, 9.78%), 111 (M+ - C3H7, 16.56%), 109 (M+
-
C3H5, 19.51%), 97 (M+ - C4H9, 14.12%), 95 (M+ - C4H7,
53.00%); HRMS (CI) calcd (C11H19, MH+) 151.1450, obsd
151.1486.
(35) Frimer, A. A.; Bartlett, P. D.; Boschung, A. F.; J ewett, J . G. J .
Am. Chem. Soc. 1977, 99, 7977-7986.
32a : 1H NMR (CDCl3) δ 6.61 (t, J 3,4 ) 6.5 Hz, 1H), 2.61
and 2.57 (m, J gem ) 14 Hz, 2H), 2.41 (m, 2H), 1.98 (sept, J 3′,4′
) J 3′,4′′ ) 7 Hz, 1H), 1.80 (m, 2H), 1.69 (m, 2H), 1.23 (s, 3H),
0.93 (d, J 3′,4′ ) 7 Hz, 3H), 0.79 (d, J 3′,4′′ ) 7 Hz, 3H); 13C NMR
(CDCl3) δ 141.2, 139.34, 43.42, 35.97, 27.51, 24.64, 22.25, 21.90,
(36) (a) The dione was prepared according to Wittig, G.; Krebs, A.
Chem. Ber. 1961, 94, 3260 and purified by column chromatography
[Rf (50% acetone in petetroleum ether) 0.70]. See also: Hach, C. C.;
Banks, C. V.; Diehl, H. Organic Syntheses; Wiley: New York, 1963;
Coll. Vol. IV, pp 229-232. (b) Spectral data is reported in Amon, C.
M.; Banwell, M. G.; Gravatt, G. L. J . Org. Chem. 1987, 52, 4851-
4855. See also: Bauer, D. P.; Macomber, R. S. J . Org. Chem. 1975,
40, 1990-1992.
18.24, 17.02; MS (EI, 70 eV) m/z 196 (M+, 1.1%), 178 (M+
-
H2O, 4.08%), 163 (M+ - H2O-CH3, 2.98%), 153 (M+ - C3H7,
100%), 125 (M+ - C4H7O, 8.15%), 111 (M+ - C5H9O, 15.0%),
97 (M+ - C6H11O, 25.23%), 83 (M+ - C7H13O, 24.49%), 69 (M+
(37) Alexandre, C.; Rouessac, F. Bull. Chim. Soc. Fr. 1977, 117-
119.

1816 J . Org. Chem., Vol. 65, No. 6, 2000
Frimer et al.
- C8H15O, 37.75%); HRMS calcd (C12H20O2, M+, 1.1%) 196.1463,
obsd 196.1081; HRMS calcd (C9H13O2, M+ - C3H7, 100%)
153.0915, obsd 153.0914.
and elute together with the â-chloroketones 46. The stability
of the latter increases as the alkyl group on the ketone becomes
more bulky; thus 46a is markedly more stable than 46c, and
only a small amount of 47a is generated on chromatography.
The â-chloroketones 46 were primarily characterized by their
molecular weight (HRMS) and by the characteristic NMR data
for H-2, i.e., the hydrogen R to the carbonyl and â to the
chlorine, which consistently appears at 4.45 ppm as a ddd. The
large J 1,2 ) 9.8 Hz in 46a indicates that H1 and H2 are quasi-
trans a,a to each other (i.e., that the chlorine and acyl group
are in a quasi trans e,e relationship). Finally, reacting enones
47a , 47b, and 47c, respectively, with methyl, ethyl, and sec-
butyltriphenylphosphonium bromide and BuLi in THF gener-
ated the corresponding dienes 30a , 30b, and 31b. The Wittig
reagents were commercially available (Aldrich) or prepared
by heating a 1:1 mixture of the bromides and Ph3P in a sealed
ampule overnight at 150 °C; the resulting salt was recrystal-
lized from ethanol and vacuum-dried. Interestingly, in the case
of the preparation of 31b, NMR studies on the isolated product
revealed it to be ca. 80% in the E-conformation (31b), and 20%
in the corresponding Z-conformation (31c; see eq 12).
32b: 1H NMR (CDCl3) δ 6.63 (t, J 3,4 ) 6 Hz, 1H), 4.10-
4.00 (m, 1H), 2.60 (t, J 6,7 ) 7.5 Hz, 1H), 2.58 (t, J 6,7 ) 7.5 Hz,
1H), 2.45-2.38 (m, 2H), 1.88-1.80 (m, 2H), 1.79-1.71 (m, 2H),
1.69 (q, J 3′,4′ ) 7.5 Hz, 2H), 1.33 (s, 3H), 0.81 (t, J 3′,4′ ) 7.5 Hz,
3H); 13C NMR (CDCl3) δ 209.25, 147.16, 139.36, 75.65, 43.34,
34.42, 27.35, 26.29, 24.57, 22.03, 8.93; FTIR (CDCl3) 3422 (bs,
OH), 1710 (s, CdO), 1661 (s, CdC) cm-1; MS (EI, 70 eV) m/z
182 (M+, 7.7%), 167 (M+ - CH3, 16.2%), 153 (M+ - CH2CH3,
100%); HRMS calcd (C11H18O2, M+) 182.1307, obsd 182.1281.
34a : 1H NMR (CDCl3) δ 6.47 (t, J 3,4 ) 6.5 Hz, 1H), 4.94 (t,
J 1′,1′′ ) J 1′,3′ ) 1.5 Hz, 1H), 4.87 (dd, J 1′,1′′ ) 1.5 Hz, J 1′′,3′ ) 1
Hz, 1H), 2.62 (inv quint, J ) 3 Hz, 2H), 2.44 (sept, J 3′,4′ ) 7
Hz, 1H), 2.40 (q, J 3,4 ) J 4,5 ) 6.5 Hz, 2H), 1.84 (m, 2H), 1.74
(m, 2H), 1.01 (d, J 3′,4′ ) 7 Hz, 6H); 13C NMR (CDCl3) δ 139.91,
135.77, 129.20, 111.40, 42.89, 31.41, 24.86, 24.69, 22.32, 21.79;
MS (CI, methane, 70 eV) m/z 179 (MH+); MS (EI, 70 eV) m/z
178 (M+, 12.3%), 135 (M+ - C3H7, 12.86%), 107 (M+ - C5H11
,
11.11%), 91 (21.43%); HRMS calcd (C12H18O, M+) 178.1358,
obsd 178.1353.
34b: 1H NMR (CDCl3) δ 6.38 (t, J 3,4 ) 7 Hz, 1H), 4.88 (m,
1H), 4.85 (m, 1H), 2.40-2.10 and 1.90-1.10 (m, 8H), 2.06 (q,
J 3′,4′ ) 7.5 Hz, 2H), 0.97 (t, J 3′.4′ ) 7.5 Hz, 3H); 13C NMR (CDCl3)
δ 136.88, 134.20, 131.88, 112.73, 43.75, 30.73, 29.70-24.38,
12.25; HRMS (CI) calcd (C11H17O1, MH+) 165.1279, obsd
165.1334.
35a : 1H NMR (CDCl3) δ 6.38 (t, J 3,4 ) 6.5 Hz, 1H), 2.55
(inv quint, J ) 3 Hz, 2H), 2.42 (m, 2H), 1.93 (m, 2H), 1.77 (m,
2H), 1.71 (s, 6H), 1.57 (s, 3H); 13C NMR (CDCl3) δ 195.80,
142.18, 138.91, 130.10, 129.77, 42.90, 27.71, 29.71, 25.22,
21.81, 21.81, 20.03; MS (CI, 70 eV) m/z 179 (MH+, 26.40%),
163 (MH+ - O, 29.9%), 150 (MH+ - C2H5, 10.96%), 135 (MH+
- C3H7, 29.4%), 121 (MH+ - C4H9, 15.8%), 107 (MH+ - C5H11
,
28.3%), 93 (MH+ - C6H13, 40.8%), 71 (C5H11, 100%); HRMS
calcd (C12H18O, M+) 178.1358, obsd 178.1344.
31c: 1H NMR* (CDCl3) δ 5.55 (t, J 2,3 ) t Hz,1H), 5.13 (q,
J 3′,4′ ) 6.5 Hz, 1H), 2.32 (m, 2H), 2.18 (m, 2H), 1.75 (m, 2H),
1.73 (m, 3H), 1.59 (d, J 3′,4′ ) 6.5 Hz, 3H), 1.48 (m, 4H); 13C
NMR (CDCl3) δ 144.51, 141.17, 128.87, 118.22, 32.88, 32.27,
28.62, 27.41, 27.09, 23.19, 14.71.
42: Rf (10% acetone in hexane) 0.23; 1H NMR (CDCl3) δ
6.06 (t, J 3,4 ) 5.5 Hz, 1H), 2.53 (inv quint, J ) 4 Hz, 2H), 2.25
(m, 2H), 1.88 (m, 2H), 1.61 (m, 4H), 1.37 (s, 6H); 13C NMR
(CDCl3) δ 215.04, 143.91, 129.38, 71.29, 46.42, 29.41, 22.01,
22.54, 29.04, 29.27; HRMS calcd (C11H18O2, M+) 182.1306, obsd
182.1345.
46a : 1H NMR (CDCl3) δ 4.45 (ddd, J 1,2 ) 9.8 Hz, J 2,3 ) 7.5
and 4.2 Hz, 1H), 3.07 (J 1,2 ) 9.8 Hz, J 1,7 ) 9.0 and 2.8 and 4.2
Hz, 1H), 2.74 (sept, J 2′,3′ ) 7 Hz, 1H), 2.18 (dddd, J gem ) 15
Hz, J 3,4 ) 8.8 and 1.8 Hz, J 2,3 ) 4.2 Hz, 1H), 2.06 (dddd, J gem
) 15, J 3,4 ) 10 and 1.8 Hz, J 2,3 ) 7.5 Hz, 1H), 1.85-1.49 (m,
6H), 1.65 and 1.45 (m, 1H), 1.127 (d, J 2′,3′ ) 7 Hz, 3H), 1.129
(d, J 2′,3′′ ) 7 Hz, 3H); 13C NMR (CDCl3) δ 62.22, 59.61, 40.65,
37.44, 28.34, 28.24, 26.84, 23.41, 18.14, 17.98; MS (CI, 70 ev)
1
43: Rf (20% acetone in hexane) 0.33; H NMR* (CDCl3) δ
5.15 (bs, 1H), 5.06 (bs, 1H), 4.20 (bs, 1H), 2.88 and 2.15 (m,
2H), 2.58 and 2.01 (m, 2H), 1.90 and 1.74 (m, 2H), 1.83 and
1.43 (m, 2H), 1.78 and 1.34 (m, 2H), 1.66 and 0.96 (m, 2H);
13C NMR (CDCl3) δ 217.31, 145.91, 113.97, 82.32, 35.41, 30.15,
30.46, 25.36, 24.37, 22.72, 18.26; HRMS calcd (C11H19O2, MH+)
183.1385, obsd 183.1350.
m/z 205 (MH+ + 2, 35.62%), 203 (MH+, 100%), 167 (MH+
-
HCl, 53.75%); HRMS (CI) calcd (C11H20OCl, MH+) 203.1203,
obsd 203.1240.
44: 1H NMR (CDCl3) δ 5.88 (t, J 3,4 ) 7 Hz, 1H), 4.94 (bs,
1H), 4.69 (bs, 1H), 2.51 (inv quint., J ) 3.5 Hz, 2H), 2.29-
2.22 (m, 2H), 1.90 (t, J 3′,1′ ) J 3′,1′′ ) 0.75 Hz, 3H), 1.88-1.82
(m, 2H), 1.69-1.62 (m, 4H); HRMS (CI) calcd (C11H27O, MH+)
165.1279, obsd 165.1289.
46b: 1H NMR (CDCl3) δ 4.45 (ddd, 1H); 13C NMR (CDCl3)
δ 62.35, 61.02, 37.38, 35.96, 28.09, 27.97, 26.61, 23.39, 7.62;
MS (CI, 70 ev) m/z 191 (MH+ + 2, 9.33%), 189 (MH+, 26.94%);
153 (MH+ - HCl, 53.75%); HRMS (CI) calcd (C10H18OCl, MH+)
189.1046, obsd 189.0990.
45: Rf (10% acetone in hexane) 0.60; 1H NMR (CDCl3) δ
6.29 (d, J 3,4 ) 12 Hz, 1H), 5.61 (dt, J 3,4 ) 12 Hz, J 4,5 ) 9 Hz,
1H), 2.48 (inv quint., J ) 3 Hz, 2H), 2.09-2.01 (m, 2H), 1.94-
1.87 (m, 2H), 1.85 (bs, 6H), 1.58-1.50 (m, 2H); MS (EI, 70 eV)
m/z 164 (M+, 64.56%), 149 (M - CH3, 32.71%), 135 (M - HCO,
30.19%); 121 (M - CO - CH3, 46.30%); HRMS (CI) calcd
(C11H27O, MH+) 165.1279, obsd 165.1291.
46c: 1H NMR (CDCl3) δ 4.45 (ddd, 1H); MS (CI, 70 ev) m/z
174 (MH+, 39.03%); 139 (MH+ - Cl, 73.75%); HRMS (CI) calcd
(C9H15OCl, MH+) 174.0811, obsd 174.0720; calcd (C9H15O, MH+
- Cl) 139.1140, obsd 139.1138.
47a : 1H NMR (CDCl3) δ 7.06 (t, J 2,3 ) 7 Hz, 1H), 3.31 (sept,
J 2′,3′ ) 4 Hz, 1H), 2.50 (m, 2H), 2.35 (m, 2H), 1.79 (m, 2H),
1.55 (m, 2H), 1.46 (m, 2H), 1.08 (d, J 2′,3′ ) 4 Hz, 6H); 13C NMR
(CDCl3) δ 205.77, 145.12, 143.29, 33.64, 32.37, 29.08, 26.22,
26.05, 25.90, 19.73; MS (CI, 70 ev) m/z 167 (MH+, 100%), 97
(MH+ - CH3CO, 13.05%); HRMS (CI) calcd (C11H19O, MH+)
167.1435, obsd 167.1370.
P r epar ation of 1-(1-Isopr opylvin yl)cycloh epten e (30a),
1-(1-Eth ylvin ycycloh ep ten e (30b), a n d 1-(1-Meth ylp r o-
p en yl)cycloh ep ten e (31b). The title compounds were inde-
pendently synthesized as outlined in eq 4. Friedel-Crafts
acylation of cycloheptene (7)9,13 yielded â-chloroketones 46a ,
46b, and 46c, which underwent elimination in hot ethanolic
NaOH to the known corresponding enones 47.33 It should be
pointed out that this elimination of HCl is a very facile process
and often occurs spontaneously. Indeed, it is difficult to obtain
pure samples of â-chloroketones 46 (which is contaminated
with cycloheptene and other unidentified alkanes), since
during silica column chromatography (eluting with ethyl
acetate/hexane) variable amounts of enones 47 are generated
47b: 1H NMR (CDCl3) δ 7.09 (t, J 2,3 ) 6 Hz, 1H), 2.67 (q,
J 2′,3′ ) 7 Hz, 2H), 2.49 (m, 2H), 2.34 (m, 2H), 1.78 (m, 2H),
1.55 (m, 2H), 1.47 (m, 2H), 1.09 (t, J 2′,3′ ) 7 Hz, 3H); 13C NMR
(CDCl3) δ 201.85, 145.78, 143.74, 32.17, 30.05, 28.92, 26.04,
25.73, 25.51, 8.91; MS (CI, 70 ev) m/z 153 (MH+, 100%); HRMS
(CI) calcd (C10H17O, MH+) 153.1279, obsd 153.1300.
47c: 1H NMR (CDCl3) δ 7.09 (t, J 2,3 ) 7 Hz, 1H), 2.49 (m,
2H), 2.35 (m, 2H), 2.30 (s, 3H), 1.77 (m, 2H), 1.55 (m, 2H),

Thermolysis and Oxygenation of Substituted Cyclopropenes
J . Org. Chem., Vol. 65, No. 6, 2000 1817
1.46 (m, 2H); MS (CI, 70 ev) m/z 139 (MH+, 2.81%), 123 (MH+
- CH3, 4.23%), 95 (MH+ - CH3CO, 100%); HRMS (CI) calcd
(C9H15O, MH+) 139.1122, obsd 139.1120.
213 (MH+, 0.18%), 195 (MH+ - H2O, 10%), 179 (MH+ - H2O2,
83%), 153 (MH+ - CH3CO2H, 100%); HRMS (CI) calcd
(C12H21O3, MH+) 213.1490, obsd 213.1450. cis-52: 1H NMR
(CDCl3) δ 5.59 (m, 1H), 2.92 (bs, 1H), 2.27 (m, 1H), 2.10-1.80
(m, 8H), 1.21 (s, 3H), 0.98 (d, J 3′,4′ ) 6.6 Hz, 3H), 0.92 (d, J 3′,4′′
) 6.6 Hz, 3H); 13C NMR (CDCl3) δ 152.46, 123.82, 105.76,
90.35, 33.95, 32.52, 27.87, 26.81, 25.37, 17.76, 17.05; trans-
52: 1H NMR (CDCl3) δ 5.54 (m, 1H, 3.00 (bs, 1H), 2.27 (m,
1H), 2.10-1.80 (m, 8H), 1.38 (s, 3H), 1.05 (d, J 3′,4′ ) 6.6 Hz,
3H), 1.02 (d, J 3′,4′′ ) 6.6 Hz, 3H); 13C NMR (CDCl3) δ 151.71,
122.06, 105.76, 90.29, 34.32, 33.34, 27.35, 26.81, 24.79, 24.90,
17.60, 17.20.
P h ot ooxid a t ion of E n on es 12a a n d 12b . The title
compounds were dissolved in CH3CN or CHCl3 and irradiated,
according to the above general oxidation procedure. After an
uptake of an equivalent of O2, concentration of the solvent and
direct preparative TLC (eluting with 10% acetone in hexane
for 12a , or 10% ethyl acetate in hexane for 12b) yielded the
hydroperoxide products 51 and 52. The latter can then be
reduced to alcohols 33 and 32, respectively, with 1.1 equiv of
Ph3P. Hydroperoxide 51, but not cyclic peroxide 52, can be
reduced by 2 equiv of thiourea (in 1:1 CHCl3/CH3OH). Alter-
natively, upon conclusion of the irradiation, Ph3P can be added
prior to chromatography, in which case the corresponding
alcohols 33 and 32 are isolated. In the singlet oxygenation of
12a , peroxides 51a and 52a (or the corresponding alcohols 33a
and 32a ) are formed in a 1:3 ratio, respectively. In the case of
12b, 51b and 52b (or 33b and 32b) are formed in a 1:2 ratio,
respectively. The spectral data for 32a [Rf (10% acetone in
hexane) 0.33] and 32b [Rf (5% ethyl acetate in hexane) 0.21]
appear above. We found it difficult to purify compound 33b
by TLC since it has essentially the same Rf as enone 12b. The
NMR spectral data for 52 reveals the presence of two diaster-
eomeric forms in a ratio of 2:1 for 52a and 1.2:1 for 52b. NOE
studies reveal the major isomer in each case has the methyl
and hydroxyl groups cis to each other. The numbering of the
carbons of compounds 51, 52, 54, and 55 are as shown in
Scheme 4.
52b: Mixture of diastereomers: Rf (10% acetone in hexane)
0.29; FTIR (CHCl3) 3457 (br, s, OH), 1701.8 (m, conjugated
CO) cm-1; MS (CI, 70 ev) m/z 199 (MH+, 4.40%), 181 (MH+
-
H2O, 100%), 165 (MH+ - H2O2, 68%); HRMS (CI) calcd
(C11H19O3, MH+) 199.1334, obsd 199.1374. cis-52b: 1H NMR*
(CDCl3) δ 5.565 (dd, J 3,4 ) 8.5 Hz and 4 Hz, 1H), 2.94 (bs,
1H), 2.32 (m, 1H), 2.18 (m, 1H), 2.04 (m, 1H), 1.90 (m, 2H),
1.58 (m, 1H), 1.40 (m, 2H), 1.75 and 1.51 (ABq, J gem ) 14 Hz,
J 3′,4′ ) 7.5 Hz, 2H), 1.31 (s, 3H), 0.93 (t, J 2′,3′ ) 7.5 Hz, 3H);13C
NMR (CDCl3) δ 155.04, 123.23, 105.47, 87.65, 32.92, 32.46,
27.79, 26.81, 25.24, 21.22, 8.33. trans-52b: 1H NMR* (CDCl3)
δ 5.51 (dd, J 3,4a ) 8.5 Hz, J 3,4b ) 4 Hz, 1H), 2.88 (bs, 1H), 2.32
(m, 1H), 2.18 (m, 1H), 2.08 (m, 1H), 1.90 (m, 2H), 1.60 (m,
1H), 1.40 (m, 2H), 1.78 and 1.66 (ABq, J gem ) 14 Hz, J 3′,4′
)
7.5 Hz, 2H), 1.34 (s, 3H), 0.94 (t, J 3′,4′ ) 7.5 Hz, 3H); 13C NMR
(CDCl3) δ 153.81, 122.84, 105.62, 87.41, 33.24, 29.93, 27.63,
26.81, 25.08, 26.62, 8.22.
1
33a : Rf (10% acetone in hexane) 0.42; H NMR (CDCl3) δ
P h otooxid a tion of En on e 20. The title compounds were
dissolved in CH3CN and irradiated, according to the above
general oxidation procedure. After an uptake of an equivalent
of O2, concentration of the solvent and direct preparative TLC
(eluting with 20% acetone in hexane) yielded products 55, 54,
and 43 in a ratio of 2:1:1. We were not able to obtain pure
samples of hydroperoxide 54, presumably because it decom-
poses on silica to alcohol 43. It was tentatively identified in
the photooxidation mixture by its vinyl absorptions at 5.18 (bs,
1H, H1′′) and 4.85 (bs, 1H, H1′). These absorptions disappeared
completely upon the addition of Ph3P or thiourea, replaced by
the corresponding vinyl absorptions of 43 at 5.15 (bs, 1H, H1′′),
5.06 (bs, 1H, H1′).
5.09 (bs, 1H), 5.07 (bs, 1H), 4.12 (bs, 1H), 2.81 (dt, J gem ) 12
Hz, J 6,7 ) 3 Hz, 1H), 2.44 (m, 1H), 2.43 (sept, J 3′,4′ ) 7.5 Hz,
1H), 2.25-1.66 and 1.56-1.36 (8H, m), 1.08 (d, J 3′,4′ ) 7.5 Hz,
3H), 0.97 (d, J 3′,4′′ ) 7.5 Hz, 3H); 13C NMR (CDCl3) δ 214.73,
157.65, 110.08, 80.00, 39.14, 34.85, 30.50, 28.30, 27.87, 24.60,
24.54, 23.28; MS (CI, 70 eV) m/z 195 (MH+, 12.7%), 179 (MH+
- H2O, 100%); FTIR (CDCl3) 3450 (br, s, OH), 1699 (s, CdO)
cm-1; HRMS calcd (C12H20O2, M+) 196.1463, obsd 196.1425.
33b: Rf (20% acetone in hexane) 0.33; 1H NMR* (CDCl3) δ
5.02 (bs, 1H), 4.98 (bs, 1H), 4.61 (bs, 1H), 2.79 (ddd, J gem ) 13
Hz, J 6,7 ) 11 and 3 Hz, 1H), 2.43 (m, 1H), 2.13 (m, 1H), 2.01
(m, 1H), 1.98 (m, 1H), 1.97 (m, 1H), 1.82 (m, 1H), 1.48-1.38
(m, 4H), 1.45 (m, 1H), 1.04 (t, J 3′,4′ ) 7.5 Hz, 3H); MS (CI, 70
eV) m/z 165 (MH+-H2O, 70%); HRMS calcd (C11H18O2, M+)
182.1306, obsd 182.1372.
1
55: Rf (20% acetone in hexane) 0.30; H NMR* (CDCl3) δ
5.47 (dd, J 3,4 ) 10 and 7.5 Hz, 1H), 2.92 (bs, 1H), 2.71 and
2.03 (m, 2H), 2.16 and 1.76 (m, 2H), 1.78 and 1.37 (m, 2H),
1.68 and 1.57 (m, 2H), 1.65 and 1.48 (m, 2H), 1.39 (s, 3H),
1.38 (s, 3H); 13C NMR (CDCl3) δ 152.70, 122.46, 106.46, 85.46,
38.49, 27.05, 26.44, 26.28, 24.31, 23.13, 20.74; MS (CI, 70 ev)
m/z 181 (MH+ - OH, 85.07%), 165 (MH+ - H2O2, 100%);
HRMS calcd (C11H17O2, MH+ - OH) 181.1229, obsd 181.1246.
1
51a : Rf (10% acetone in hexane) 0.43; H NMR (CDCl3) δ
5.25 (bs, 1H), 4.85 (bs, 1H), 2.90 (m, 1H), 2.72 (td, J gem ) 12
Hz, J 6,7 ) 3 Hz, 1H), 2.48 (m, 3H), 2.25 (m, 1H), 2.00-1.75
(m, 5H), 1.15 (d, J 3′,4′ ) 6 Hz, 3H), 1.10 (d, J 3′,4′′ ) 6 Hz, 3H);
13C NMR (CDCl3) δ 217.48, 154.30, 113.92, 94.62, 42.35, 33.16,
30.11, 27.89, 27.43, 24.73, 24.61, 24.06; FTIR (CDCl3) 3410
(br s, OH), 1712 (s, CO) cm-1; MS (CI, 70 ev) m/z 213 (MH+,
10%), 195 (MH+ - H2O, 100%), 179 (MH+ - H2O2, 84%);
HRMS (CI) calcd (C12H21O3, MH+) 213.1490, obsd 213.1510.
Ack n ow led gm en t. S.D.B. acknowledges the kind
support of the 1986 Michael Landau Research Award.
1
51b: Rf (10% acetone in hexane) 0.37; H NMR (CDCl3) δ
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra of
E-12a , Z-12a , E-12b, Z-12b, 15, 18, 19, 20, E-29a and Z-29a ,
E-29b and Z-29b, 30a , 30b, 31b, 32a , 32b, 33a , 33b, 34a ,
34b, 35a , 42, 43, 44, 45, 46a , 46b, 47a , 47b, 47c, 51a , 51b,
52a , 52b, and 55, and the complete 1H and 13C NMR peak
assignments for the compounds described in the Experimental
Section. This material is available free of charge via the
Internet at http://pubs.acs.org.
5.15 (bs, 1H), 4.80 (bs, 1H), 2.72 (td, J gem ) 12 Hz, J 6,7 ) 3 Hz,
1H), 2.48 (ddd, J gem ) 12 Hz, J 6,7 ) 6 and 3 Hz, 1H), 2.20 (m,
2H), 2.17-1.38 (m, 8H), 1.12 (t, J 3′,4′ ) 7.5 Hz, 3H); 13C NMR
(CDCl3) δ 212.29, 148.49, 113.75, 94.13, 42.43, 33.16, 30.28,
27.43, 24.21, 22.84, 11.78; MS (CI, 70 ev) m/z 199 (MH+,
7.79%), 181 (MH+ - H2O, 61.51%), 165 (MH+ - H2O2, 100%);
HRMS (CI) calcd (C11H19O3, MH+) 199.1334, obsd 199.1370.
52a : Mixture of diastereomers: Rf (10% acetone in hexane)
0.29; FTIR (CHCl3) 3419 (br, s, OH) cm-1; MS (CI, 70 ev) m/z
J O991803B