On the susceptibility of organic peroxy bonds to hydride reduction

Article abstract of DOI:10.1021/jo050139y

Reduction of organic molecules that contain a peroxy bond is broadly considered as a "risky" and uncertain operation when cleavage of the peroxy linkage is not desired. For this reason, such reduction steps are normally avoided at the planning stage of the synthesis when possible. As a natural consequence, the information in the literature about the susceptibility of organic peroxy bonds to reducing species is scant. In this work the tolerance of organic peroxy bonds to some common hydride reductants was examined systematically for the first time. Using reduction of ester group to alcohol as a probe, LiAlH₄, LiAlH(O^tBu)₃, LiBHEt ₃, and LiBH₄ were found to be significantly better than other reductants examined when taking into consideration both the completeness of the reduction of ester groups and the peroxy bond survival rate. LiBH ₄ appeared to be the most suitable reductant for the reduction under discussion, not only because of the high reduction yields/excellent compatibility with peroxy bonds, but also because of the advantages in practical aspects. The results disclosed herein may (hopefully) provide a handy reference for dealing with reduction of other peroxy bond-containing molecules in the future.

Full text of DOI:10.1021/jo050139y

On the Susceptibility of Organic Peroxy Bonds to Hydride
Reduction
Hong-Xia Jin, He-Hua Liu, Qi Zhang, and Yikang Wu*
State Key Laboratory of Bioorganic & Natural Products Chemistry, Shanghai Institute of Organic
Chemistry, Chinese Academy of Sciences, 354 Fenglin Road, Shanghai 200032, China
yikangwu@mail.sioc.ac.cn
Received January 23, 2005
Reduction of organic molecules that contain a peroxy bond is broadly considered as a “risky” and
uncertain operation when cleavage of the peroxy linkage is not desired. For this reason, such
reduction steps are normally avoided at the planning stage of the synthesis when possible. As a
natural consequence, the information in the literature about the susceptibility of organic peroxy
bonds to reducing species is scant. In this work the tolerance of organic peroxy bonds to some
common hydride reductants was examined systematically for the first time. Using reduction of
ester group to alcohol as a probe, LiAlH4, LiAlH(OtBu)3, LiBHEt3, and LiBH4 were found to be
significantly better than other reductants examined when taking into consideration both the
completeness of the reduction of ester groups and the peroxy bond survival rate. LiBH4 appeared
to be the most suitable reductant for the reduction under discussion, not only because of the high
reduction yields/excellent compatibility with peroxy bonds, but also because of the advantages in
practical aspects. The results disclosed herein may (hopefully) provide a handy reference for dealing
with reduction of other peroxy bond-containing molecules in the future.
Introduction
as artemisinin in the West), where the six-membered
lactone functionality was smoothly reduced to lactol in
high yield. It should be noted, however, that this reduc-
tant usually is not powerful enough to reduce acyclic ester
functionality. Treatment of qinghaosu with LiAlH4 led3
to substantial cleavage of the peroxy linkage. Reduction
of the lactone carbonyl to the alcohol level is also known,
either using the one-pot procedure (NaBH4/BF3‚OEt2)
reported4 by Jung and co-workers or the two-step proce-
dure (first DIBAL-H (iBu2AlH)/-78 °C then Et3SiH/BF3‚
OEt2) developed by Avery5 and co-workers. Ziffer and co-
workers6 also used DIBAL-H to reduce an ester group
(to corresponding aldehyde and alcohol in 50% and
Peroxy bonds are among the most “fragile” covalent
bonds found in organic compounds, with an average1a
bond energy of only 34 kcal/mol (less than half of that
for C-C single bond). Apart from heat and UV light, they
are also sensitive to various reducing species. Therefore,
reduction of any functional group in the presence of a
peroxy functionality is considered by most chemists as a
potentially insecure/risky operation when cleavage of the
peroxy bond is not desired. As a consequence, the number
of existing examples of such reductions in the literature
is rather limited. Because “reduction without cleaving
peroxy bonds” has not been covered by the major refer-
ence books1b-e on organic peroxides and has never been
designed as an entry point in documentation, retrieval
of the scant relevant information in the literature is also
a time-consuming task.
(2) (a) Liu, J.-M.; Ni, M.-Y.; Tu, A.-A.; Wu, Z.-H.; Wu, Y.-L.; Zhou,
W. S. (Hua Hsueh Hsueh Pao) Hua Xue Xue Bao (now often translated
as Acta Chim. Sinica) 1979, 37, 129-143; Chem. Abstr. 1979, 92,
94594. (b) Li, Y.; Wu, Y.-L. Curr. Med. Chem. 2003, 10, 2197-2230
and the references therein.
(3) Wu, Y.-L.; Zhang, J.-L. You Ji Hua Xue (now often translated
as Chin. J. Org. Chem.) 1986, 154-156; Chem. Abstr. 1986, 105,
191426n.
The earliest one appeared to be the NaBH4 reduction2a
of qinghaosu2b (an outstanding antimalarial agent, known
(4) Jung, M.; Li, X.; Bustos, D. A.; ElSohly, H. N.; McChesney, J.
D.; Milhous, W. K. J. Med. Chem. 1990, 33, 1516-1518.
(5) (a) Avery, M. A.; Mehrotra, S.; Johnson, T. L.; Bonk, J. D.;
Vroman, J. A.; Miller, R. J. Med. Chem. 1996, 39, 4149-4155. (b)
Avery, M. A.; Alvim-Gaston, M.; Vroman, J. A.; Wu, B.; Ager, A.; Peters,
W.; Robinson, B.; Charman, W. J. Med. Chem. 2002, 45, 4321-4335.
(6) Mekonnen, B.; Weiss, E.; Katz, E.; Ma, J.; Ziffer, H.; Kyle, D.
Bioorg. Med. Chem. 2000, 8, 1111-1116.
(1) (a) Cremer, D. In The Chemistry of Peroxides; Patai, S., Ed.; John
Wiley & Sons: Chichester, U.K., 1983; Chapter 1, p 3. (b) Davies, A.
G. Organic Peroxides; Butterworth: London, 1961. (c) In Organic
Peroxides; Swern, D., Ed.; Wiley-Interscience: New York, 1970; Vol.
1-2. (d) In Organic Peroxides; Swern, D., Ed.; Wiley-Interscience: New
York, 1972; Vol. 3. (e) In The Chemistry of Peroxides; Patai, S., Ed.;
Wiley: Chichester, 1983.
10.1021/jo050139y CCC: $30.25 © 2005 American Chemical Society
Published on Web 04/26/2005
4240
J. Org. Chem. 2005, 70, 4240-4247

Susceptibility of Organic Peroxy Bonds
SCHEME 1
29%yield, respectively) in the side-chain of a qinghaosu
analogue. Their results showed that at least 79% of the
peroxy bond was not broken at -78 °C after 1.5 h.
In their efforts to establish the structure of some
marine sponges metabolites, Stierle and Faulkner7 suc-
cessfully reduced an ester group (88% yield on a 20-mg
scale) with LiAlH(OtBu)3 (a mild reductant often used to
reduce ketones/aldehydes but normally not esters) in
refluxing diethyl ether without breaking the peroxy bond
in the molecule. Later, the same reagent was also
employed by Kitagawa8 in reduction of a similar peroxy
bond-containing ester but without giving any information
about the yield or experimental details.
organic chemists mainly as radical initiators, such a
knowledge deficiency perhaps did not cause much incon-
venience (because at that time there were rarely occa-
sions, where a reduction must be done in the presence of
a peroxy functionality). Now, as organic peroxides become
an important13
class of antimalarial agents (many other
activities are also known) and synthesis of new organic
peroxides is already a commonplace, the implicit yet
strongly influencing “no reduction” convention perhaps
deserves reconsiderationsis it really necessary and
worthy for a synthesis to take extra steps and a round-
about route just for avoiding involvement of a reduction
step? Given the large number of the syntheses of organic
peroxides in recent years and obvious absence of convinc-
ing experimental data showing how sensitive the peroxy
bonds are to for instance the hydride reductants, we
believe efforts to find out an answer to this question are
well-warranted now.
To gain a general knowledge of the stability of peroxy
linkage to commonly employed hydride reducing agents,
we conducted a systematic investigation. The main
results are reported below, which hopefully may serve
as a quick reference for dealing with other peroxy bond-
containing substrates in the future when a reducing
agent is involved.
Reduction of ester groups with9 LiAlH4 without cleav-
ing the co-present peroxy bond was also known. In one
case9a (on a 20 mg scale) the yield was essentially
quantitative, and in another case9b the yield was not
reported. Slight variation in, for example, the stereo-
chemistry of the substrate might lead9c to substantially
changed yields. Prolonged reaction time from 3 to 7 h
also led to high levels of peroxy bond-breaking products.
Dussault10 and co-workers examined LiAlH4/-78 °C or
0 °C, NaBH4/rt or 0 °C, and DIBAL-H/-78 °C in the
reduction of a peroxy-containing aldehyde. The yield with
LiAlH4 at -78 °C was excellent. However, the reduction
required very careful operation and did not work so well
with another closely related substrate. DIBAL-H was also
successfully employed by Porter11 and co-workers in
reducing an ester group in the presence of a hindered
(with a tertiary carbon on each side of the peroxy linkage)
peroxy bond.
Reduction of peroxy bond-containing ketones (51-62%
yield)/cyclic carbonate (yield not specified) with LiBH4
was briefly communicated by Xu12a and co-workers. Very
recently, in a major breakthrough12b,c in developing novel
antimalarial agents Vennerstrom and co-workers cleanly
reduced12d an ethyl ester functionality in a highly stable
ozonide (with an adamantanyl and a cyclohexanyl on the
two terminals of the peroxy bond, respectively) with
LiBH4 in the presence of 10 mol % of LiBHEt3.
Because in most of the above-mentioned investigations
the reductions were performed either as part of structural
elucidation of natural products or a single step of a total
synthesis, the available information on the susceptibility
of peroxy bonds to the reductants is rather limited. In
the past, when organic peroxides were useful to most
Results and Discussion
Reduction of Peroxy Ester 1 with Various Com-
mon Reducing Agents. Our investigation on the stabil-
ity of peroxy bonds to hydride reductants started with
reduction of compound 1 (Scheme 1). We chose ester
functionality here for the reaction because on one hand
this type of transformations is very common in organic
synthesis, and on the other, such reductions usually need
more forcing conditions than reduction of aldehydes or
ketones. Thus, if a combination of reagent/conditions can
successful reduce an ester to alcohol without breaking
the peroxy bond, it is probably also safe and effective for
reducing similar peroxy-containing aldehdes/ketones to
corresponding alcohols. Some reductants that are known
to be able to convert ester groups into alcohols were then
tested (Table 1). Because reaction temperature was also
an important parameter in the present context, for those
potentially useful reductants, the reduction was often
examined at several temperatures commonly employed
in synthesis.
(7) Stierle, D. B.; Faulkner, D. J. J. Org. Chem. 1980, 45, 3396-
3401.
(8) Kobayashi, M.; Kondo, K.; Kitagawa, I. Chem. Phar, Bull. 1993,
41, 1324-1326.
(9) (a) Quinoa, E.; Kho, E.; Manes, L. V.; Crews, P.; Bakus, G. J. J.
Org. Chem. 1986, 51, 4260-4264. (b) Braekman, J. C.; Daloze, D.; De
Groote, S.; Fernandes, J. B.; Van Soest, R. W. M. J. Nat. Prod. 1998,
61, 1038-1042. (c) Albericci, M.; Braekman, J. C.; Daloze, D.; Tursch,
B. Tetrahedron 1982, 38, 1881-1890. (d) Muellner, U.; Huefnerm, A.;
Haslinger, E. Tetrahedron 2000, 56, 3893-3900.
(10) Dussault, P.; Sahli, A.; Westermeyer, T. J. Org. Chem. 1993,
58, 5469-5474.
(11) Porter, N. A.; Caldwell, S. E.; Lowe, J. R. J. Org. Chem. 1998,
63, 5547-5554.
(12) (a) Xu, X.-X.; Zhu, J.; Huang, D.-Z.; Zhou, W.-S. Tetrahedron
Lett. 1991, 32, 5785-5788. (b) Vennerstrom, J. L.; Arbe-Barnes, S.;
Brun, R.; Charman, S. A.; Chiu, F. C.; Chollet, J.; Dong, Y.; Dorn, A.;
Hunziker, D.; Matile, H.; McIntosh, K.; Padmanilayam, M.; Santo
Tomas, J.; Scheurer, C.; Scorneaux, B.; Tang, Y.; Urwyler, H.; Wittlin,
S.; Charman, W. N. Nature 2004, 430, 900-904. (c) O’Neill, P. M.
Nature 2004, 430, 838-839. (d) Tang, Y.; Dong, Y.; Karle, J. M.;
DiTusa, C. A.; Vennerstrom, J. L. J. Org. Chem. 2004, 69, 6470-6473.
LiAlH4 is one of the most common reagents for reduc-
ing esters. Therefore, in the present work we first
(13) See: e. g., (a) Vennerstrom, J. L.; Arbe-Barnes, S.; Brun, R.;
Charman, S. A.; Chiu, F. C.; Chollet, J.; Dong, Y.; Dorn, A.; Hunziker,
D.; Matile, H.; McIntosh, K.; Padmanilayam, M.; Santo Tomas, J.;
Scheurer, C.; Scorneaux, B.; Tang, Y.; Urwyler, H.; Wittlin, S.;
Charman, W. N. Identification of an antimalarial synthetic trioxolane
drug development candidate. Nature 2004, 430, 900-904. (b) Tang,
Y.; Dong, Y.; Vennerstrom, J. L. Synthetic peroxides as antimalarials.
Med Res Rev. 2004, 24, 425-448. (c) O’Neill, P. M. Medicinal
chemistry: a worthy adversary for malaria. Nature 2004, 430, 838-
839. (d) Casteel, D. A. Peroxy Natural Products. Nat. Prod. Rev. 1992,
289-312.
J. Org. Chem, Vol. 70, No. 11, 2005 4241

Jin et al.
TABLE 1. Reaction of the Peroxy Ester 1 to 2 with Various Hydride Reducing Agentsa
entry
reductant (mol equiv)
solvent
temperature
time
yield of 2
1
2
LiAlH4 (2.0)
LiAlH4 (2.4)
LiAlH4 (2.1)
Red-Al (1.5)
DIBAL-H (4.5)
THF
THF
THF
THF
CH2Cl2
Et2Of
Et2Of
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
Et2O
Et2O
Et2O
Et2O
0 °C
13 min
35 min
15 min
2.3 h
89%
0 °C
80%
3
24 °C
55%b
48%c
53%d
83%g
80%h
52%i
Tracesj
Tracesk
66%
4
0 °C∼29 °C
-78 °C
20 °C
5
3.6 h
e
6
LiAlH(OtBu)3 (6.7)
25 min
15 min
32 min
3 h
25 min
2.8 h
50 min
1 h
30 min
15 min
25 min
14 min
1 h
e
7
LiAlH(OtBu)3 (6.7)
Reflux
40∼55 °C
Reflux
-78 °C
-78 °C
0 °C
e
8
LiAlH(OtBu)3 (6.7)
9
BH3-THF (2.0)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
L-Selectride (3.0)
L-Selectride (2.5)
L-Selectride (2.5)
L-Selectride (2.3)
K-Selectride (2.0)
K-Selectride (2.0)
LiBHEt3 (2.5)
59%
29%
30 °C
l
-100 °C
-78 °C
0 °C
-
l
-
86%m
76%
LiBHEt3 (2.5)
33 °C
l
NaBH4/HSC2H4SH (1.5)
NaBH4/HSC2H4SH (1.5)
LiBH4 (2.0)
30∼40 °C
-
29 °C
30 min
1 h
8 min
25 min
2 h
N. R.n
89%o
87%p
91%
29 °C
LiBH4 (2.0)
reflux
0 °C
LiBH4 (2.0)
LiCl+KBH4 (3.4)
29 °C
85%
a For the general procedure see Experimental Section. b The starting 1 disappeared on TLC within 5 min. c The starting 1 was recovered
in 7% yield. d Some aldehyde was still left after reduction at -78 °C for 3.6 h. Warming to room temperature for 10 min led to full
reduction to alcohol. e About 30% solution in THF. f Final reaction medium was 1:4 THF (from the reagent solution)/Et2O (the added
solvent). g About 5% of the starting 1 was recovered. h The starting 1 disappeared on TLC within 4 min. i About 22% of the starting 1 was
recovered. j The reaction mixture comprised mainly the starting 1. k The starting 1 was near quantitatively recovered. l The product mixture
was very complicated. m The starting 1 disappeared on TLC after 14 min. n No reaction. o TLC showed disappearance of 1 within 15 min.
p About 5% of the starting 1 was recovered.
examined the tolerance of peroxy bond to LiALH4 reduc-
tion closely. The highest yield (89%) of 2 was obtained
at 0 °C with the reaction terminated within 13 min (entry
1). Longer reaction time or higher reaction temperature
all led to substantially reduced yields (entries 2 and 3).
In all these runs, the starting 1 was fully consumed,
which suggested that the mass imbalance was most likely
caused by cleavage of the peroxy bond/over reduction
(leading to highly polar/water soluble polyols and thus
lost in the aqueous workup). The yield of the isolated
peroxy alcohol therefore reflected the survival rate of the
peroxy bond in the reduction.
Reduction of 1 with Red-Al (NaAlH2(OCH2CH2OMe)2,
a reagent that has not been employed in reductions of
peroxy bond containing substrates) in THF was very
sluggish at 0 °C. At high temperatures (up to 29 °C) the
reaction proceeded substantially faster. After 2 h, more
than 90% of the starting 1 was consumed. The yield of
2, however, was only 48% (along with 7% of recovered 1,
entry 4). Thus, in this case the overall survival rate of
the peroxy bond was only about 55%.
DIBAL-H did not work so well in the reduction of 1.
When using 3 equiv of DIBAL-H to reduce 1 at -78 °C,
substantial amounts of the intermediate aldehyde re-
mained in the reaction mixture after reaction for up to
3.6 h. Quenching the reaction at this stage always led to
extensive formation of side-products. Addition of another
1.5 equiv of DIBAL-H did not make much difference.
Warming the reduction mixture from -78 °C to the
ambient temperature (then kept at the same temperature
for 10 min.) before the workup helped to transform all
intermediate aldehyde to alcohol. The yield of 2, however,
was still not so good (entry 5), signaling substantial
cleavage of the peroxy bond.
LiAlH(OtBu)3 normally reduces14 ketones and alde-
hydes but not esters. However, apart from the examples7,8
mentioned above, Ayers15 successfully utilized LiAlH(Ot-
Bu)3 (THF solution, not the commonly employed powder)
to reduce malonates into â-hydroxy propionates. For
comparison with other reducing agents, we also examined
this reductant with 1. We observed that the reducing
power of LiAlH(OtBu)3 depended on the state/form of the
reagent (we are unaware of any records of similar
observations in the literature). The reagent that came
as THF solution was apparently more powerful than the
powdered one (vide infra). It is also interesting to note
that addition of diethyl ether to the reaction system
significantly facilitated the reduction of ester group,
although the solubility of LiAlH(OtBu)3 in THF is much
higher than in diethyl ether. For instance, in 1:4 Et2O-
THF at 20 °C (entry 6) the starting 1 was reduced to 2
in 83% yield within 25 min, along with ca. 5% of
recovered 1 (i.e., 88% of the starting peroxy bond survived
or 12% of the peroxy bond was cleaved). Treatment of 1
in the same media at refluxing for 15 min led to full
consumption of the starting ester, giving the alcohol 2
in 80% yield (entry 7). In the absence of Et2O, 22% of
the 1 remained intact after reaction at 40-55 °C for 32
min and the yield of 2 was only 52% (entry 8). The
“survival rate” of the starting peroxy bond in this case
was 77%.
The reaction with L-Selectride (LiBH(iBu)3) at 30, 0,
-20, and -78 °C showed very similar TLC chromatogram
after 5-10 min’ reaction, with the product 2 being the
only spot developed away from the origin. In all runs the
(14) See, e. g.,: (a) Malek, J.; Cerny, M. Synthesis 1972, 217-234.
(b) Brown, H. C.; Yoon, N. M. J. Am. Chem. Soc. 1966, 88, 1464-1472.
(15) Ayers, T. A. Tetrahedron Lett. 1999, 40, 5467-5470.
4242 J. Org. Chem., Vol. 70, No. 11, 2005

Susceptibility of Organic Peroxy Bonds
SCHEME 2
starting 1 completely disappeared soon. However, more
careful inspection revealed that the rapid disappearance
of 1 on TLC observed with this reductant did not
necessarily mean consumption of 1, because most of the
starting 1 was recovered (entry 10) after reaction at -78
°C for 25 min and oxidative workup (H2O2/NaOH).
Extending the reduction time from 25 min to 2.8 h led to
substantially increased yield of 2 (66%, entry 11). If
running the reduction at 0 °C for 50 min, 2 was isolated
in 59% yield (entry 12). Raising the reaction temperature
to 30 °C lowered the yield of 2 even further (29%, entry
13). These results showed that L-Selectride could be
tolerated by the peroxy linkage only at low temperatures.
Because of the difficulty in monitoring the reaction,
relatively tedious workup (compared with e. g., LiBH4
mentioned below), and requirement for low-temperature
cooling facility, we did not continue with this reagent any
further.
all required syringe operation) was also a great advan-
tage. Finally, it should be mentioned that the LiBH4
prepared in situ from KBH4 and LiCl (entry 23) gave
more or less the same results as the commercially
available powder reagent.
K-Selectride (KBH(iBu)3) appeared to be much less
suitable than L-Selectride. Even at -100 °C the reaction
mixture was also rather complicated, while the starting
1 was still present in substantial quantities on TLC.
These observations showed that the peroxy bond was
significantly cleaved (entry 14). At -78 °C, the mixture
became even more complicated, although 1 disappeared
on TLC (entry 15). In contrast, LiBHEt3 (super-hydride,
which alone has not been examined on any peroxy bond-
containing substrate to our knowledge) seemed to be
better than L-Selectride. At 0 °C the starting 1 was fully
consumed after 14 min’s reaction and the alcohol 2 could
be isolated in 86% yield. Running the same reaction at
33 °C for 14 min lowered the yield to 76% (entry 17).
Formation of side products also became apparent. Similar
to what is mentioned above for L-Selectride, LiBHEt3 also
tended to coordinate with the substrates (rapid disap-
pearance of 1 on TLC) and thus made the TLC following
of the progress of the reduction rather difficult.
Tolerance of Ascaridole to Hydride Reduction. To
gain further knowledge of tolerance of peroxy bonds to
hydride reduction, we next examined the reaction (Scheme
2) of ascaridole 317a,b (or dihydroascaridole 517c,d) with
some of the most promising hydride reductants according
to the results mentioned above (with 417e or 617f,g as the
peroxy bond cleaved product, respectively). Table 2
summarizes the main results. Here a higher recovery rate
of the starting peroxy substrate (or lower yields of the
diol/side-products) corresponds to a set of safer reduction
conditions for the peroxy bond. In some of the experi-
ments, an equal molar amount of phenylethyl pivalate
was introduced to the reaction system to serve as an
indicator for the extent of exposure18 to the hydride
(showing relative ease of reduction of ester functionality
and cleavage of the peroxy bond).
As observed in the earlier experiments, LiAlH4 was
most satisfactorily used for reduction of esters at 0 °C
(Table 2, entry 1). Under such conditions the ester
function was often completely reduced, whereas the
peroxy bond was only partially cleaved (91% vs 8%). More
forcing conditions led to substantially increased cleavage
of the peroxy bond (entries 2-4).
We also examined NaBH4/HSCH2CH2SH, which could16
satisfactorily reduce esters according to Guida and co-
workers. Reduction of 1 with NaBH4/HSCH2CH2SH at
the ambient temperature (ca. 29 °C) resulted in es-
sentially no reactions within time up to 30 min (entry
18). Raising the reaction temperature to 40 °C (oil bath)
for 1 h led to complete disappearance of 1. However, the
product mixture was too complicated to allow for isolation
of the expected 2.
Lithium borohydride (LiBH4) is a well-known reagent
that is capable of reducing ester functionality to alcohol.
It is hence interesting to see whether peroxy bonds can
tolerate this hydride. We were very pleased to find that
treatment of 1 with 2 mol equiv of LiBH4 in Et2O at the
ambient temperature (ca. 29 °C) for 1 h resulted in the
expected alcohol 2 in 89% yield after chromatographic
isolation (entry 20). The reaction was quite clean and the
workup was very simple. More importantly, the yield was
apparently not so sensitive to the variation in tempera-
ture/reaction time (entries 21-22) as observed with, e.g.,
LiAlH4 or LiBHEt3. The possibility of performing the
reduction at the ambient temperature without special
precautions against moisture/air during the addition of
the reagent (cf. DIBAL-H, LiBHEt3 and L/K-Selectride
LiBH4 was remarkably milder and safer than LiAlH4.
Treatment of ascaridol (or dihydroascaridol) with this
reductant at 25 °C in diethyl ether for 3.3 h led to only
negligible amounts of peroxy bond cleaved products and
most of the starting peroxy substrate could be recovered
(entry 5). The peroxy bond cleavage, however, became
more evident with time. If extending the reaction time
to 16 h, the survived starting substrate dropped from 92%
to 60% (entry 6), and the yield of the diol and other
(17) (a) Jefford, C. W.; Jaber, A.; Boukouvalas, J. J. Chem. Soc.
Chem. Commun. 1989, 1916-1917. (b) Sels, B. F.; De Vos, D. E.;
Grobet, P. J.; Pierard, F.; Kirsch-De Mesmaeker, F.; Jacobs, P. A. J.
Phys. Chem. B 1999, 103, 11114. (c) Posner, G. H.; Tao, X.-L.;
Cumming, J. N.; Klinedinst, D.; Shapiro, T. A. Tetrahedron Lett. 1996,
37, 7225. (d) Jin, H.-X.; Liu, H.-H.; Wu, Y.-K. Chin. J. Chem. 2004,
22, 999-1002. (e) Suzuki, M.; Ohtake, H.; Kameya, Y.; Hamanaka,
N.; Noyori, R. J. Org. Chem. 1989, 54, 5292-5302. (f) Carman, R. M.;
Rayner, A. C. Aust. J. Chem. 1994, 47, 195-202. (g) Donkers, R. L.;
Workentin, M. S. Chem. Eur. J. 2001, 7, 4012-4020.
(18) The conditions that allowed for full reduction of the added
pivalate using LiALH4 or DIBAL-H were 18 °C/18 min or -78 °C/1 h,
respectively. With LiBH4 or LiALH(OtBu)3 as the reductant, however,
3-4 h at the ambient temperature was required.
(16) Guida, W. C.; Entreken, E. E.; Guida, A. R. J. Org. Chem. 1984,
49, 3024-3026.
J. Org. Chem, Vol. 70, No. 11, 2005 4243

Jin et al.
TABLE 2. Treatment of Ascaridol with LiAlH4, LiBH4, DIBAL-H, LiALH(OtBu)3, and LiBHEt3
entry
reductant (mol equiv)
solvent
temp. (°C)
time (h)
substrate recovery
diol 4
other productsa
1b,c
2b,c
3
LiAlH4 (1.5)
THF
0
25
1.0
2.0
6.5
2.0
3.3
15.8
6.5
5.8
2.0
1.0
2.7
1.0
5.8
13.9
1.7
6.5
2.5
79%
65%
57%
16%
92%
60%
65%
70%
83%
80%
78%
15%
86%
56%
75%
50%
21%
8%
13%
23%
40%
traces 6
17%
14%
11%
6%
LiAlH4 (1.5)
THF
∼7%
∼8%
LiAlH4 (1.5)
THF
32
4
LiAlH4 (1.5)
THF
reflux
25
25
32
∼24%
5b,d
6
LiBH4 (2.0)
Et2O
LiBH4 (2.0)
Et2O
∼10%
∼7%
∼7%
∼5%
∼3%
∼6%
∼48%
7
LiBH4 (2.0)
Et2O
8
LiBH4 (2.0)
Et2O
reflux
reflux
-78
-78
-30
25
9
LiBH4 (2.0)
THF
10b,e
11
12
13b,f
14
15b,f
16
17
DIBAL-H (3.0)
DIBAL-H (3.0)
DIBAL-H (3.0)
LiAlH(OtBu)3 (6.7)
LiAlH(OtBu)3 (6.7)
LiBHEt3 (2.0)
LiBHEt3 (2.0)
LiBHEt3 (2.0)
CH2Cl2
CH2Cl2
CH2Cl2
Et2O/ THF
Et2O/ THF
THF
3%
6%
26%
traces
32%
7%
30
0
4%
10%
15%
THF
THF
19
14%
50%
reflux
a Weight percentage with respect to the substrate. b Competition experiment (i.e., reduction of the Piv ester vs cleavage of the peroxy
bond) run in the presence of an equal molar amount of PhCH2CH2OPiv (Piv ) pivaloyl). c The yield of PhCH2CH2OH was 91%. d With
dihydroascaridole 5 as the substrate. e The yield of PhCH2CH2OH was 74% (along with 11% of recovered starting PhCH2CH2OPiv). f The
yield of PhCH2CH2OH was 92%.
TABLE 3. Treatment of 7 with LiAlH4, LiBH4, DIBAL-H,
LiBHEt3, and LiAlH(OtBu)3
SCHEME 3
reductant
mol equiv)
temp. time substrate
entry
solvent
THF
(°C)
(h) recoverya
1
2
LiAlH4 (1.5)
LiAlH4 (1.5)
LiAlH4 (1.5)
LiAlH4 (1.5)
LiBH4 (2.0)
0
2.0
2.0
88%
85%
64%
64%
86%
85%
42%
95%
90%
83%
86%
THF
THF
THF
Et2O
Et2O
CH2Cl2
THF
THF
THF
28
19
50
19
3
11
2.0
11
4
products climbed drastically. Raising the temperature did
not have so profound effects as extending the reaction
time. As shown by entries 7 and 8, substantially more
peroxy substrate survived at higher temperatures but
within shorter reaction time.
5
6
LiBH4 (2.0)
reflux 10
7
DIBAL-H (3.0)
LiBHEt3 (2.0)
LiBHEt3 (2.0)
LiBHEt3 (2.0)
26
0
3.0
2.0
22
2.0
19
8
9
18
10
11
reflux
19
DIBAL-H was a very powerful reductant. Even at -78
°C, after reaction for 1 h, the yield of the diol and other
products was already more than negligible. When the
reaction temperature increased to -30 °C, most of the
peroxy bond was already broken within 1 h.
LiAlH(OtBu)3 (6.7) Et2O/THF
a Because of its relatively low boiling point., it was not conve-
nient to calculate the yield of 8.
The results with LiAlH(OtBu)3 were more or less
similar to those observed in the reductions with LiBH4.
At the ambient temperature, the majority of the starting
ascaridol remained intact for several hours (entry 13),
which was long enough for complete reduction of the
pivalate. However, if the exposure time was longer than
10 h, the cleavage products also became significant (entry
14).
The profile of peroxy bond cleavage by LiBHEt3 was
examined in a similar way. This reagent was more
powerful than LiBH4, but less powerful than LiALH4.
After treatment with LiBHEt3 at 0 °C for 1.7 h, the
recovery yield of intact 3 was 75%. Raising the temper-
ature and extending the reaction time increased the
percentage of peroxy bond cleavage further as clearly
shown by the raised yields of the diol and other products.
Tolerance of a Teraoxane to Hydride Reduction. We
also briefly examined the hydride reduction compatibility
with tetraoxane 719 (Scheme 3, Table 3), a robust
compound readily prepared from cyclohexanone. Perhaps
because the peroxy bonds were somewhat more hindered
than those in compound 1 or ascaridole 3, 7 was rather
stable to LiAlH4. In sharp contrast to what was observed
with the other peroxy substrates, the survival rate for 7
was essentially the same at 0 and 28 °C (the ambient
temperature).
Similarly, stirring with LiBH4 at the ambient temper-
ature (19 °C) for a few hours did not led to any discernible
changes. Even after refluxing for 11 h, the starting 7 still
could be recovered in 85% yield. DIBAL-H, however,
cleaved 7 much more effectively. Only 42% of the starting
7 was recovered after treatment with DIBAL-H at 26 °C
for 3 h (entry 7). LiBHEt3 and LiAlH(OtBu)3 were all well-
tolerated as expected (entries 8-11).
Tolerance of the Peroxy Bond in a Plakoric Acid
Analogue to Hydride Reduction. The three experi-
ments (Table 4) with a plakoric acid analogue (Scheme
4) carried out under parallel conditions provided an even
closer comparison of the peroxy-bond cleavage power of
LiAlH4, LiBH4, and LiAlH(OtBu)3, these three promising
reductants.
The reactions (reduction of the ester group and the
cleavage of the peroxy bond) with LiAlH4 proceeded
remarkably faster than with the other two reagents. The
peroxy bond-cleavage products appeared on TLC within
minutes. Apparently, utilization of this reductant on
peroxy bond-containing substrates requires greater care
(19) (a) Sanderson, J. R.; Paul, K.; Story, P. R.; Denson, D. D.; Alford,
J. A. Synthesis 1975, 159-161. (b) Xu, W.-L.; Chen, Y.-F.; Zhang, S.-
L. Chem. Res. Chin. Univ. 1999, 15, 329-332.
4244 J. Org. Chem., Vol. 70, No. 11, 2005

Susceptibility of Organic Peroxy Bonds
TABLE 4. Comparison of LiAlH4, LiBH4, and
LiAlH(OtBu)3 in Reduction of 9 (cf Scheme 4)
SCHEME 4
reductant
temp. time
side
entry
(mol equiv)
solvent
THF
(°C) (min) 10 productsa
1
2
3
LiAlH4 (2.0)
LiBH4 (2.0)
0
0
0
13
39
81%
86%
18%
20%
20%
Et2O
LiAlH(OtBu)3 (6.7) Et2O/THF
52b 50%c
other substrates, demonstrating again the great potential
of this reductant.
a Weight percentage with respect to the substrate. b Then
stirred at 34 °C for 106 min before workup. c Along with ca. 30%
of recovered starting 9.
The reducing power of LiAlH(OtBu)3 was remarkably
lower than LiAlH4 or LiBH4 here. Even under the most
suitable conditions (using the reagent came as THF
solution and reacting in a mixture of Et2O-THF), the
reaction was very sluggish at 0 °C. Raising the temper-
ature to 34 °C after stirring at 0 °C for 52 min still failed
to drive the reaction to complete (ca. 30% of starting 9
was recovered). On the other hand, the peroxy bond
cleaved products were as much as observed with the
than using other reagents. To minimize over-reduction,
monitoring the progress of the reaction must be done soon
after the addition of all reactants, especially for those
substrates where the peroxy bond is not highly hindered.
LiBH4 was well-tolerated by the peroxy bond while
reducing the ester group effectively as observed with
a
TABLE 5. Reduction of Some Plakoric Acid Analogues with LiBH4
a All starting esters (except for 11, 17, and 21, which were single diastereomer isolated by column chromatography) employed in the
reduction were a mixture of the two diastereomers (almost inseparable on TLC). As the chiral centers were not touched during the reduction,
the products were presumably also mixtures of diastereomers. However, the diastereomeric mixtures served equally well as far as the
compatibility of the peroxy bond to hydride reductants was concerned. b Data taken from Table 1 for comparison. c Data taken from
Table 4 for comparison.
J. Org. Chem, Vol. 70, No. 11, 2005 4245

Jin et al.
TABLE 6. Reduction of PhCH2CO2Me with
NMR (taken on the crude product mixture after simple
workup).
LiAlH(OtBu)3 (6.7 mol equiv) under Different Conditions
entry reagenta
solventb
temperature/time
ratioc
Again, in the presence of Et2O the reduction (with the
commercially available LiAlH(OtBu)3 THF solution as the
reductant) of PhCH2CO2Me proceeded rather fast and
complete (Table 6, entries 1, 3, and 4). In the absence of
Et2O, only 76% of the starting ester was reduced under
the otherwise the same conditions (entry 2). Powdered
LiAlH(OtBu)3 was much less active. At the ambient
temperature for 4 h, the reduction was still rather
incomplete in either Et2O or THF (entries 5 and 6).
Prolonged reaction time did not help much if the reaction
medium was Et2O alone (entry 7). In THF (in which the
solubility of the reductant is significantly higher than in
Et2O), however, the yield could be raised to 75% after
14.5 h (entry 8). Finally, it is interesting to note that
reflux of the powdered reagent in THF before the
reduction and with the added Et2O still could not gave
the same reactivity as observed with the commercially
available LiAlH(OtBu)3 THF solution (entries 10-12).
1
2
A
A
A
A
B
B
B
B
B
B
B
B
3:2 Et2O/THF 20 °C/30 min
100:0
76:24
100:0
100:0
21:79
28:62
31:69
75:25
THF 20 °C/30 min
3
3:2 Et2O/THF 25 °C/15 min
1:1 Et2O/THF 20 °C/15 min
4
5
Et2O
THF
Et2O
THF
THF
20 °C/4 h
6
20 °C/4 h
7
19 °C/14.5 h
19 °C/14.5 h
20 °C/1 h then reflux 2 h 75:25
8
9
10d
11e
12f
2:1 Et2O/THF 20 °C/1 h then reflux 2 h 55:45
2:1 Et2O/THF 30 °C/1 h then reflux 2 h 69:31
5:4 Et2O/THF 28 °C/1 h then reflux 2 h 39:61
a A, LiAlH(OtBu)3 solution in THF; B, LiAlH(OtBu)3 powder.
b Final composition of the reaction medium after adding Et2O.c The
ratio of PhCH2CH2OH/PhCH2CO2Me in the product mixture as
estimated from the integrals in 1H NMR of the product mixture.
d The LiAlH(OtBu)3 powder was suspended in THF at the ambient
temperature. e The LiAlH(OtBu)3 powder was refluxed in THF
(still a suspension) for 0.5 h before cooling down and introducing
the Et2O. f The LiAlH(OtBu)3 powder was refluxed in THF for 2 h
(still a suspension) before cooling down and introducing the Et2O.
Conclusions
other two more powerful reductants. The results here are
in sharp contrast to the reduction of ethyl (1, cf. Table 1)
or methyl esters (cf. ref 7 and the results in the last
section of this paper), indicating that the ease of the
reduction with this reagent may depend significantly on
the alkyl group in the ester functionality.
We have conducted a systematic investigation on the
tolerance of organic peroxy bonds to some hydride reduc-
ing agents commonly utilized in organic synthesis. Using
reduction of ester group to the corresponding alcohol as
a reference reaction to define the end point of exposure
to the reducing agent, we measured the survival rate of
organic peroxy bonds after treatment with each indi-
vidual reducing agent selected. The results show that
LiAlH4, LiAlH(OtBu)3, LiBHEt3, and LiBH4 are signifi-
cantly better than other reductants examined when
taking into consideration both the efficiency of the ester
reduction and survival of the peroxy bond. Reduction with
LiAlH4 requires careful control (best carried out at 0 °C
for only a few minutes in most cases). Slight increase in
temperature or reaction time may lead to significantly
increased over-reductions (cleavage of the peroxy bond).
The powdered LiAlH(OtBu)3 could not reduce ester
functionality to any synthetically useful extents without
breaking the peroxy bond present in the substrate. The
commercially available LiAlH(OtBu)3 solution in THF is
a reagent powerful enough to reduce methyl or ethyl
esters while still weak enough to be tolerated by the
peroxy bonds. The presence of Et2O facilitates the reduc-
tion of esters with the LiAlH(OtBu)3 solution in THF,
although the solubility of the reagent is much higher in
THF than in ether. LiBHEt3 is also a potentially useful
reagent for the reduction of the peroxy-esters. LiBH4 is
much more attractive than all other reagents so far
tested. Apart from the excellent reducing power and
peroxy bond compatibility, the advantages in practical
aspects over the other reductants are also very impres-
sive. It is hoped that the results disclosed herein may
add a useful piece to the existing knowledge of organic
peroxides and provide a quick reference for dealing with
other peroxy bond-containing molecules when a hydride
reducing agent is involved.
Reduction of the Ester Functionality in Some
Plakoric Acid Analogues with LiBH4. The results
above show that among all the reductants tested, LiAlH4,
LiAlH(OtBu)3, LiBHEt3, and LiBH4 gave the best results.
Because LiBH4 was particularly attractive when taking
all aspects into consideration, we next examined this
reductant with a number of other peroxy substrates
(Table 5).
The yields are generally high. As observed in reduction
of other substrates, variation of the temperature or time
did not have much influence on the yield. Benzyl ester
was also reduced satisfactorily. Compared with LiAlH-
(OtBu)3 (which is only effective for methyl/ethyl esters
but not the benzyl esters even under the optimal condi-
tions), the application scope of LiBH4 is apparently
broader. Taking the reducing power, compatibility with
peroxy functionality, convenience of operations as well
as reagent handling into consideration, LiBH4 is evi-
dently superior to all the other reductants in the reduc-
tion of peroxy bond-containing substrates.
On the Reducing Power of LiAlH(OtBu)3. We have
mentioned in the first section that the reducing power
of LiAlH(OtBu)3 seemed to be dependent on the state/
form of the reagent. Addition of diethyl ether to the
reaction medium (THF) appeared to have an accelerating
effect on the reduction. As we were unable to find related
reports in the literature, it would be interest to see
whether this effect occurred only in the presence of a
peroxy bond or existed generally. Hence, we also briefly
tested LiAlH(OtBu)3 on an ordinary ester that did not
contain any peroxy functionality. For convenience of
monitoring the progress of the reduction, methyl phenyl-
acetate was chosen as the substrate. The reduction was
conducted under different conditions as shown in Table
6. The substrate/product ratio was then measured by 1H
Experimental Section
Typical Procedure 1 (Reduction of 1 with L-Select-
ride, Preparation of 2-(1,6,7-Trioxa-spiro[4.5]dec-8-yl)-
ethanol (2)). L-Selectride (1.0 M, 0.85 mL) was added (via a
4246 J. Org. Chem., Vol. 70, No. 11, 2005

Susceptibility of Organic Peroxy Bonds
syringe) to a solution of 1 (78 mg, 0.34 mmol) in dry THF (2.3
mL) stirred at -78 °C under N2. After stirring at the same
temperature for 2.75 h, MeOH (0.2 mL) was added to the
reaction mixture. The -78 °C bath was then replaced by an
ice-water bath. Aqueous NaOH (10%, 3 mL) was introduced
slowly, followed by H2O2 (30%, 2 mL). The stirring was
continued at the ambient temperature for another 4.5 h. The
phases were separated and the aqueous layer was back-
extracted with Et2O. The combined organic layers were washed
in turn with H2O, saturated NaHSO3 solution, brine, and dried
over anhydrous Na2SO4. The residue after removal of the
solvents was purified by column chromatography on silica gel
(2:1 Et2O/hexanes) to give alcohol 2 (42 mg, 65.7%) as a
colorless oil: 1H NMR (300 MHz, CDCl3) δ 4.35-4.26 (m, 1H),
4.12-4.01 (m, 2H), 3.82-3.72 (m, 2H), 2.05-1.66 (m, 11H);
IR (film): 3425, 2949, 1440, 1357, 1164, 1110, 1043 cm-1; ESI-
MS (m/z): 189 ([M + H]+); ESI-HRMS calcd. for C9H16O4Na
([C9H16O4Na]+) 211.0937, found 211.0941. Anal. Calcd. for
C9H16O4: C, 57.43; H, 8.57. Found: C, 57.44; H, 8.73.
Typical Procedure 2 (Reduction of 1 with Red-Al).
Red-Al (wt 65% in toluene, 0.14 mL) was added dropwise (via
a syringe) to a solution of the 1 (74 mg, 0.32 mmol) in dry
THF (3.2 mL) stirred at 0 °C under N2. The stirring was
continued at the ambient temperature for 2.3 h. With cooling
in ice-water bath aqueous NaOH (10%, 4 mL) was added
dropwise. The phases were separated and the aqueous phase
was back-extracted with Et2O. The combined organic layers
were washed with brine and dried over Na2SO4. Removal of
the solvent left an oily residue, which was purified by column
chromatography on silica gel to give 2 (29 mg, 48%).
Typical Procedure 3 (Reduction of 1 with DIBAL-H).
DIBAL-H (1.0 M solution in cyclohexane, 2.3 mL) was added
(via a syringe) to a solution of 1 (174 mg, 0.76 mmol) in dry
CH2Cl2 (3.8 mL) stirred at -78 °C under N2. The stirring was
continued at -78 °C for 3.6 h. Then the bath was allowed to
warm to the ambient temperature. MeOH (0.02 mL) was
added, followed by aqueous saturated potassium sodium
tartrate solution (5 mL) and Et2O (7 mL). The phases were
separated. The aqueous layer was back-extracted with Et2O.
The combined organic layers were filtered and washed with
copious amount of H2O and brine, dried over Na2SO4, and
concentrated on a rotary evaporator. The residue was chro-
matographed on silica gel to give 2 (75 mg, 53%).
rated. The aqueous layer was extracted with EtOAc. The
combined organic layers were washed with saturated NaHCO3,
water and brine, dried over Na2SO4, and concentrated on a
rotary evaporator. The residue was chromatographed on silica
gel to give 2 (36 mg, 83%).
Typical Procedure 6 (Reduction of 1 with LiBH4).
LiBH4 (13 mg) was added to a 1.0 M solution of 1 (66 mg, 0.287
mmol) in anhydrous Et2O stirred at the ambient temperature.
The mixture was stirred until TLC showed complete disap-
pearance of 1. The reaction was quenched with saturated
aqueous NH4Cl with cooling (ice-water bath). The phases were
separated and the aqueous layer was extracted with Et2O. The
combined organic layers were washed with brine, dried over
Na2SO4, and concentrated on a rotary evaporator. The residue
was chromatographed on silica gel to give 2 (49 mg, 91%).
Typical Procedure 7 (Reduction of 1 with LiBHEt3).
LiBHEt3 (1.0 M, 0.5 mL) was added (via a syringe) to a solution
of 1 (57 mg, 0.25 mmol) in dry THF (2.0 mL) stirred at 0 °C
under N2. The stirring was continued at the same temperature
until TLC showed complete disappearance of 1. The reaction
was quenched with saturated aqueous NH4Cl, followed by
aqueous HCl (2 N, two drops). The phases were separated and
the aqueous layer was extracted with Et2O. The combined
organic layers were washed with saturated NaHCO3, water
and brine, dried over Na2SO4, and concentrated on a rotary
evaporator. The residue was purified by column chromatog-
raphy on silica gel to give 2 (40 mg, 86%).
Reduction of Plakoric Acid Analogues 9′, 11, 13, 15,
17, 19, and 21 with LiBH4 (Preparation of 2-(6-Butyl-6-
methoxy-1,2-dioxinan-3-yl)-ethanol (10), 2-(Spiro[2,3-di-
oxo-benzocyclohexane-4,2′- tetrahydrofuran]-1-yl)-eth-
anol (12), 2-(Spiro[5-dihydro-2,3-dioxo-benzocyclohep-
tane-4,2′- tetrahydropyran]-1-yl)-ethanol (14), 2-(Spiro-
[5-dihydro-2,3-dioxo-benzocycloheptane-4,2′- (3′,6′-dihy-
dro)-benzotetrahydropyran]-1-yl)-ethanol (16), 2-(4-Hexyl-
2,3,7-trioxa- bicyclo[2.2.1]hept-1-yl)-ethanol (18), 2-(6-
Benzyl-6-methoxy-1,2-dioxinan-3-yl)-ethanol (20), and
2-(6-Methoxy-6-methyl-1,2-dioxinan-3-yl)-ethanol (22)).
The procedure was the same as that for the reduction of 1 with
LiBH4 described above (Typical Procedure 6). The reaction
temperature/time and the yield for each individual substrate
are listed in Table 5.
Acknowledgment. This work has been supported
by the National Natural Science Foundation of China
(20025207, 20272071, 20372075, 20321202), the Chinese
Academy of Sciences (“Knowledge Innovation” project,
KGCX2-SW-209), and the Major State Basic Research
Development Program (G2000077502).
Typical Procedure 4 (Reduction of 1 with LiAlH4).
With stirring and cooling (0 °C), LiAlH4 (20 mg) was added in
portions to a 1.0 M solution of 1 (59 mg, 0.26 mmol) in dry
THF (2.5 mL). The resulting mixture was stirred at the same
temperature until TLC showed complete disappearance of 1.
Saturated aqueous NH4Cl was added, followed by EtOAc.
Acidification of the mixture with aqueous HCl (2 N, two drops)
led to disappearance of the gray precipitates. The phases were
separated and the aqueous layer was extracted with EtOAc.
The combined organic layers were washed with saturated
aqueous NaHCO3, water and brine, dried over Na2SO4, and
concentrated on a rotary evaporator. The residue was chro-
matographed on silica gel to give 2 (43 mg, 89%).
Note Added after ASAP Publication. An oxygen
atom was missing in the structure of the starting
material in the abstract/toc graphic in the version posted
April 26, 2005; the corrected version posted April 28,
2005.
Typical Procedure 5 (Reduction of 1 with LiAlH-
(OtBu)3). A 30% solution of LiAlH(OtBu)3 in THF (1.3 g) was
added dropwise (via a syringe) to a solution of 1 (54 mg, 0.23
mmol) in dry Et2O (4.6 mL) stirred at the ambient tempera-
ture. When TLC showed complete disappearance of 1, the
reaction was quenched with H2O. The mixture was acidified
with aqueous HCl (2 N, two drops, which turned the reaction
mixture into a clear solution) before the phases were sepa-
Supporting Information Available: The general re-
marks for the Experimental Section, the data for all new
compounds not mentioned in the Experimental Section, reduc-
tion of 3, 7, and 9, 1H NMR of 9, 9′, 17, and 13C NMR spectra
of compounds 18 and 22. This material is available free of
charge via the Internet at http://pubs.acs.org.
JO050139Y
J. Org. Chem, Vol. 70, No. 11, 2005 4247