Ozonolyses of 1-alkyl-substituted 1-tert-butylethylenes and highly methylated methylenecycloalkanes. The influence of the substituent steric bulk on the direction of cleavage of the primary ozonides

Article abstract of DOI:10.1016/S0040-4020(01)01183-8

Ozonolyses of 1-alkyl-substituted 1-tert-butylethylenes and of highly methylated methylenecycloalkanes were conducted in the presence of trifluoroacetophenone (7) in ether. The ozonolysis of 2,2,6-trimethyl-1-methylenecyclohexane provided only the crossed-ozonide 10 derived from capture of formaldehyde O-oxide with the ketone 7 in 42% yield, while in the case of the relevant 2,2,5-trimethyl-1-methylenecyclopentane the alternative crossed-ozonide 15e derived from cycloaddition of 2,2,5-trimethylcyclopentanone O-oxide with the ketone 7 was the sole isolable product. The total energies of two possible cycloreversion processes for the primary ozonide 12c and for 12e, calculated at B3LYP/6-31G**//B3LYP/3-21G* level of theory, seem to reproduce the observed difference in the regiochemistry of fragmentation between these two primary ozonides.

Full text of DOI:10.1016/S0040-4020(01)01183-8

TETRAHEDRON
Pergamon
Tetrahedron 58 ,2002) 891±896
Ozonolyses of 1-alkyl-substituted 1-tert-butylethylenes
and highly methylated methylenecycloalkanes. The in¯uence
of the substituent steric bulk on the direction of cleavage
of the primary ozonides
Shin-ichi Kawamura, Hideyuki Yamakoshi, Manabu Abe, Araki Masuyama
and Masatomo Nojimap
Department of Materials Chemistry and Frontier Research Center, Faculty of Engineering, Osaka University, Suita, Osaka 565-0871, Japan
Received 17 October 2001; accepted 26 November 2001
AbstractÐOzonolyses of 1-alkyl-substituted 1-tert-butylethylenes and of highly methylated methylenecycloalkanes were conducted in the
presence of tri¯uoroacetophenone ,7) in ether. The ozonolysis of 2,2,6-trimethyl-1-methylenecyclohexane provided only the crossed-
ozonide 10 derived from capture of formaldehyde O-oxide with the ketone 7 in 42% yield, while in the case of the relevant 2,2,5-
trimethyl-1-methylenecyclopentane the alternative crossed-ozonide 15e derived from cycloaddition of 2,2,5-trimethylcyclopentanone
O-oxide with the ketone 7 was the sole isolable product. The total energies of two possible cycloreversion processes for the primary ozonide
12c and for 12e, calculated at B3LYP/6-31Gpp//B3LYP/3-21Gp level of theory, seem to reproduce the observed difference in the regio-
chemistry of fragmentation between these two primary ozonides. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
ozonolyses of a series of 1-alkyl-substituted 1-tert-butyl-
ethylenes 1a±c and of methylenecycloalkanes, 11a±d and
17, in the presence of tri¯uoroacetophenone in ether and
have determined in each case the ratios of two possible
crossed-ozonides.
Extensive investigation of the mechanism of alkene ozonoly-
sis has con®rmed the essential features of the pathway
originally proposed by Criegee.1 Much of the current
interest in this process centers on the factors affecting the
direction of cleavage of the primary ozonide ,PO; 1,2,3-
trioxolane).2 Three factors have been found to play an
important role, i.e. the electron-donating effect of the
alkyl substituent attached to the C±C double bond
,Markovnikov effect),3 the electron-withdrawing effect of
the hetero-atom substituent at the allylic position,4 and the
steric effect of the allylic dialkyl substituents.5 The most
clear-cut example indicating the importance of the latest
factor is the ozonolysis of 1,5,5-trimethylcyclopentene.5c
Of two possible carbonyl oxide intermediates, only the
less-hindered aldehyde oxide, with the geminal methyl
groups remote from the carbonyl oxide group, is exclusively
produced. Thus, in this case the steric effect of the allylic
dimethyl groups seems to overcome the electronic effect
of the methyl substituent at C-1. In contrast, ozonolysis of
1-methylcyclopentene proceeds mainly by the correspond-
ing ketone oxide.3b To obtain further insight into the origin
of the substituent steric effects, we have conducted the
2. Results and discussion
2.1. Ozonolysis of 1-alkyl-substituted 1-tert-butyl-
ethylenes
Ozonolysis of 1-tert-butyl-1-methylethylene ,1a) in ether
gave the normal ozonide 9a in a yield of 36%. When the
same reaction was repeated in the presence of tri¯uoro-
acetophenone ,7), however, the formation of the ozonide
9a was completely suppressed and instead, only the
crossed-ozonide 8a derived from the cycloaddition of tert-
butyl methyl ketone O-oxide ,3a) with the ketone 7 was
obtained in 60% yield ,Table 1 and Scheme 1). This
suggests that ,i) the direction of cleavage of PO 2a is very
selective, affording exclusively the ketone oxide 3a and
,ii) the reaction of the ketone O-oxide 3a with the reactive
ketone 7 is much faster than that with formaldehyde 4. From
the ozonolysis of 1-tert-butyl-1-isopropylethylene ,1b) in
the presence of 7 was obtained the crossed-ozonide
8b ,56%). In the case of 1,1-di-tert-butylethylene ,1c),
however, the 1H NMR spectra of the crude reaction mixture
showed that only the alternative cross-ozonide 10, derived
Keywords: ozonolysis; direction of cleavage of primary ozonide; steric
effect.
Corresponding author. Tel./fax: 181-6-6879-7928;
e-mail: nojima@ap.chem.eng.osaka-u.ac.jp
p
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020,01)01183-8

892
S. Kawamura et al. / Tetrahedron 58 ,2002) 891±896
Table 1. Ozonolysis of 1-alkyl-substituted tert-butylethylenes and highly
methylated methylenecycloalkanes
the crossed-ozonides 8a,b from alkenes 1a,b suggests that
for these substrates the electronic effects of the alkyl-sub-
stituents are decisive, thereby providing predominantly the
pairs of the ketone O-oxides 3a,b and formaldehyde ,4)
,path a in Scheme 1). In the case of highly congested
di-tert-butylethylene ,1c), however, the steric effect of two
tert-butyl groups determines the regiochemistry of the PO
cleavage such that the sterically less-congested formalde-
hyde O-oxide ,6) is exclusively produced ,path b in Scheme
1). In the case of 1a,c the difference in total energies
between two cleavage pathways, determined by the
quantum chemical calculations at B3LYP/6-31Gpp//
B3LYP/3-21Gp level of theory,6 is consistent with these
conclusions ,Table 2). In the alkene 1a the pathway a
leading to the formation of the corresponding ketone
O-oxides 3a is more exothermic ,15.1 kJ/mol). In contrast,
the alternative pathway b is more favorable ,7.4 kJ/mol) in
the case of the sterically most-congested di-tert-butylethyl-
ene ,1c). However, the DFT calculations could not repro-
duce the result for the behavior of 1b; the calculation
suggests the predominance of path b ,5.4 kJ/mol), while
the experimental result demonstrates the predominant
contribution of the alternative path a. At present, we can
not offer proper explanation for this discrepancy.
Substrate
Additivea
Products ,% yield)
1a
1a
1b
1b
1c
1c
11a
11a
11b
11b
11c
11c
11d
11d
11e
11e
9a ,36)
8a ,60)
PhCOCF3
PhCOCF3
PhCOCF3
PhCOCF3
PhCOCF3
PhCOCF3
PhCOCF3
PhCOCF3
9b ,48)
8b ,56)
5c ,45)
10 ,44)b
16a ,81)
15a ,59)
16b ,75)
15b ,65)
14c ,56)
10 ,42)b, 14c ,75)
14d ,76)
10 ,42)b, 14d ,80)
16e ,53)c
15e ,59)d
Reaction with 1 equiv. of ozone in diethyl ether at 2708C.
The reaction was conducted in the presence of 1 equiv. of tri¯uoroaceto-
a
phenone.
The yields of the ozonide 10 determined by the H NMR spectra of the
b
1
crude product mixture were 67% from 1c, 83% from 11c, and 83% from
11d, respectively.
c
The 1H NMR spectra of the crude product mixture showed that 16e was
produced in 80% yield.
1H NMR spectra of the crude product mixture showed that 15e was
produced in 80% yield.
d
2.2. Ozonolyses of highly-methylated methylene-
cycloalkanes
from cycloaddition of formaldehyde O-oxide ,6) to the
ketone 7, was produced in 67% yield ,Scheme 1). Since
the ozonide 10 was volatile, the isolated yield was only
44% ,Table 1, see footnote).
As expected from the substituent electronic effects, ozonoly-
sis of methylenecyclohexane ,11a) in the presence of
tri¯uoroacetophenone ,7) in ether proceeded mainly by
cyclohexanone O-oxide ,13a) and as the sole isolable
product, the crossed-ozonide 15a was obtained in 59%
yield ,Table 1 and Scheme 2). From 2,2-dialkyl-substituted
1-methylenecyclohexane 11b also, the corresponding
crossed-ozonide 15b derived from cycloaddition of the
ketone O-oxides 13b and the ketone 7 was exclusively
produced. The presence of additional methyl group,s) at
C-6, however, altered dramatically the regiochemistry of
POcleavage. Ozonolyses of 2,2,6-trimethyl-1-methylene-
cyclohexane ,11c) and 2,2,6,6-tetramethyl-1-methylene-
cyclohexane ,11d) gave exclusively the alternative
crossed-ozonide 10 derived from capture of formaldehyde
O-oxide ,6) by the ketone 7. Thus, in the ozonolyses of
substrates 11c,d the steric effects of the additional methyl
substituent,s) seem to be important in determining the
direction of cleavage of PO.
These results clearly demonstrate that for a series of 1-alkyl-
substituted 1-tert-butylethylenes 1a±c, the electronic and/or
steric effects of the alkyl substituents play an important role
in determining the direction of cleavage of the correspond-
ing PO 2a±c. It would be reasonable to expect that, on the
basis of the electronic considerations, the formation of the
corresponding ketone O-oxide 3 should be favored, whereas
a preponderance of formaldehyde O-oxide ,6) would arise
from the in¯uence of steric effects. Exclusive formation of
If consideration is given to the behavior of 1,1-di-tert-butyl-
ethylene ,1c), the result for the highly crowded methylene-
cyclohexane 11d is understandable.5 Less easily understood
is the fact that the substitution pattern of 2,2,6-trimethyl-1-
methylenecyclohexane ,11c) is very similar to that of 1-tert-
butyl-1-isopropylethylene ,1b) and nevertheless, the regio-
chemistry of the POcleavage is quite different between two
PO, 12c and 2b. As an approach to understanding this appar-
ent discrepancy, ozonolysis of 2,2,5-trimethyl-1-methylene-
cyclopentane ,11e) was conducted in the presence of
tri¯uoroacetophenone ,7). Surprisingly, the sole isolable
product was the crossed-ozonide 15e derived from the
cycloaddition of the ketone O-oxide 13e and the ketone 7,
suggesting that in this case the substituent electronic effects
overcome the steric effect of the methyl substituent at C-5
Scheme 1. ,a) RˆMe; ,b) Rˆisopropl; ,c) Rˆtert-butyl.

S. Kawamura et al. / Tetrahedron 58 ,2002) 891±896
893
Table 2. The difference in heats of two cycloreversion processes for primary ozonides, 2a±c, respectively
R
2
3
314 ,path a)
5
516 ,path b)
Me
iPr
tBu
2500.674911 hartree
,44.0 kJ/mol)
2579.298819 hartree
,50.1 kJ/mol)
2618.602473 hartree
,57.0 kJ/mol)
2386.189269 hartree
2464.813413 hartree
2504.119943 hartree
2500.691689 hartree
,0 kJ/mol)
2579.316641 hartree
,5.4 kJ/mol)
2618.621363 hartree
,7.4 kJ/mol)
2311.111119 hartree
2389.743092 hartree
2429.049360 hartree
2500.685933 hartree
2579.317906 hartree
,0 kJ/mol)
2618.624174 hartree
,0 kJ/mol)
Heats of formation of 4 and 6 were 2144.502420 and 2189.574814 hartree, respectively.
,Table 1 and Scheme 3). In connection with this, ozonolysis
of the methylenecyclopentane 11e in ether in the absence of
the ketone 7 was found to give the normal ozonide 16e in a
high yield of 80%, suggesting that the cycloaddition of the
ketone O-oxide 1e with formaldehyde ,4) proceeds
smoothly. In contrast, under the similar conditions the
ozonolysis of the methylenecyclohexane 11c did not yield
the normal ozonide 16c and instead, the ketone 14c was
isolated in 56% yield. This implies that the sterically
congested ketone 14c has a negligible reactivity even
toward the small and reactive formaldehyde O-oxide ,6).
Exactly the same trend was observed for the sterically
most-congested methylenecyclohexane 11d.
To understand the remarkable difference in behavior
between the two substrates, 11c and 11e, the difference in
total energies between two cycloreversion processes were
Scheme 2.
Scheme 3.

894
S. Kawamura et al. / Tetrahedron 58 ,2002) 891±896
Table 3. The difference in heats of two cycloreversion processes for primary ozonides, 12c and 12e, respectively
examined in each case by the quantum chemical calcu-
lations at B3LYP/6-31Gpp//B3LYP/3-21Gp level of theory.
,Table 3). In the case of PO 12c path b leading to the forma-
tion of formaldehyde O-oxide ,6) and ketone 14c was more
favorable than the alternative pathway a leading to the
ketone O-oxide 13c and formaldehyde ,4). In contrast,
exactly a reverse trend was observed for PO 12e. Thus,
path a leading to the ketone O-oxide 13e and 4 was more
favorable than the alternative pathway b leading to a pair of
the ketone 14e and 6. These results are clearly consistent
with the experimental observations. The dramatic ring-size
effect ,cyclohexanone O-oxide 13c and cyclopentane
O-oxide 13e) on the energy difference can be reasonably
explained by the steric effect ,Fig. 1). In the case of 13c a
severe repulsion ,1,3-allylic strain) between the oxygen and
the methyl group, which are located in almost the same
plane, would be expected. Alternatively, in 13e the steric
repulsion seems to be negligible, as can be visualized in the
calculated structure ,Fig. 1).
3. Conclusion
The ozonolyses of 1-alkyl-substituted 1-tert-butylethylenes
and of highly methylated methylenecylcloalkanes, together
with the former observations,5 demonstrate that the steric
effect of the allylic alkyl groups plays an important role in
determining the direction of cleavage of POsuch that the
formation of the electronically less-favored but sterically
more-favored carbonyl oxide predominates. However, the
remarkable difference in the regiochemistry of the PO
cleavage observed between 2,2,6-trimethyl-1-methylene-
cyclohexane and 2,2,5-trimethyl-1-methylenecyclopentane
suggests that the extent of appearance of the steric effect
would be signi®cantly in¯uenced by the apparently minor
change in the structure of substrates.
4. Experimental
4.1. General procedures
1H and13C NMR spectra were obtained in CDCl3 ,unless
otherwise noted) with SiMe4 standard. The method of
ozonolysis was previously described.7 The alkenes 1b,8
1c,9 and 11b,10 11c,9 11d,11 were prepared from the corre-
sponding ketones 5b,12 5c,12 and 14b,13 14c,14 14d15 by
the reported methods. 2,2,5-Trimethyl-1-methylenecyclo-
pentane ,11e) was prepared as follows.10 To dry dimethyl
sulfoxide ,30 mL) was added 1.58 g of sodium hydride
followed by 13.6 g ,38 mmol) of methyltripheylphos-
phonium bromide. This mixture was stirred at 50±558C
for 1 h before 4.0 g ,32 mmol) of 2,2,5-trimethylcyclo-
pentanone16 was added as a solution in 20 mL dimethyl
sulfoxide. After the mixture was stirred at 558C overnight,
it was added to 50 mL of water and extracted four times
with pentane. The pentane extracts were combined, washed
with saturated brine, and then dried over MgSO4. The
Figure 1. B3LYP-3-21Gp optimized structures of the carbonyl oxides 13c
and 13e.

S. Kawamura et al. / Tetrahedron 58 ,2002) 891±896
895
pentane was removed by distillation and the residue was
distilled under reduced pressure to give 1.2 g ,31%) of
2,2,5-trimethyl-1-methylenecyclopentane ,11e): colorless
7.7 ,5H, m); 13C NMR d 95.56, 103.03 ,q, Jˆ34 Hz),
121.49 ,q, Jˆ289 Hz), 126.77, 128.73, 129.08, 130.58,
135.51. Anal. Calcd for C9H7F3O3: C, 49.1; H, 3.2. Found:
C, 49.0; H, 3.0.
1
oil, bp 808C ,20 mmHg); H NMR d 1.03 ,3H, s), 1.08
,3H, s), 1.09 ,3H, d, Jˆ7 Hz), 1.2±1.6 ,4H, m), 1.8±1.9
,1H, m), 4.74 ,1H, d, Jˆ2 Hz), 4.78 ,1H, d, Jˆ2 Hz); 13C
NMR d 19.75, 29.88, 29.96, 31.99, 39.26, 40.11, 42.39,
101.89, 167.40. Anal. Calcd for C9H16: C, 87.0; H, 13.0.
Found: C, 87.0; H, 13.1.
4.4. Ozonolysis of methylenecycloalkanes in ether
Ozonolysis of methylenecyclohexane ,11a) is representa-
tive. To a diethyl ether solution ,20 mL) of an alkene 11a
,192 mg, 2.00 mmol) was passed a slow stream of ozone
,1 equiv.) at 2708C. After evaporation of the solvent, the
products were separated by column chromatography on
silica gel. Elution with ether±hexane ,1:20; v/v) gave the
ozonide 16a ,234 mg, 81%).
4.2. Ozonolysis of 1-alkyl-substituted 1-tert-butylethyl-
enes 1a±c in ether
The ozonolysis of 1-tert-butyl-1-methylethylene ,1a) is
representative. To a solution of an ethylene 1a ,196 mg,
2.00 mmol) in diethyl ether ,20 mL) was passed a slow
stream of ozone ,1 equiv.) at 2708C. After evaporation of
the solvent, the crude products were separated by column
chromatography on silica gel ,column, 2£50 cm; 20 g of
silica gel). Elution with ether±hexane ,1:20, v/v) gave
ozonide 9a ,106 mg, 36%).
4.4.1. 1,2,4-Trioxaspiro[4.5]decane 216a). Colorless oil;
1H NMR d 1.2±1.8 ,10H, m), 5.27 ,2H, s); 13C NMR d
23.94, 24.91, 33.94, 93.75, 108.89. Anal. Calcd for
C7H12O3: C, 58.3; H, 8.4. Found: C, 58.2; H, 8.4.
4.4.2. 1,2,4-Trioxadispiro[4.0.4.4]tetradecane 216b).
1
Colorless oil; H NMR d 1.2±2.1 ,16H, m), 5.01 ,1H, s),
4.2.1. 3-tert-Butyl-3-methyl-1,2,4-trioxolane 29a). Color-
less oil; 1H NMR d 1.00 ,9H, s), 1.35 ,3H, s), 4.88 ,1H, s),
5.15 ,1H, s); 13C NMR d 19.57, 25.41, 37.70, 94.28, 113.35.
Anal. Calcd for C7H14O3; C, 57.5; H 9.65. Found: C, 57.7;
H, 9.6.
5.22 ,1H, s); 13C NMR d 21.57, 23.18, 25.25, 25.41, 32.69,
33.64, 33.96, 35.40, 36.89, 94.41, 112.53. Anal. Calcd for
C11H18O3: C, 66.6; H, 9.2. Found: C, 66.8; H, 9.1.
4.4.3. 6,6,9-Trimethyl-1,2,4-trioxaspiro[4.4]nonane 216e).
Colorless oil; 1H NMR d 0.98 ,3H, s), 0.99 ,3H, d, Jˆ7 Hz),
1.07 ,3H, s), 1.4±1.6 ,4H, m), 1.8±1.9 ,1H, m), 5.04 ,1H, s),
5.11 ,1H, s); 13C NMR d 22.70, 25.88, 27.51, 31.63, 36.84,
38.90, 43.16, 94.00, 118.06. Anal. Calcd for C9H16O3: C,
62.8; H, 9.4. Found: C, 62.6; H, 9.6.
4.2.2. 3-tert-Butyl-3-isopropyl-1,2,4-trioxolane 29b).
1
Colorless oil; H NMR d 0.98 ,6H, d, Jˆ3 Hz), 1.10 ,9H,
s), 2.0±2.3 ,1H, m), 4.93 ,1H, s), 5.10 ,1H, s); 13C NMR d
18.56, 20.08, 25.86, 32.24, 39.66, 94.23, 114.03. Anal. Calcd
for C9H18O3: C, 62.0; H, 10.4. Found: C, 62.0; H, 10.6.
4.5. Ozonolysis of methylenecycloalkanes in the presence
of tri¯uoroacetophenone
4.3. Ozonolysis of 1-alkyl-substituted 1-tert-butylethyl-
enes 1a±c in the presence of tri¯uoroacetophenone in
ether
Ozonolysis of methylenecyclohexane ,11a) is representa-
tive. After treating a diethyl ether solution ,20 mL) of an
alkene 11a ,160 mg, 1.67 mmol) and tri¯uoroacetophenone
,291 mg, 1.67 mmol) with ozone ,1 equiv.) at 2708C, the
crude products were separated by column chromatography
on silica gel. The ®rst fraction ,elution with ether±hexane,
1:20) contained the ozonide 15a ,285 mg, 59%).
Ozonolysis of 1-tert-butyl-1-methylethylene ,1a) is repre-
sentative. To a solution of an alkene 1a ,196 mg, 2.00
mmol) and tri¯uoroacetophenone ,348 mg, 2 mmol) in
diethyl ether ,20 mL) was passed a slow stream of ozone
,1 equiv.) at 2708C. After evaporation of the solvent, the
products were separated by column chromatography on
silica gel ,elution with ether±hexane, 1:20). As the ®rst
fraction, the ozonide 8a ,348 mg, 60%) was eluted.
4.5.1. 3-Phenyl-3-tri¯uoromethyl-1,2,4-trioxaspiro[4.5]-
1
decane 215a). Colorless oil; H NMR d 1.3±2.2 ,10H, m),
7.3±7.8 ,5H, m); 13C NMR d 23.38, 23.92, 24.69, 31.77,
33.93, 103.27 ,q, Jˆ34 Hz), 113.08, 121.69 ,q, Jˆ288 Hz),
126.61, 128.27, 130.21, 132.09, 135.49. Anal. Calcd for
C14H15F3O3: C, 58.3; H, 5.3. Found: C, 58.6; H, 5.4.
4.3.1. 3-tert-Butyl-3-methyl-5-phenyl-5-tri¯uoromethyl-
1
1,2,4-trioxolane 28a). Colorless oil; H NMR d 1.13 ,9H,
s), 1.25 ,3H, s), 7.3±7.7 ,5H, m). Anal. Calcd for
C14H17F3O3: C, 57.9; H, 5.9. Found: C, 58.2; H, 6.2.
4.5.2. 3-Phenyl-3-tri¯uoromethyl-1,2,4-trioxadispiro-
[4.0.4.4]tetradecane 2a 2:1 mixture of two isomers)
215b). Colorless oil; H NMR d 1.0±2.3 ,16H, m), 7.3±
4.3.2. 3-tert-Butyl-3-isopropyl-5-phenyl-5-tri¯uoromethyl-
1,2,4-trioxolane 28b). Colorless oil; 1H NMR d 0.63 ,3H, d,
Jˆ7 Hz), 0.90 ,3H, d, Jˆ7 Hz), 1.15 ,9H, s), 2.27 ,1H,
septet, Jˆ7 Hz), 7.3±7.8 ,5H, m); 13C NMR d 18.24,
19.00, 20.36, 26.18, 28.16, 32.60, 39.26, 103.16 ,q, Jˆ
33 Hz), 121.74 ,q, Jˆ288 Hz), 126.82, 128.01, 129.84,
132.44. Anal. Calcd for C16H21F3O3: C, 60.4; H, 6.7.
Found: C, 60.1; H, 6.7.
1
7.8 ,5H, m); 13C NMR d major isomer: 21.69, 22.79,
25.05, 25.38, 31.29, 33.10, 33.98, 36.88, 48.09, 103.02 ,q,
Jˆ34 Hz), 116.73, 121.65 ,q, Jˆ288 Hz), 126.51, 128.05,
128.14, 130.01, 133.14; minor isomer: 13C NMR d 21.46,
23.13, 25.14, 25.25, 31.81, 33.25, 34.54, 37.02, 48.86,
103.59 ,q, Jˆ34 Hz), 117.93, 121.36 ,q, Jˆ287 Hz),
126.51, 127.04, 128.14, 129.92, 133.53. Anal. Calcd for
C18H21F3O3: C, 63.2; H, 6.2. Found: C, 63.3; H, 6.5.
4.3.3. 3-Phenyl-3-tri¯uoromethyl-1,2,4-trioxolane 210).
1
Colorless oil; H NMR d 5.13 ,1H, s), 5.45 ,1H, s), 7.3±

896
S. Kawamura et al. / Tetrahedron 58 ,2002) 891±896
J.; Will, E.; Yamakoshi, Y.; Teshima, K.; Nojima, M. J. Org.
Chem. 1994, 59, 5055. ,d) Teshima, K.; Kawamura, S.;
Ushigoe, Y.; Nojima, M.; McCullough, K. J. J. Org. Chem.
1995, 60, 4755. ,e) Yamakoshi, H.; Kawamura, S.; Nojima,
M.; Mayr, H.; Baran, J. J. Org. Chem. 1996, 61, 5939.
,f) Kawamura, S.; Takeuchi, R.; Masuyama, A.; Nojima,
M.; McCullough, K. J. J. Org. Chem. 1998, 63, 5617.
6. ,a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. Sect. B 1998, 37,
785. ,b) Becke, A. D. Phys. Rev. Sect. A 1988, 38, 3098.
,c) Beche, A. D. J. Chem. Phys. 1993, 98, 5648. ,d) The
calculations were performed within TITAN; Wavefunction,
Inc., Schrodinger, Inc., 1999.
4.5.3. 3-Phenyl-3-tri¯uoromethyl-6,6,9-trimethyl-1,2,4-
trioxaspiro[4.4]nonane 215e). Colorless oil; H NMR d
1
1.06 ,3H, s), 1.10 ,3H, d, Jˆ7 Hz), 1.23 ,3H, s), 1.5±1.7
,4H, m), 1.8±1.9 ,1H, m), 7.4±7.6 ,5H, m); 13C NMR d
22.41, 25.50, 28.75, 31.70, 38.38, 41.63, 43.43, 102.88 ,q,
Jˆ33 Hz), 121.18 ,q, Jˆ289 Hz), 122.05, 126.79, 128.17,
128.25, 130.15, 132.29. Anal. Calcd for C16H19F3O3: C,
60.8; H, 6.1. Found: C, 60.5; H, 6.0.
References
7. McCullough, K. J.; Nakamura, N.; Fujisaka, T.; Nojima, M.;
Kusabayashi, S. J. Am. Chem. Soc. 1991, 113, 1786.
8. Kimpe, N.; Buyck, L.; Verhe, R.; Schamp, N. Tetrahedron
1984, 40, 3291.
1. Criegee, R. Angew. Chem., Int. Ed. Engl. 1975, 11, 745.
2. ,a) Bailey, P. S. Ozonation in Organic Chemistry, Vol. 2;
Academic: New York, 1978; p 1982. ,b) Bunnelle, W. H.
Chem. Rev. 1991, 91, 335. ,c) McCullough, K. J.; Nojima,
M. In Organic Peroxides; Ando, W., Ed.; Wiley: New York,
1992.
9. Lutz, F.; Ulrike, Q. Synth. Commun. 1985, 15, 855.
10. Christiansen, G. D.; Lightner, D. A. J. Org. Chem. 1971, 36,
948.
3. ,a) Fliszar, S.; Belzecki, C.; Chylinska, J. B. Can. J. Chem.
1967, 45, 221. ,b) Griesbaum, K.; Kiesel, G. Chem. Ber. 1989,
122, 145. ,c) McCullough, K. J.; Takeuchi, R.; Masuyama, A.;
Nojima, M. J. Org. Chem. 1999, 64, 2137.
11. Irngartinger, H.; Kurda, E.; Rodewald, H.; Berndt, A. Chem.
Ber. 1982, 115, 967.
12. Whitmore, F. C.; Stahly, E. E. J. Am. Chem. Soc. 1933, 55,
4153.
13. Krapcho, A. P. Synthesis 1974, 383.
4. ,a) Bunnelle, W. H.; Isbell, T. A. J. Org. Chem. 1992, 57, 729.
,b) Hayes, R.; Wallace, T. W. Tetrahedron Lett. 1990, 31,
3355. ,c) Kawamura, S.; Yamakoshi, H.; Nojima, M. J. Org.
Chem. 1996, 61, 5953.
14. Stevens, C. L.; Weinheimer, A. L. J. Am. Chem. Soc. 1958, 80,
4072.
15. Millard, A. A.; Rathke, M. W. J. Org. Chem. 1978, 43, 1834.
16. Pierre, F.; Emile, D. J. Tetrahedron 1978, 34, 1349.
5. ,a) Criegee, R. Chem. Ber. 1975, 108, 743. ,b) Griesbaum, K.;
Volpp, W. Chem. Ber. 1988, 121, 1795. ,c) Mayr, H.; Baran,