A New Methodology for the Bis(oxocyclohexadienyl) Peroxide Formation

Article abstract of DOI:10.1021/jo971307s

Symmetrically substituted bis(4-oxocyclohexa-2,5-dienyl) peroxide 5 (R = R') as well as unsymmetrically substituted 5 (R ≠ R') can be prepared efficiently by treating 4-halogenocyclohexa-2,5-dienone 3 with 4-hydroperoxycyclohexa-2,5-dienone 4 in the presence of an appropriate positive halogen compound such as N-iodosuccinimide. Acetonitrile is a suitable solvent for the reaction. The formation of 5 is suggested to take place via electrophilic attack by the positive halogen species upon 3 generating the 4-oxocyclohexa-2,5-dienyl cation (or the phenoxy cation), followed by nucleophilic attack by 4 upon the cation. It is emphasized that some of the peroxides obtained by this means have not been prepared by the classical method, coupling of phenoxy radicals with O₂.

Full text of DOI:10.1021/jo971307s

8790
J . Org. Chem. 1997, 62, 8790-8793
A New Meth od ology for th e Bis(oxocycloh exa d ien yl) P er oxid e
F or m a tion
Kanji Omura*
College of Nutrition, Koshien University, Momijigaoka, Takarazuka, Hyogo 665, J apan
Received J uly 16, 1997X
Symmetrically substituted bis(4-oxocyclohexa-2,5-dienyl) peroxide 5 (R ) R′) as well as unsym-
metrically substituted 5 (R * R′) can be prepared efficiently by treating 4-halogenocyclohexa-2,5-
dienone 3 with 4-hydroperoxycyclohexa-2,5-dienone 4 in the presence of an appropriate positive
halogen compound such as N-iodosuccinimide. Acetonitrile is a suitable solvent for the reaction.
The formation of 5 is suggested to take place via electrophilic attack by the positive halogen species
upon 3 generating the 4-oxocyclohexa-2,5-dienyl cation (or the phenoxy cation), followed by
nucleophilic attack by 4 upon the cation. It is emphasized that some of the peroxides obtained by
this means have not been prepared by the classical method, coupling of phenoxy radicals with O2.
Sch em e 1
Bis(oxocyclohexadienyl) peroxides such as 2 can be
prepared by coupling of stable or relatively stable phe-
noxy radicals such as 1 with O2 (eq 1).1 Unfortunately,
the majority of phenoxy radicals have short lives. Hence,
the number of peroxides obtainable by this means is
limited. The thermochemical nature of the peroxides is
unique in that homolytic fission of not only O-O but also
C-O bonding is possible.2 Addition of new and varied
bis(oxocyclohexadienyl) peroxides may contribute to get-
ting deeper insight into their chemical nature. In this
paper, we report an entirely new principle for the
formation of bis(oxocyclohexadienyl) peroxides, which
appears to proceed through an ionic process.
strongly accelerated by an electrophilic halogen species
such as N-iodosuccinimide (NIS).4 It seemed reasonable
to us to anticipate that in the presence of a positive hal-
ogen compound 3 would react analogously with 4-hydro-
peroxy-4-alkyl-2,6-di-tert-butylcyclohexa-2,5-dienone 4,
giving rise to bis(oxocyclohexadienyl) peroxide 5, as
illustrated in Scheme 1. This was found to be the case.
4-Chloro-4-methyl-2,6-di-tert-butylcyclohexa-2,5-dien-
one (3a ) was allowed to react with an excess (3 molar
equiv) of 4-hydroperoxy-4-methyl-2,6-di-tert-butylcyclo-
hexa-2,5-dienone (4a ) in MeCN containing NIS (1 molar
equiv). The reaction was complete within 5 min at 35
°C. After workup, bis(1-methyl-3,5-di-tert-butyl-4-oxo-
cyclohexa-2,5-dienyl) peroxide (5a ) was isolated in 74%
yield (run 2, Table 1). It is noteworthy that 5a cannot
be formed from coupling of 4-methyl-2,6-di-tert-butylphe-
noxy radicals (6) with O2 (cf. eq 1) since it cannot compete
with extremely rapid disproportionation of 6 (eq 2).5,6 It
Resu lts a n d Discu ssion
In a recent investigation on the mechanism of the
reaction of 2,6-di-tert-butylphenol with an iodinating
agent, I2 and H2O2, in MeOH, it has been proposed that
the reaction involves electrophilic assistance by I2 of
deiodination of the intermediary 4-iodocyclohexa-2,5-
dienone, generating the 4-oxocyclohexa-2,5-dienyl cation
(or the phenoxy cation).3 In accordance with this postu-
lation, it has been more recently proved that nucleophilic
displacement of 4-halogeno-4-alkyl-2,6-di-tert-butylcy-
clohexa-2,5-dienones 3 with alcohols affording 4-alkoxy-
4-alkyl-2,6-di-tert-butylcyclohexa-2,5-dienones can be
* To whom correspondence should be addressed. Tel.: 0797-87-
5111. Fax: 0797-87-5666.
X Abstract published in Advance ACS Abstracts, November 15, 1997.
(1) (a) Altwicker, E. R. Chem. Rev. (Washington, D.C.) 1967, 67, 475
and references therein. (b) Cook, C. D.; Kuhn, D. A.; Fianu, P. J . Am.
Chem. Soc. 1956, 78, 2002. (c) Roginskii, V. A.; Plekhanova, L. G.;
Dubinskii, V. Z.; Nikiforov, G. A.; Miller, V. B.; Ershov, V. V. Izv. Akad.
Nauk SSSR, Ser. Khim. 1975, 1327. (d) Batanov, I. A.; Nikiforov, G.
A.; Ershov, V. V. Izv. Akad. Nauk SSSR, Ser. Khim. 1982, 359.
(2) (a) Reference 1a, p 509. (b) Forrester, A. R.; Hey, J . M.; Thomson,
R. H. In Organic Chemistry of Stable Free Radicals; Academic Press:
New York, 1968; p 281.
is clear that NIS plays an essential role in the formation
(3) Omura, K. J . Org. Chem. 1996, 61, 2006.
(4) The results will be published separately.
S0022-3263(97)01307-8 CCC: $14.00 © 1997 American Chemical Society

Bis(oxocyclohexadienyl) Peroxide Formation
J . Org. Chem., Vol. 62, No. 25, 1997 8791
Ta ble 1. P er oxid e 5a fr om Rea ction of Ch lor o Dien on e
3a a n d Hyd r op er oxy Dien on e 4a a
Ta ble 2. P er oxid es 5 fr om NIS-In d u ced Rea ction s
of Ch lor o Dien on es 3 a n d Hyd r op er oxy Dien on es 4
in MeCNa
positive
halogen
time
(min)
recovery of
yield of
run
solvent
3a (%b)
5a c (%d)
1
2
3
4
5
6
7
8
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
DME
60
5
30
5
5
5
85
0
0
98
10
35
40
0
trace
74
58
2
36
23
1
NIS
NISe
NBS
I2
Br2
NIS
NIS
60
60
PhH
14
a
A mixture of 3a (1 mmol), 4a (3 mmol), and a positive halogen
reagent (1 mmol) in a solvent (15 mL) was stirred at 35 °C for the
b
time indicated in the table. Estimated from the 1H NMR
d
spectrum of the crude reaction product. c Isolated product. (Mol/
mol 3a employed) × 100. e With 0.33 mmol of NIS.
of 5a , since if NIS is omitted, little 5a was yielded and a
large quantity of 3a was recovered unchanged despite
an extended reaction time (run 1, Table 1). However, it
is questionable whether employment of a stoichiometric
amount (1 molar equiv) of NIS is mandatory for complet-
ing the reaction, as explained below. The reaction of NIS
with 3a may produce an equivalent amount of ICl (cf.
Scheme 1), which may be also active as an electrophilic
halogen species. Thus, the ICl may also react with more
of 3a (hereby generating fresh ICl) and contribute toward
forming 5a . When the reaction of 3a , 4a , and NIS was
started, the originally light yellow or light brown mixture
soon turned brownish orange, indicative of rapid forma-
tion and persistence of a molecular halogen (ICl). As the
result, employment of a catalytic amount of NIS may in
principle suffice to complete the reaction between 3a and
4a (eq 3). The reaction of 3a with 4a in MeCN in the
run
3
4
5b (yield, %c)
1d,e
2e
3a
3b
3b
3c
3d
3e
3a
3b
3c
3d
3e
4a
4a
4a
4b
4b
4c
4b
4b
4a
4a
4b
5a (74)
5a (57)
5a (48)
5b (26)
5b (0)
5c (61)
5d (61)
5d (49)
5d (58)
5d (0)
3e,f
4
5g,h
6
7
8
9
10h
11
5e (37)
a
A mixture of 3 (1 mmol), 4 (3 mmol), and NIS (1 mmol) in
MeCN (15 mL) was stirred for 10 min at 35 °C. Consumption of
3 was complete or essentially complete. Isolated product. c (Mol/
b
mol 3 employed) × 100. Identical with run 2, Table 1. e Reaction
d
for 5 min. f NBS was employed in place of NIS. Reaction for 60
g
h
min. Compound 3d was recovered quantitatively.
or Br2 proceeded considerably fast, but 5a was furnished
in an unsatisfactory yield (runs 5 or 6, Table 1). The
effect of solvent was also studied. Nonpolar solvents were
found to be much less suitable than MeCN. Thus, the
reaction of 3a , 4a and NIS in DME was quite slow and
gave little 5a (run 7, Table 1). The reaction in benzene
for an extended time could complete the consumption of
3a , but 5a was obtained only in a limited amount (run
8, Table 1). It deserves mentioning that MeCN is also
the most appropriate solvent for the aforementioned
positive-halogen assisted nucleophilic displacement of 3
with alcohols.4
The NIS-induced reactions in MeCN of various com-
binations of halogenodienones 3 and hydroperoxy di-
enones 4 deserved investigation. The results are sum-
marized in Table 2. Peroxide 5a was also obtained by
the rapid reaction of 4-bromo-4-methylcyclohexa-2,5-
dienone (3b) with 4a , although the yield was moderate
(run 2, Table 2). Replacement of NIS by NBS in this
reaction also afforded 5a rapidly (run 3, Table 2).
Exposure of 4-bromo-4-tert-butylcyclohexa-2,5-dienone
(3c) to 4-hydroperoxy-4-tert-butylcyclohexa-2,5-dienone
(4b) produced bis(1,3,5-tri-tert-butyl-4-oxocyclohexa-2,5-
dienyl) peroxide (5b) in a relatively small yield (run 4,
Table 2). Complete lack of the reaction was observed
between 4-chloro-4-tert-butylcyclohexa-2,5-dienone (3d )
and 4b despite an extended reaction time (run 5, Table
2). There appears to be little interaction between 3d and
NIS. The new methodology successfully afforded a new
peroxide, bis(1-ethyl-3,5-di-tert-butyl-4-oxocyclohexa-2,5-
dienyl) peroxide (5c) in a fair yield. This was ac-
complished when 4-chloro-4-ethylcyclohexa-2,5-dienone
(3e) and 4-hydroperoxy-4-ethylcyclohexa-2,5-dienone (4c)
NIS (catalyst)
3a + 4a
(3)
5a + HCl
presence of 0.33 molar equiv of NIS was complete within
30 min and afforded 0.58 molar equiv (58% yield) of 5a
(run 3, Table 1).7 This yield was appreciably lower than
that with 1 molar equiv of NIS. In this regard, it is
pointed out that different products accompany 5a de-
pending on by which positive halogen compound the
reaction is mediated. When mediated by NIS, practically
neutral and presumably inactive succinimide may be
generated (Scheme 1), but when mediated by ICl, an acid,
HCl, is formed (eq 3), which might cause a reaction(s)
disadvantageous to the formation or preservation of 5a .
In our present study, all the reactions of 3 and 4 other
than run 3, Table 1, were carried out with a stoichio-
metric amount (1 molar equiv) of positive halogen species.
The effect of electrophilic halogen compounds other
than NIS was also investigated on the formation of 5a
from the reaction between 3a and 4a in MeCN, and the
results were compared (see also Table 1). N-Bromosuc-
cinimide (NBS) was far less effective than NIS, and only
a very small amount of 5a was obtained, the rest having
been unreacted 3a (run 4, Table 1). The reaction with I2
(5) (a) Mu¨ller, E.; Mayer, R.; Heilmann, U.; Scheffler, K. Liebigs.
Ann. Chem. 1961, 645, 66. (b) Coppinger, G. M. J . Am. Chem. Soc.
1964, 86, 4385.
(6) However, peroxide 5a has been isolated in an unspecified yield
among other products from the reaction of 4-hydroxy-4-methyl-2,6-di-
tert-butylcyclohexa-2,5-dienone with performic acid. See: Benjamin,
B. M.; Raaen, V. F.; Hagaman, E. W.; Brown, L. L. J . Org. Chem. 1978,
43, 2986.
(7) Since ICl can play the role that NIS does in the peroxide
formation, it is possible that not all of NIS employed initially in runs
2 or 3 was consumed.

8792 J . Org. Chem., Vol. 62, No. 25, 1997
Omura
were employed as the reactants (run 6, Table 2). Syn-
thesis of 5c by coupling of 4-ethyl-2,6-di-tert-butylphe-
noxy radicals with O2 (cf. eq 1) has not been feasible due
to rapid disproportionation of the radical.5a
As exemplified above, symmetrically substituted bis(oxo-
cyclohexadienyl) peroxides 5 (R ) R′) including ones
unobtainable by the classical procedure (eq 1) can be
synthesized by our new method.
The concept of the new method predicts that prepara-
tion of unsymmetrically substituted bis(oxocyclohexadi-
enyl) peroxides 5 (R * R′) may likewise be possible. It
should be emphasized that, as Altwicker has previously
pointed out, formation of a “mixed peroxide” such as 5
(R * R′) from a competition of two different phenoxy
radicals for O2 is difficult,1a possibly owing to different
stability and/or different reactivity toward O2, of the
different radicals. As far as we are aware, a very rare
example of an isolated “mixed peroxide” is 8.8 Peroxide
8 is formed during thermal decomposition of a transan-
nular ozonide of an anthracene derivative in the presence
of 2,4,6-tri-tert-butylphenol. This decomposition is sug-
gested to generate phenoxy radical 7, 2,4,6-tri-tert-
butylphenoxy radical and O2. Peroxide 5b is formed
simultaneously, as may be anticipated (eq 4). It was
It is our feeling that a bulky 4-alkyl substituent in
halogeno dienone 3 or hydrperoxy dienone 4 tends to
decrease the yield of bis(cyclohexadienyl) peroxide 5 in
the NIS reaction in MeCN, but the data accumulated
above may not be enough to make a conclusive statement.
In general, bis(oxocyclohexadienyl) peroxides 5 obtained
in this study were isolated simply by passing the crude
reaction product through neutral alumina (Activity II)
with petroleum ether. Products other than 5 were not
investigated in detail.
By our new method, synthesis of a variety of fresh
bis(oxocyclohexadienyl) peroxides may be expected when
it is taken into account that diverse methods are avail-
able for preparing 4-(or 6-)halogeno-9 and 4-(or 6-)hydro-
peroxycyclohexa-2,5-(or 2,4-)dienones10 as the starting
materials.
Exp er im en ta l Section
1H (90 MHz) and 13C (22.6 MHz) NMR and IR spectra were
taken in CDCl3 and in CHCl3, respectively. TLC was run on
SiO2.
Sta r tin g Ma ter ia ls. Halogeno dienones 3a ,9d 3b,5b 3c,5b
and 3d 9d were synthesized according to the reported methods.
Compound 3e was prepared in a nearly quantitative yield by
adaptation of the reported method for the preparation of 3a
and 3d ,9d using 4-ethyl-2,6-di-tert-butylphenol (0.1 mol), sul-
furyl chloride (0.12 mol), and trimethyl phosphate (190 mL in
total) at ca. 3 °C. Light yellow crystals from hexane: mp 46-
1
47 °C; H NMR δ 6.54 (s, 2H), 2.08 (q, J ) 7.4 Hz, 2H), 1.25
(s, 18H), 0.86 (t, J ) 7.4 Hz, 3H); IR 1659, 1640 cm-1. Anal.
Calcd for C16H25OCl: C, 71.49; H, 9.37; Cl, 13.19. Found: C,
71.21; H, 9.64; Cl, 13.33.
Hydroperoxy dienones 4a and 4b were synthesized accord-
ing to the reported procedure.10g Employment of a mixture of
EtOH and H2O (20:1) in place of EtOH alone as the solvent
made completion of the reaction appreciably faster. Compound
4c was obtained in 70% yield by adaptation of the method
reported for the preparation of 4a and 4b10g by using 4-ethyl-
2,6-di-tert-butylphenol (50 mmol), KOH (0.6 mol), a mixture
of EtOH (400 mL) and H2O (20 mL), and O2. Colorless crystals
from petroleum ether: mp 89-91 °C; 1H NMR δ 7.79 (s, 1H),
6.51 (s, 2H), 1.70 (q, J ) 7.5 Hz, 2H), 1.25 (s, 18H), 0.76 (t, J
) 7.5 Hz, 3H); IR 3520, 3300 (br), 1665, 1642 cm-1
. Anal.
proved that our procedure can indeed make “mixed
peroxides”, 5 (R * R′). For instance, with the assistance
of NIS in MeCN, 3a successfully coupled with 4b, forming
a new “mixed peroxide”, 1-methyl-3,5-di-tert-butyl-4-
oxocyclohexa-2,5-dienyl 1,3,5-tri-tert-butyl-4-oxocyclohexa-
2,5-dienyl peroxide (5d ) in 61% yield (run 7, Table 2).
As expected, the same peroxide (5d ) was obtainable also
from combination of 3b with 4b (run 8, Table 2) or of 3c
with 4a (run 9, Table 2). From runs 7-9, Table 2, neither
symmetrically substituted bis(oxocyclohexadienyl) per-
oxide 5a nor 5b was detectable, which is compatible with
our originally proposed mechanism (Scheme 1). Allowing
3d to react with 4a for a long period, however, did not
produce 5d at all, again supporting the assumed lack of
interaction between 3d and NIS (run 10, Table 2).
Another new “mixed peroxide”, (1-ethyl-3,5-di-tert-butyl-
4-oxocyclohexa-2,5-dienyl) 1,3,5-tri-tert-butyl-4-oxocyclo-
hexa-2,5-dienyl peroxide (5e) was furnished from the
reaction of 3e with 4b (run 11, Table 2). Finally, the NIS-
induced reaction of 3a with 4c or of 3e with 4a gave a
crystalline product, presumed to be “mixed peroxide” 5f,
in a good yield (70% or less). Unfortunately, stubborn
contamination by an impurity (impurities) has prevented
purification of the product.
Calcd for C16H26O3: C, 72.14; H, 9.84. Found: C, 72.09; H,
10.08.11
Other reagents were used as received.
P er oxid e 5a fr om Ch lor o Dien on e 3a a n d Hyd r op er -
oxy Dien on e 4a As Assisted by P ositive Ha logen Re-
a gen ts (Ta ble 1). To 3a (255 mg, 1 mmol) and a positive
halogen compound (1 mmol) placed in a bottle was added in
one portion a solution of 4a (756 mg, 3 mmol) in a solvent (15
mL). In run 1, Table 1, a positive halogen reagent was
omitted. In run 3, Table 1, 0.33 mmol (75 mg) of NIS was
employed as the positive halogen compound. The mixture in
the stoppered bottle was stirred at 35 °C for the time indicated
(9) (a) Ershov, V. V.; Volod’kin, A. A. Izv. Akad. Nauk SSSR, Otd.
Khim. Nauk 1962, 2150. (b) Ershov, V. V.; Volod’kin, A. A. Izv. Akad.
Nauk SSSR, Otd. Khim. Nauk 1963, 893. (c) Calo`, V; Lopez, L.; Pesce,
G.; Ciminale, F.; Todesco, P. E. J . Chem. Soc., Perkin Trans. 2 1974,
1192. (d) Pearson, D. E.; Venkataramu, S. D.; Childers, W. E., J r.
Synth. Commun. 1979, 9, 5. (e) Fischer, A.; Henderson, G. N. Can. J .
Chem. 1979, 57, 552. (f) Fischer, A.; Henderson, G. N. Can. J . Chem.
1983, 61, 1045.
(10) (a) Coppinger, G. M. J . Am. Chem. Soc. 1957, 79, 2758. (b)
Bickel, A. F.; Gersmann, H. R. Proc. Chem. Soc. 1957, 231. (c)
Gersmann, H. R.; Bickel, A. F. J . Chem. Soc. 1962, 2356. (d) Bacon,
R. G. R.; Kuan, L. C. Tetrahedron Lett. 1971, 3397. (e) Barton, D. H.
R.; Magnus, P. D.; Quinney, J . C. J . Chem. Soc., Perkin Trans. 1 1975,
1610. (f) Nishinaga, A.; Itahara, T.; Shimizu, T.; Matsuura, T. J . Am.
Chem. Soc. 1978, 100, 1820. (g) Nishinaga, A.; Shimizu, T.; Matsuura,
T. J . Org. Chem. 1979, 44, 2983. (h) Futamura, S.; Yamazaki, K.; Ohta,
H.; Kamiya, Y. Bull. Chem. Soc. J pn. 1982, 55, 3852. (i) J efford, C.
W.; Bernardinelli, G.; McGoran, E. C. Helv. Chim. Acta 1984, 67, 1952.
(j) Yumbie, N. P.; Thompson, J . A. Chem. Res. Toxicol. 1988, 1, 385.
(8) Rigaudy, J .; Chelu, G.; Cuong, N. K. J . Org. Chem. 1985, 50,
4474.

Bis(oxocyclohexadienyl) Peroxide Formation
J . Org. Chem., Vol. 62, No. 25, 1997 8793
in Table 1. The mixture was poured into a stirred, cold
solution of NaHSO3 (0.9 g) in water (150 mL). The stirring
was continued for 2 min. The mixture was extracted with
ether (150 mL × 2). The extract was washed with water, dried
over anhydrous Na2SO4 and evaporated to dryness. The
residue was chromatographed on neutral Al2O3 (Merck, 70-
230 mesh, Activity II, 130 g) with petroleum ether, affording
5a as nearly colorless crystals. Colorless crystals from hex-
mL) for 2 min resulted in quantitative recovery of unreacted
5b. Replacement of 5b in this reaction by 3c (100 mg) resulted
in quantitative recovery of unreacted 3c, while that by 4b (100
mg) gave 2,6-di-tert-butyl-p-benzoquinone in ca. 90% yield as
deep yellow crystals, identical with an authentic sample13 (1H
NMR and TLC): mp 64-67 °C (lit.13 mp 65-68 °C).
Peroxide 5c (303 mg, 61%) was obtained from run 6, Table
2, by using 3e (269 mg) and 4c (798 mg). Light yellow crystals
1
1
ane: mp 124 °C (lit.6 mp 125 °C); IR 1665, 1642 cm-1. The H
from hexane: mp 132-133 °C; H NMR δ 6.47 (s, 4H), 1.57
and 13C NMR spectra were compatible with those reported for
5a .6 Anal. Calcd for C30H46O4: C, 76.55; H, 9.85. Found: C,
76.35; H, 9.94. The highest amount of 5a (347 mg, 74%) was
obtained from run 2, Table 1.
(q, J ) 7.4 Hz, 4H), 1.26 (s, 36H), 0.71 (t, J ) 7.4 Hz, 6H); 13
C
NMR δ 186.6, 148.3, 140.3, 80.2, 34.9, 30.1, 29.6, 7.8; IR 1662,
1639 cm-1
.
Anal. Calcd for C32H50O4: C, 77.06; H,10.10.
Found: C, 76.96; H, 10.38.
In runs 5 or 6, Table 1, a solution of 4a (3 mmol) in MeCN
(10 mL) and a solution of I2 (254 mg, 1 mmol) (run 5, Table 1)
or Br2 (160 mg, 1 mmol) (run 6, Table 1) in MeCN (5 mL) was
added successively to 3a (1 mmol). The mixture in a stoppered
bottle was stirred at 35 °C for the time indicated in Table 1.
Treatment of 5a (100 mg) in MeCN (15 mL) with a stirred,
cold solution of NaHSO3 (0.9 g) in water (150 mL) for 2 min
resulted in quantitative recovery of unreacted 5a . Replace-
ment of 5a in this reaction by 3a (100 mg) gave nearly
quantitative recovery of unreacted 3a , while that by 4a (100
mg) provided 4-hydroxy-4-methyl-2,6-di-tert-butylcyclohexa-
2,5-dienone in ca. 90% yield as colorless crystals with mp 111-
113 °C (lit.10a mp 112-113 °C).
P er oxid es 5 fr om NIS-In d u ced Rea ction s of Ha logen o
Dien on es 3 a n d Hyd r op er oxy Dien on es 4 in MeCN
(Ta ble 2). To 3 (1 mmol) and NIS (225 mg, 1 mmol) placed
in a bottle was added in one portion a solution of 4 (3 mmol)
in MeCN (15 mL). The mixture in the stoppered bottle was
stirred at 35 °C for the time indicated in Table 2. The reaction
mixture was worked up and the crude product mixture was
analyzed in the manner described above for the runs in Table
1.
Peroxide 5d (313 mg, 61%) was obtained from run 7, Table
2, by using 3a (255 mg) and 4b (882 mg). Colorless crystals
1
from MeOH: mp 122.5-124 °C; H NMR δ 6.63 (s, 2H), 6.53
(s, 2H), 1.26 (s, 36H), 1.21 (s, 3H), 0.83 (s, 9H); 13C NMR δ
186.4, 148.6, 147.2, 141.7, 140.4, 83.2, 76.8, 41.1, 35.1, 34.7,
29.6, 29.5, 25.9, 23.6; IR 1663, 1640 cm-1
. Anal. Calcd for
C33H52O4: C, 77.28; H, 10.23. Found: C, 77.00; H, 10.25.
Peroxide 5e (196 mg, 37%) was obtained from run 11, Table
2, by using 3e (269 mg) and 4b (883 mg). Light yellow crystals
1
from hexane: mp 107-108 °C; H NMR δ 6.65 (s, 2H), 6.50
(s, 2H), 1.54 (q, J ) 7.9 Hz, 2H), 1.27 (s, 36H), 0.84 (s, 9H),
0.72 (t, J ) 7.7 Hz, 3H); 13C NMR δ 186.7, 186.4, 148.5, 148.3,
140.8, 140.4, 83.1, 80.1, 40.9, 35.1, 34.9, 29.6, 25.9, 7.7; IR 1663,
1640 cm-1
. Anal. Calcd for C34H54O4: C, 77.52; H, 10.33.
Found: C, 77.50; H, 10.58.
Ack n ow led gm en t. The author is thankful to Yuhko
Nagumo and Yuhko Akahori (students) for experimental
contributions.
J O971307S
(11) Compound 4c appears in the literature, but its physical and
spectral data do not appear to have been cited. Cf. (a) Nishinaga, A.;
Nakamura, K.; Matsuura, T. Tetrahedron Lett. 1978, 3557. (b) Wand,
M. D.; Thompson, J . A. J . Biol. Chem. 1986, 261, 14049. (c) Reference
10j.
(12) Omura, K. J . Org. Chem. 1984, 49, 3046.
(13) Omura, K. J . Org. Chem. 1989, 54, 1987.
Peroxide 5b (143 mg, 26%) was obtained from run 4, Table
2, by using 3c (341 mg) and 4b (881 mg). Colorless crystals
from hexane, identical with an authentic sample12 (1H NMR
and TLC): mp 138-139 °C (lit.12 mp 139-140.5 °C).
Treatment of a solution of 5b (100 mg) in MeCN (15 mL)
with a stirred, cold solution of NaHSO3 (0.9 g) in water (150