Chemical consequences of long-range orbital interactions in cholestane-3,7-diol monosulfonate esters. A seven-center fragmentation

Article abstract of DOI:10.1021/jo951655z

The manifestation of through-five-bond interactions in the reactions of the rigid 1,5-diol monosulfonate esters (1-3), having an ideal all-trans geometry of the σ-relay, with sodium tert-amylate is investigated. It has been shown that the deprotonation of the alcohol group in 2 and 3, which results in an increased electrofugal ability of this group, finds expression in a seven-center fragmentation. This fragmentation also illustrates that effective through-five-bond interactions exist between the alcoholate group and the carbocationic center which is generated during the heterolysis. The reaction outcome of mesylate 1 does not indicate effective through-bond interactions, and only elimination is observed. This difference can be attributed to the alkyl substituents on the γ- and α-positions to the mesylate group in 2 and 3, respectively, which stimulate the seven-center fragmentation. Though a reasonable amount of fragmentation product is obtained from 3, the through-five-bond interaction is not strong enough to dominate the reaction course completely and typical E1-like processes, i.e., elimination and rearrangement, are competitive. As expected, only a 1,2 Me-shift is observed in the reaction of the axial mesylate 4 where a gauche interaction is present in the geometry of the σ-relay. The presence of through-bond interactions in the reactions of 3 and 4 becomes apparent by comparison of the reactivity of 3 and 4 with their O-silylated analogs 5 and 6.

Full text of DOI:10.1021/jo951655z

J . Org. Chem. 1996, 61, 859-867
859
Ch em ica l Con sequ en ces of Lon g-Ra n ge Or bita l In ter a ction s in
Ch olesta n e-3,7-d iol Mon osu lfon a te Ester s. A Seven -Cen ter
F r a gm en ta tion
†
‡
†
Petrus M. F. M. Bastiaansen, Ladislav Kohout, Maarten A. Posthumus,
,
†
,†
J oannes B. P. A. Wijnberg,* and Aede de Groot*
Laboratory of Organic Chemistry, Agricultural University, Dreijenplein 8, 6703 HB Wageningen,
The Netherlands, and the Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences,
Flemingovo 2, 166 10 Praha 6, Czech Republic
Received September 8, 1995X
The manifestation of through-five-bond interactions in the reactions of the rigid 1,5-diol monosul-
fonate esters (1-3), having an ideal all-trans geometry of the σ-relay, with sodium tert-amylate is
investigated. It has been shown that the deprotonation of the alcohol group in 2 and 3, which
results in an increased electrofugal ability of this group, finds expression in a seven-center
fragmentation. This fragmentation also illustrates that effective through-five-bond interactions
exist between the alcoholate group and the carbocationic center which is generated during the
heterolysis. The reaction outcome of mesylate 1 does not indicate effective through-bond
interactions, and only elimination is observed. This difference can be attributed to the alkyl
substituents on the γ- and R-positions to the mesylate group in 2 and 3, respectively, which stimulate
the seven-center fragmentation. Though a reasonable amount of fragmentation product is obtained
from 3, the through-five-bond interaction is not strong enough to dominate the reaction course
completely and typical E1-like processes, i.e., elimination and rearrangement, are competitive. As
expected, only a 1,2 Me-shift is observed in the reaction of the axial mesylate 4 where a gauche
interaction is present in the geometry of the σ-relay. The presence of through-bond interactions in
the reactions of 3 and 4 becomes apparent by comparison of the reactivity of 3 and 4 with their
O-silylated analogs 5 and 6.
7
In tr od u ction
Paddon-Row et al. used the Birch reduction as a tool for
exploring the chemical consequences of long-range orbital
interactions through five and six σ-bonds and came to
the conclusion that any chemical consequences arising
The best-known example of participation of the σ-frame-
work in an intramolecular chemical reaction is the
heterolytic Grob fragmentation,1 which is almost cer-
tainly the result of orbital interactions through three
σ-bonds.2 The base-induced fragmentation of cyclic 1,3-
diol monosulfonate esters, known as the Wharton reac-
tion, is a typical example of the Grob fragmentation and
finds widespread application in organic synthesis.3
The synthetic value of reactions in which orbital
interactions through four σ-bonds are supposed to be
operative is best illustrated with the intramolecular base-
induced rearrangement and elimination reactions of
trans-fused perhydronaphthalene-1,4-diol monosulfonate
esters.4 More detailed studies revealed that when these
compounds possess an all-trans arrangement of the
8
from such interactions should be small. Grob et al.
examined the solvolysis of a number of 4-substituted (X)
bicyclo[2.2.2.]octylsulfonate esters (X ) COO , CONH ,
2 2 2
CH OH, and CH NH ) to find out whether these satu-
rated compounds are capable of undergoing seven-center
fragmentation in which five σ-bonds are involved. How-
ever, careful scrutiny of the reaction products revealed
that only substitution and no seven-center fragmentation
to 1,4-dimethylidenecyclohexane had occurred (Scheme
-
2
9
1
). Grob concluded that for the weakly electron-donating
,5
substituents X mentioned above double hyperconjugation,
as illustrated in structure I, does not progress to the point
where fragmentation occurs and an energetically more
favorable substitution reaction takes place.
More recently, Adcock et al.10 reported the fragmenta-
tion to 1,4-dimethylidenecyclohexane in the reaction with
the strongly electron-donating trimethylstannyl group as
σ-bonds homofragmentation is the characteristic reaction
and that a more substituted σ-framework
results in a higher reactivity of these compounds.
Reports on the chemical consequences of orbital inter-
actions through more than four σ-bonds are scarce.
pathway6
a-c
6
d
†
Agricultural University.
Czech Academy of Sciences.
Abstract published in Advance ACS Abstracts, J anuary 15, 1996.
‡
X
(
1) Grob, C. A. Angew. Chem. 1969, 81, 543.
(7) Paddon-Row, M. N.; Hartcher, R. J . Am. Chem. Soc. 1980, 102,
671.
(2) Gleiter, R.; Stohrer, W.-D.; Hoffmann, R. Helv. Chim. Acta 1972,
5
5, 893.
3) For an extensive review of the Wharton reaction, see: Caine, D.
Org. Prep. Proced. Int. 1988, 20, 1.
4) Wijnberg, J . B. P. A.; J enniskens, L. H. D.; Brunekreef, G. A.;
de Groot, A. J . Org. Chem. 1990, 55, 941.
5) J enniskens, L. H. D.; Wijnberg, J . B. P. A.; de Groot, A. J . Org.
Chem. 1991, 56, 6585.
(8) Grob, C. A.; Rich, R. Helv. Chim. Acta 1979, 62, 2793 and
references cited therein.
(
(9) Seven-center Beckmann-type fragmentations are known in the
literature: (a) Eisele, W.; Grob, C. A.; Renk, E.; von Tschammer, H.
Helv. Chim. Acta 1968, 51, 817. (b) Ibuka, T.; Mitsui, Y.; Hayashi, K.;
Minakata, H.; Inubushi, Y. Tetrahedron Lett. 1981, 22, 4425. Also,
seven-center fragmentations in which double bonds and/or highly
strained ring systems are involved have been reported. For example,
see: (c) Ho, T.-L. In Heterolytic Fragmentation of Organic Molecules;
Wiley-Interscience: New York, 1993; Chapter 9 and references cited
therein.
(
(
(6) (a) Orr uj , R. V. A.; Wijnberg, J . B. P. A.; J enniskens, L. H. D.; de
Groot, A. J . Org. Chem. 1993, 58, 1199. (b) Orr uj , R. V. A.; Wijnberg,
J . B. P. A.; Bouwman, C. T.; de Groot, A. J . Org. Chem. 1994, 59, 374.
(c) Bastiaansen, P. M. F. M.; Wijnberg, J . B. P. A.; de Groot, A. J . Org.
Chem. 1995, 60, 4240. (d) Orr uj , R. V. A.; Wijnberg, J . B. P. A.; de Groot,
A. J . Org. Chem. 1995, 60, 4233.
(10) Adcock, W.; Krstic, A. R.; Duggan, P. J .; Shiner, V. J ., J r.; Coope,
J .; Ensinger, M. W. J . Am. Chem. Soc. 1990, 112, 3140.
0
022-3263/96/1961-0859$12.00/0 © 1996 American Chemical Society

8
60 J . Org. Chem., Vol. 61, No. 3, 1996
Bastiaansen et al.
Sch em e 1
Sch em e 2
Ch a r t 1
composition and reactivity of compounds having a non-
zigzag σ-relay.
Additionally, the O-silylated mesylates 5 and 6 were
subjected to the same strongly basic conditions as well
to demonstrate again that the generation of an alcoholate
function is crucial for the reactivity of these compounds.
Herein we report the results of our study.
the substituent X. This and other related results11
supported Grob’s view that fragmentation can occur only
when the group X is a sufficiently active electrofuge.12
In order to test this hypothesis, we started an inves-
tigation into the reaction behavior of 1,5-diol monosul-
Resu lts a n d Discu ssion
The mesylate 1 was prepared from the known 3â-
1
8
hydroxy-5R-cholestan-7-one by mesylation and succes-
sive treatment with MeMgI at 0 °C.19 The synthesis of
mesylate 2 started with the readily available dimethy-
lated ketone 7.20 Protection of the carbonyl group of 7
as its ethylene acetal 8 was followed by an allylic
oxidation with PDC in refluxing pyridine21 to give the
enone 9 (Scheme 2). Treatment of 9 with Na in liquid
fonate esters under strongly basic nonsolvolytic condi-
tions.6 We hoped to demonstrate the seven-center
fragmentation Grob had in mind because deprotonation
of the hydroxyl group in these sulfonate esters would lead
to a strongly increased electrofugal ability of the carbinol
function.13 Furthermore, it is known that structural
1
4
features such as (1) an all-trans (zigzag) σ-relay, (2) a
tertiary alcohol function,15 and (3) a higher degree of
substitution of the σ-relay16 all may favor fragmentation
3
NH in the presence of t-BuOH (3 equiv) gave the 7â-
hydroxy compound 10 as the sole product. Hydrolysis
of the acetal function of 10 and subsequent treatment
with MeLi afforded the diol 11.22 Finally, the secondary
alcohol group of 11 was converted into a mesylate group
to afford compound 2 in 34% overall yield from 7.
For the synthesis of the mesylates 3 and 4, the double
reactions. We therefore decided to investigate, as 1,5-
diol monosulfonate esters, the easily accessible choles-
tane-3,7-diol monomesylates 1-417 (Chart 1). In par-
ticular, the mesylates 2 and 3 possess all the favorable
structural features for the occurrence of seven-center
fragmentation.
3
bond of 9 was reduced with Li in liquid NH at reflux
temperature (Scheme 3). The resulting ketone 12 was
treated with an excess of MeMgI and then with HCl to
afford predominantly the keto alcohol 13. The stereo-
chemistry of the hydroxyl group at C(7) in 13 was
confirmed by X-ray crystallography.23 After reduction of
13 with LAH and subsequent treatment with MsCl, the
mesylate 3 was obtained in 66% overall yield from 9.
Reduction of 13 with L-Selectride (Aldrich) and mesyla-
tion of the resulting diol gave the mesylate 4 in 61%
overall yield from 9. Treatment of 3 and 4 with TMSCl
There were two other reasons for selecting these
particular cholestane derivatives for study. First, from
the reactions of the mesylates 1-3, having different alkyl
substitution patterns, it was expected that further evi-
dence for the accelerating effect of alkyl substituents in
these processes would be obtained.6d Second, from the
reaction of the axial mesylate 4, we expected to get
information on whether through-bond interactions (TBI)
over five σ-bonds would still be significant for the product
(
11) (a) Adcock, W.; Coope, J .; Shiner, V. J ., J r.; Trout, N. A. J . Org.
Chem. 1990, 55, 1411. (b) Lambert, J . B.; Salvador, L. A.; So, J .-H.
Organometallics 1993, 12, 697.
(18) Davies, A. R.; Summers, G. H. R. J . Chem. Soc. C 1967, 1227.
(19) The stereochemistry of the hydroxyl group at C(7) could not be
established by NMR measurements. Since in the Grignard reaction of
12 the favored attack of the reagent is from the â-side (see Scheme 3),
a similar preferential approach is assumed to occur in the Grignard
reaction leading to 1.
(12) An electrofugal group can act as an electron donor in heterolytic
fragmentation and splits off without the bonding electron pair: Grob,
C. A. Angew. Chem., Int. Ed. Engl. 1969, 8, 535.
(
active electrofuge in both the Wharton fragmentation and the base-
induced homofragmentation of 1,4-diol monosulfonate esters in apolar
13) It has been demonstrated that an alcoholate group is a very
3
(20) Woodward, R. B.; Patchett, A. A.; Barton, D. H. R.; Ives, D. A.
J .; Kelly, R. B. J . Chem. Soc. 1957, 1131.
6
a-c
solvents like benzene or toluene.
14) This structural feature guarantees that (a) the extent of orbital
(21) Parish, E. J .; Wei, T.-Y. Synth. Commun. 1987, 17, 1227.
(22) The stereochemistry of the hydroxyl group at C(3) could not be
established by NMR measurements. Since the nucleophilic addition
of MeLi to cholestan-2-one and 4a-methyl-trans-2-decalone at -78 °C
proceeds preferentially from the equatorial â-side, the stereochemistry
at C(3) was tentatively assigned as given in structure 11. See:
Macdonald, T. L.; Clark Still, W. J . Am. Chem. Soc. 1975, 97, 5280
and references cited therein. In this context, it is important to note
that the reactivity of 1,4-diol monosulfonate esters upon treatment with
sodium tert-amylate in refluxing benzene is not affected by the
(
interactions through σ-bonds is maximized and (b) the five σ-bonds
involved are properly aligned for fragmentation. See: (a) J ordan, K.
D.; Paddon-Row, M. N. Chem. Rev. 1992, 92, 395. (b) Reference 11b.
(15) In general, fragmentations that lead to ketones occur more
3
readily than those producing aldehydes.
16) Fragmentation products containing more highly substituted
(
olefins are more easily formed than those containing less highly
substituted olefins. See ref 3 and ref 9c, p 7.
6
b
(17) The numbering system for the cholestane derivatives used in
orientation of the hydroxyl group.
(23) Kooijman, H.; Spek, A. L. Manuscript in preparation.
this study follows the original steroid numbering.

Seven-Center Fragmentation
Sch em e 3
J . Org. Chem., Vol. 61, No. 3, 1996 861
Sch em e 5
Ta ble 1. Rea ction s of th e Mesyla tes 1-6 w ith Sod iu m
a
ter t-Am yla te
entry mesylate
products (%)b
recoveryc
d
d
1
2
3
4
5
6
1
2
3
4
5
6
14 (4) + 15 (3)
75
51
17
2
66
58
d
d
16 (3) + 17 (6) + 18 (32)
Sch em e 6
19 (20) + 20 (13) + 21 (22) + 22 (6)
d
d
23 (6) + 24 (48) + 25 (21)
e
- (17)
d
d
26 (13) + 27 (14)
a
All reactions were performed in refluxing toluene with ca. 5
b
equiv of sodium tert-amylate for 10 min. Yield in parentheses.
Percentage of recovered starting material. The yield of this
product is based on NMR analysis. Inseparable mixture of five
probably olefinic compounds.
c
d
e
Sch em e 4
and HMDS afforded the corresponding O-silylated
mesylates 5 and 6, respectively, in high yields.
In order to obtain comparable data about the reactivity
of the mesylates 1-6, all reactions were run in refluxing
toluene24 with ca. 5 equiv of sodium tert-amylate for 10
min. Comparison of the quantities of recovered starting
material gave a rough estimate of the relative reaction
rates.25 The results of these studies are collected in Table
1
.
material, and afforded a small quantity (3%) of the
fragmentation product 1628 together with an inseparable
ca. 1:5 mixture (38%) of the olefins 17 and 18, respec-
tively (entry 2, Scheme 5). The NMR spectral data of
The reaction of the mesylate 126 gave a small amount
7%) of an inseparable ca. 3:2 mixture of the olefins 14
(
and 15, respectively, together with 75% of recovered
starting material (entry 1, Scheme 4). In the H NMR
spectrum of the mixture, the olefinic signals of the major
1
1
5
6 are fully consistent with the structure given in Scheme
. Although the reaction pathway to 16 is only a minor
olefin 14 appear as a broad singlet (W1/2 ≈ 5 Hz) at δ
side reaction, it is, to the best of our knowledge, the first
example of a seven-center fragmentation over five single
bonds in unstrained compounds.
5
1
5
.59, while the corresponding signals of the minor olefin
5 resonate at δ 5.20 (br dd, J ) 2.2, 10.0 Hz, 1 H) and
.56 (m, 1 H). Similar differences in splitting pattern
The mesylate 3 reacted considerably faster than the
mesylates 1 and 2 (17% of 3 was regained) (entry 3,
Scheme 6). Four compounds were formed in this reac-
and chemical shift have been reported for the signals of
the olefinic protons of 5R-cholest-2-ene and 5R-cholest-
3
-ene.27
2
8
tion: the seven-center fragmentation product 19 (20%)
The mesylate 2 reacted slightly faster than 1, as follows
and the olefins 20, 21, and 22 in 13, 22, and 6% yields,
respectively. Also, in this case the NMR spectral data
of the seven-center fragmentation product 19 are fully
consistent with the assigned structure.
The mesylate 4 reacted the fastest of all the compounds
studied. After the standard treatment, only a small
amount of starting material (2%) could be regained (entry
from the lesser amount (51%) of recovered starting
(24) Toluene as solvent was preferred over benzene because it was
expected that the reactivity of the mesylates studied would be
moderate.
(25) The percentages of recovered starting material were somewhat
lowered (up to 7%) by competing S-O bond scission.
(26) Treatment of (3â,5R,7R)-cholestane-3,7-diol 3-methanesulfonate
with sodium tert-amylate in refluxing toluene afforded a complex
mixture of products, probably as a result of alkoxide-induced inter-
6
c
molecular hydride shifts.
27) Cragg, G. M. L.; Davey, C. W.; Hall, D. N.; Meakins, G. D.;
Richards, E. E.; Whateley, T. L. J . Chem. Soc. C 1966, 1266.
(28) It might be possible that the yield of this product was somewhat
diminished due to aldol condensations under the influence of sodium
tert-amylate.
(
6
a

8
62 J . Org. Chem., Vol. 61, No. 3, 1996
Bastiaansen et al.
Sch em e 7
stabilization of 2(A) can be attributed to enhanced
σ-participation6
d,29
as result of the presence of the C(8)-
C(14) bond next to the carbocationic center on C(7). It
might be possible that the electron-donating Me substit-
uents at C(4) also participate in the stabilization of 2(A)
by enlargement of the electron density of the σ-relay.
Combined with the enhanced stability of the carbocation,
this inductive donation will favor the reaction pathway
leading to fragmentation via the double hyperconjuga-
3
0
tively stabilized intermediate (or transition state) 2(B).
The formation of a small amount of fragmentation
product 16 in the reaction of 2 can be attributed to this
effect. Elimination leading to 17 and 18, however,
remains the main reaction pathway as the elimination:
fragmentation ratio (ca. 13:1 in favor of elimination)
indicates.
Sch em e 8
The difference in reaction rate found for the mesylates
2
and 3 can be explained similarly (Scheme 6). Stabili-
zation by σ-participation will be more effective in 3(A)
than in 2(A) because the C(4) atom adjacent to the
carbocationic center in 3(A) bears two alkyl substituents
(Me groups), whereas in 2(A) only one alkyl substituent,
i.e., the C(8)-C(14) bond, is present next to the carboca-
tionic center. In addition, cationic carbon atoms are
better stabilized by alkyl substituents located at the
R-position (as in 3(A)) than by alkyl substituents at the
γ-position (as in 2(A)).31 Consequently, TBI will be more
effective in 3(A) than in 2(A), which finds expression in
the reactivity order 3 > 2 and the relatively high yield
(20%) of the seven-center fragmentation product 19.
The other compounds (20, 21, and 22) found in the
reaction of 3 must be formed via rearrangement of the
initially formed intermediate 3(A) to the thermodynami-
cally more stable intermediate 3(B). This tertiary car-
bocationic intermediate can undergo proton loss to give
the olefins 20 and 21 but can also react further by way
of a 1,2-H shift (C(3) f C(4)) to afford another tertiary
intermediate (3(C)), which in turn can give additional
21 and the olefin 22, the latter probably as the result of
an intramolecularly assisted deprotonation.32 The for-
mation of 22 can only proceed stepwise, because a
concerted mechanism starting from 3(B) is not allowed
for stereochemical reasons (both H-3 and H-5 have the
R-orientation). It might be possible that 3(C) is some-
what better stabilized than 3(B), because in 3(C) the
number of σ-bonds between the alcoholate function and
the carbocationic center is reduced by one.33
4
2
, Scheme 7). As products, the three olefins 23, 24, and
5 in a ratio of 1:8:3.5, respectively, were obtained.
Column chromatography gave pure 24 (48%) and an
inseparable mixture (27%) of 23 and 25.
Compared with the mesylates 3 and 4, the correspond-
ing O-silylated mesylates 5 and 6 reacted more slowly
(entries 5 and 6). The quantities of recovered starting
material (66 and 58%, respectively) suggest that 6 reacts
slightly faster than 5. On the other hand, the product
compositions are noticeably different. Whereas 5 pro-
duced a complex mixture (ca. 17%) of at least five olefinic
compounds, 6 only gave two products: the ∆2 olefin 26
,3
(13%) and the rearranged olefin 27 (14%) (Scheme 8).
The results collected in Table 1 clearly show that Me
substitution at C(4) leads to an increase in reaction rate
entries 1-3). The occurrence of different product-
(
forming pathways in the reactions of 1-3 can be con-
nected with the presence (or absence) of Me substituents
as well. Furthermore, it is evident that alkoxide forma-
tion is also important for the reactivity of these com-
pounds as can be concluded from the differences in
reaction rate between 3 and 5 and between 4 and 6
(entries 3-6). These observations have led to the as-
sumption that the mesylates 1-4 react by a mechanism
which is comparable with that proposed for the 1,4-diol
monosulfonate esters.6 This means that 1-4 will react
via dipolar intermediates with intramolecular alkoxide-
induced heterolysis of the mesylate bond as the rate-
determining step. The dipolar intermediates in turn may
undergo typical cationic reactions such as elimination,
rearrangement, 1,2-Me shift, and/or fragmentation. The
product compositions and reaction rates found for 1-4
are easily understood on this basis.
The main reason for examining mesylate 4 was to find
out to what extent an axially positioned leaving group
influences the rate and course of these TBI-induced
reactions. Seven-center fragmentation was not expected
in this reaction because the required antiperiplanar
relationship between the leaving group and the C(4)-
C(5) bond is lacking.1 Instead, it was expected that a
1,2-Me shift would be the main reaction pathway since
both the leaving mesylate group and the â-Me group at
C(4) are axially positioned. The formation of the rear-
1b
The equatorial mesylate 1, which is supposed to react
via the intermediate 1(A), undergoes a 1,2-elimination
as the sole product-forming pathway (Scheme 4). The
relatively slow formation of the olefins 14 and 15 suggests
that intermediate 1(A) is not very effectively stabilized
by TBI.
The enhanced reactivity of mesylate 2, compared with
that of 1, indicates that in intermediate 2(A) the car-
bocationic center on C(7) is better stabilized than the one
on C(3) in 1(A) (Scheme 5). The higher degree of
(29) (a) J ensen, F. R.; Smart, B. E. J . Am. Chem. Soc. 1969, 91,
686. (b) Traylor, T. G.; Hanstein, W.; Berwin, H. J .; Clinton, N. A.;
5
Brown, R. S. J . Am. Chem. Soc. 1971, 93, 5715.
(30) Enhanced TBI should lead to increased fragmentation.10
(
31) Bielmann, R.; Grob, C. A.; K u¨ ry, D.; Wei Yao, G. Helv. Chim.
Acta 1985, 68, 2158.
32) Since intramolecular alkoxide-assisted elimination is a highly
(
6
a
favorable process, the formation of 21 from intermediate 3(C) via
an intermolecular proton abstraction will only be a minor reaction
pathway.
(33) Paddon-Row, M. N.; Patney, H. K.; Brown, R. S.; Houk, K. N.
J . Am. Chem. Soc. 1981, 103, 5575.

Seven-Center Fragmentation
Sch em e 9
J . Org. Chem., Vol. 61, No. 3, 1996 863
TBI over five σ-bonds. However, the reaction course is
not solely determined by these interactions and typical
E1-like processes come into play, this in contrast to the
6
previously studied 1,4-diol monosulfonate esters where
TBI over four σ-bonds is considered to be the main factor
determining the reaction behavior of these compounds.
Nevertheless, the involvement of through-five-bond in-
teractions is reflected by the occurrence of seven-center
fragmentation in the reactions of 1,5-diol monosulfonate
esters in which the five σ-bonds are held in the all-trans
geometry. The presence of alkyl substituents, especially
on the R-position, which results in enhanced σ-participa-
tion is also a prerequisite for allowing seven-center
fragmentation. The occurrence of seven-center fragmen-
tation also supports Grob’s view that the enhanced
solvolysis rates of the 4-substituted bicyclo[2.2.2]octyl-
ranged olefins 24 and 25 in which the intermediates
(A) and 4(B) should be involved corresponds with this
4
expectation (Scheme 7). Again, intramolecular deproto-
nation (48% of 24) is preferred over intermolecular
deprotonation (21% of 25). An intermolecular antielimi-
nation explains the formation of olefin 23 from 4(A).
Finally, it is of interest to note that 4 reacts somewhat
faster than 3. According to the trans rule, one should
expect the reverse order because of the all-trans arrange-
ment of the σ-relay present in 3.14a In order to explain
this discrepancy, the results of the reactions of 3 and 4
are compared with those of the O-silylated mesylates 5
and 6. These silyl ethers show a similar reactivity order
as that found for 3 and 4, i.e., compound 6 with an axial
mesylate group reacts slightly faster than 5 in which an
equatorial mesylate group is present. It is known that
the tosylate esters of 4,4-dimethylcholestan-3R-ol and the
sulfonate esters mentioned in the Introduction may be
8,31
due to double hyperconjugation.
The synthetic value
of this fragmentation reaction is limited because other
more favorable reaction pathways, i.e., elimination and
rearrangement, are preferred.
Though the occurrence of long-range orbital interac-
tions through five σ-bonds still enhances the reaction rate
of 1,5-diol monosulfonate esters in comparison with
systems that lack these interactions, a dramatic decrease
in reaction rate is found with respect to the 1,4-diol
monosulfonate esters. In our opinion, the ability of an
alcoholate function to induce intramolecularly the het-
erolysis of a remote sulfonate ester group has reached
its limit with the 1,5-diol monosulfonate esters.
3
â-epimer both react by an E1-like mechanism upon
treatment with sodium tert-amylate in refluxing ben-
zene.34 Moreover, the 3R-epimer reacts two to three
3
5
times faster than the 3â-epimer, which can be at-
tributed to relief of steric strain. A similar E1-like
mechanism may also be operative in the closely related
mesylates 5 and 6, which are supposed to react without
involvement of long-range orbital interactions.6 In
contrast to 5 and 6, the structurally related O-silylated
trans-fused perhydronaphthalene-1,4-diol mesylates (equa-
torial and axial) lacking the Me groups at C(4) do not
react at all under similar reaction conditions.6a Thus,
the relatively high reactivity of 5 and 6 and their product
composition can be associated with the steric and/or
electronic effects of the Me groups at C(4). On the basis
of these facts, it is assumed that the mesylates 3 and 4
also react via an E1-like mechanism but now with
participation of TBI since both 3 and 4 react faster than
their respective O-silylated analogs 5 and 6. The influ-
ence of Me groups at C(4) will predominate over TBI in
a
Exp er im en ta l Section 36
Ma ter ia ls. All reagents were purchased from Aldrich or
J anssen and were used without further purification unless
otherwise stated. A stock solution of sodium tert-amylate (3.2
3
7
M in toluene) was prepared by the procedure of Conia and
stored under an Ar atmosphere in a refrigerator. (3â,5R)-3-
1
8
Hydroxycholestan-7-one and 4,4-dimethylcholest-5-en-3-one
2
0
(
7) were prepared following previously described procedures.
(
3â,5r)-3-Hyd r oxych olesta n -7-on e Meth a n esu lfon a te.
To a stirred solution of 2.80 g (6.95 mmol) of (3â,5R)-3-
hydroxycholestan-7-one in 2.8 mL of dry pyridine was added
3
and 4 because of the large distance between the
2
.60 g (22.7 mmol) of MsCl. The reaction mixture was stirred
3
3
alcoholate group and the mesylate ester. On the other
hand, it will be clear that the additional stabilization of
the carbocationic intermediates by TBI is somewhat more
pronounced in 3(A) than in 4(A) (trans rule), which finds
expression in the formation of the seven-center fragmen-
tation product 19 in the reaction of 3.
at rt for 18 h and then poured into H O. The aqueous layer
2
was extracted with CHCl
HCl, water, 10% KHCO , and brine. After the product was
dried and the solvent was evaporated, crystallization from
CHCl
3
and washed successively with 5%
3
3
/ether afforded 1.60 g (45%) of pure (3â,5R)-3-hydroxy-
1
cholestan-7-one methanesulfonate: mp 167-168 °C; H NMR
δ 0.62 (s, 3 H), 0.75-2.43 (m, 29 H), 0.83 (d, J ) 6.4 Hz, 6 H),
The results of this study provide a good explanation
for the nonoccurrence of fragmentation products in the
solvolysis of the ꢀ-amino tosylate studied by Grob et al.
0
.88 (d, J ) 6.5 Hz, 3 H), 1.08 (s, 3 H), 2.98 (s, 3 H), 4.56 (m,
1 H); 13C NMR δ 11.65 (q), 12.02 (q), 18.74 (q), 21.79 (t), 22.53
(q), 22.78 (q), 23.73 (t), 24.91 (t), 27.94 (d), 28.34 (2t), 34.84
(t), 35.60 (d), 35.65 (s), 35.72 (t), 36.08 (t), 38.58 (t), 38.72 (q),
(
Scheme 9).8 In spite of the proper alignment of the
3
9.41 (t), 42.44 (s), 45.59 (t), 46.37 (d), 48.79 (d), 49.86 (d), 54.70
σ-bonds, the absence of two alkyl substituents next to
the carbon atom bearing the tosylate group and the
presence of a weakly electron-donating alkylated amino
group in this trans-perhydroisoquinoline system make
seven-center fragmentation an unfavorable process, and
only substitution will take place as Grob et al. found
experimentally.
(
4
2
d), 54.95 (d), 80.61 (d), 211.04 (s); MS m/ z (relative intensity)
+
80 (M , 22), 385 (10), 384 (21), 358 (30), 357 (100), 356 (17),
+
72 (11), 253 (9); HRMS calcd for C28
H
H
48
O
4
S (M ) 480.3273,
S: C, 69.95; H,
found 480.3273. Anal. Calcd for C28
0.06. Found: C, 70.02; H, 10.21.
(3â,5r,7r)-7-Meth ylch olesta n e-3,7-d iol 3-Meth a n esu l-
fon a te (1). To a stirred solution of 0.230 g (0.48 mmol) of
48
O
4
1
Con clu d in g Rem a r k s
(36) For
employed in this research, see ref 5. All NMR spectra were taken in
CDCl unless noted otherwise. Column chromatography was performed
a general description of the experimental procedures
These results clearly indicate that an alkoxide function
can assist in the formation of a carbocationic center via
3
using Merck Silica Gel 60 (70-230 mesh). GC analysis was carried out
with FID and a DB-5MS fused silica column, 15 m × 0.25 mm i.d.,
2
film thickness 0.10 µm, and H as the carrier gas. For GC-MS analysis
(
(
34) Levisalles, J .; P e` te, J .-P. Bull. Soc. Chim. Fr. 1967, 3747.
35) Arad, A.; Levisalles, J . Bull. Soc. Chim. Fr. 1972, 1135.
the same DB-5MS column was used.
(37) Conia, M. J .-M. Bull. Soc. Chim. Fr. 1950, 17, 537.

8
64 J . Org. Chem., Vol. 61, No. 3, 1996
Bastiaansen et al.
(
3â,5R)-3-hydroxycholestan-7-one methanesulfonate in 50 mL
five 100 mL portions of ether. The combined organic layers
were dried and evaporated, and the resulting product was flash
chromatographed (10:1 petroleum ether (bp 40-60 °C)/EtOAc)
of dry ether was added 0.48 mL (0.96 mmol) of MeMgI (2.0 M
in ether) at once at 0 °C. The reaction mixture was stirred
for 1 min and then carefully quenched with a small amount
of saturated aqueous NH
H
layer was washed with 10 mL of brine, dried, and evaporated.
The remaining residue was flash chromatographed (3:1 pe-
troleum ether (bp 40-60 °C)/EtOAc) to give 0.080 g of
1
to give 1.75 g (80%) of 10: H NMR δ 0.60-2.03 (m, 28 H),
4
Cl. After the addition of 15 mL of
0.63 (s, 3 H), 0.80 (s, 3 H), 0.81 (s, 3 H), 0.84 (s, 3 H), 0.87 (d,
J ) 7.1 Hz, 3 H), 0.91 (d, J ) 6.7 Hz, 6 H), 3.32 (m, W1/2 ) 15
Hz, 1 H), 3.80-4.02 (m, 4 H); 13C NMR δ 12.05 (q), 14.29 (q),
18.74 (q), 19.90 (q), 20.77 (t), 22.49 (q), 22.76 (2q), 23.81 (t),
26.90 (t), 27.02 (t), 27.93 (d), 28.68 (t), 31.39 (t), 35.62 (d), 35.64
(s), 35.72 (t), 36.15 (t), 39.45 (t), 39.81 (t), 41.88 (s), 43.02 (d),
43.14 (s), 49.68 (d), 53.84 (d), 55.14 (d), 55.76 (d), 64.72 (t),
64.83 (t), 76.14 (d), 113.08 (s); MS m/ z (relative intensity) 474
2
O, the two-phase mixture was separated, and the organic
1
unreacted starting material and 0.101 g (42%) of 1: H NMR
δ 0.66 (s, 3 H), 0.77 (s, 3 H), 0.80-2.05 (m, 30 H), 0.85 (d, J )
6
.3 Hz, 6 H); 0.83 (d, J ) 6.5 Hz, 3 H), 1.23 (s, 3 H); 2.97 (s, 3
13
+
H), 4.60 (m, 1 H); C NMR δ 11.42 (q), 12.24 (q), 18.81 (q),
1.60 (t), 22.54 (q), 22.78 (q), 23.72 (t), 27.96 (d), 27.96 (t), 28.52
t), 28.76 (t), 32.08 (q), 34.46 (t), 34.86 (s), 35.64 (d), 36.07 (t),
6.95 (t), 38.40 (d), 38.66 (q), 39.46 (t), 39.76 (t), 44.01 (s), 44.28
(M , 2), 431 (2), 109 (1), 100 (9), 99 (100), 95 (2), 81 (2), 69 (2),
+
2
(
3
55 (2), 43 (1); HRMS calcd for C31
H
54
54
O
3
(M ) 474.4073, found
C, 78.42; H, 11.47.
474.4070. Anal. Calcd for C31
Found: C, 78.60; H, 11.68.
H
O
3
:
(
d), 45.04 (t), 49.47 (d), 51.34 (d), 54.98 (d), 72.32 (s), 82.00
(3r,5r,7â)-3,4,4-Tr im eth ylch olesta n e-3,7-d iol (11). To
a stirred solution of 0.867 g (1.84 mmol) of 10 in 200 mL of
acetone was added 4 mL of 4 N HCl. The reaction mixture
was stirred at rt for 3 h and then neutralized with 10 mL of
saturated aqueous NaHCO3. After concentration under re-
duced pressure, the remaining aqueous solution was extracted
+
(d); MS m/ z (relative intensity) 481 (M - 15, 18), 385 (19),
3
5
4
83 (19), 367 (17), 332 (30), 256 (22), 97 (67), 83 (88), 69 (100),
+
5 (78); HRMS calcd for C28
81.3353. Anal. Calcd for C29
H
49
O
4
S (M - 15) 481.3352, found
52 4
H O
S: C, 70.11; H, 10.55.
Found: C, 70.30; H, 10.75.
4
′,4′-Dim eth ylspir o[1,3-dioxolan e-2,3′-ch olest-5′-en e] (8).
2 2
with 50 mL of CH Cl . The organic layer was washed with
A solution of 8.00 g (19.4 mmol) of 7, 0.347 g of camphorsul-
fonic acid, and 11 mL of ethylene glycol in 300 mL of toluene
was refluxed in the flask equipped with a Dean-Stark column
packed with 4 Å molecular sieves for 16 h, cooled, and poured
15 mL of brine, dried, and evaporated. The so-obtained crude
(5R,7â)-7-hydroxy-4,4-dimethylcholestan-3-one (0.777 g, 99%)
was dissolved in 30 mL of THF and cooled to -78 °C. After
addition of 5.7 mL of 1.6 M MeLi, the reaction mixture was
stirred at -78 °C for 10 min and allowed to warm to rt over a
2 h period. The excess of MeLi was then cautiously destroyed
into 100 mL of saturated aqueous NaHCO
mixture was separated, and the aqueous layer was extracted
with three 100 mL portions of CH Cl . The combined organic
3
. The two-phase
2
2
4
with a small amount of saturated aqueous NH Cl. After
layers were dried and evaporated. The remaining residue was
dilution with 25 mL of ether, the organic layer was washed
with 10 mL of brine, dried, and evaporated. The remaining
flash chromatographed (25:1 petroleum ether (bp 40-60 °C)/
1
38
EtOAc) to yield 8.14 g (92%) of 8: H NMR δ 0.66 (s, 3 H),
residue was flash chromatographed (5:1 petroleum ether (bp
1
0
.85 (d, J ) 6.7 Hz, 6 H), 0.90-2.18 (m, 26 H), 0.91 (d, J ) 6.5
40-60 °C)/EtOAc) to give 0.469 g (57%) of pure 11: H NMR
Hz, 3 H), 1.03 (s, 3 H), 1.12 (s, 3 H), 1.22 (s, 3 H), 3.87-4.03
δ 0.64 (s, 3 H), 0.65 (m, 1 H), 0.72-1.92 (m, 27 H), 0.83 (s, 3
H), 0.84 (d, J ) 6.1 Hz, 6 H), 0.88 (s, 3 H), 0.89 (s, 3 H), 0.89
(d, J ) 6.3 Hz, 3 H), 1.17 (s, 3 H), 1.98 (m, 1 H), 3.34 (ddd, J
) 5.4, 10.4, 10.4 Hz, 1 H); 13C NMR δ 12.06 (q), 14.67 (q), 18.75
(q), 18.82 (q), 20.74 (t), 22.50 (q), 22.75 (q), 23.07 (q), 23.83 (t),
24.36 (q), 26.88 (t), 27.95 (d), 28.69 (t), 31.84 (t), 34.20 (t), 35.63
(d), 36.15 (t), 36.30 (s), 36.45 (t), 39.44 (t), 39.79 (t), 40.71 (s),
43.02 (d), 43.44 (s), 49.63 (d), 54.52 (d), 55.20 (d), 55.76 (d),
1
3
(
(
m, 4 H), 5.52 (m, 1 H); C NMR δ 11.83 (s), 18.66 (q), 20.53
t), 21.66 (q), 22.36 (q), 22.53 (q), 22.79 (q), 23.79 (t), 24.16 (t),
2
6.81 (t), 27.98 (d), 28.25 (t), 29.81 (q), 30.87 (d), 32.31 (t), 35.24
(t), 35.76 (d), 36.16 (t), 36.24 (q), 39.49 (t), 39.71 (t), 42.19 (s),
4
1
4
9
4.79 (s), 50.55 (d), 55.95 (d), 57.17 (d), 64.84 (t), 65.20 (t),
13.19 (s), 119.90 (d), 149.64 (s); MS m/ z (relative intensity)
+
56 (M , 3), 412 (2), 356 (1), 124 (2), 107 (1), 100 (5), 99 (100),
+
+
5 (1), 87 (1), 55 (2); HRMS calcd for C31
H
52
O
2
(M ) 456.3967,
75.05 (s), 76.33 (d); MS m/ z (relative intensity) 446 (M , 2),
found 456.3967. Anal. Calcd for C31
Found: C, 81.69; H, 11.71.
H
52
O
2
: C, 81.52; H, 11.48.
428 (5), 375 (18), 374 (28), 358 (38), 357 (100), 149 (19), 123
+
(13), 122 (10), 95 (12); HRMS calcd for C30
H
54
O
2
(M ) 446.4138,
4
′,4′-Dim eth ylsp ir o[1,3-d ioxola n e-2,3′-ch olest-5′-en -7′-
on e] (9). To a stirred solution of 8.14 g (17.9 mmol) of 8 and
.00 g of 3 Å molecular sieves in 400 mL of dry pyridine was
found 446.4127. Anal. Calcd for C30
Found: C, 80.65; H, 11.93.
54 2
H O : C, 80.65; H, 12.18.
1
(3r,5r,7â)-3,4,4-Tr im eth ylch olesta n e-3,7-d iol 7-Meth -
a n esu lfon a te (2). To a solution of 0.439 g (0.98 mmol) of 11
in 20 mL of dry pyridine was added 0.200 g (1.75 mmol) of
MsCl. The reaction mixture was stirred at rt overnight and
then concentrated under reduced pressure. The resulting
residue was taken up in 50 mL of EtOAc and washed
successively with two 15 mL portions of saturated aqueous
NaHCO and one 15 mL portion of brine. The organic layer
3
was dried and evaporated, and the remaining residue was
flash chromatographed (5:1 petroleum ether (bp 40-60 °C)/
EtOAc) to give 0.487 g (94%) of 2: H NMR δ 0.63 (s, 3 H),
added 80.0 g of PDC. The reaction mixture was heated at
reflux for 2 h, cooled, and then poured into 750 mL of brine.
The aqueous layer was extracted with six 250 mL portions of
CH Cl , and the combined organic layers were dried and
2 2
evaporated. The remaining residue was flash chromato-
graphed (25:1 petroleum ether (bp 40-60 °C)/EtOAc) to yield
1
7
6
3
.29 g (87%) of 9: H NMR δ 0.67 (s, 3 H), 0.84 (d, J ) 6.7 Hz,
H), 0.90 (d, J ) 6.5 Hz, 3 H), 0.90-2.40 (m, 24 H), 1.06 (s,
H), 1.31 (s, 3 H), 1.32 (s, 3 H), 3.87-4.03 (m, 4 H), 5.88 (s, 1
1
3
1
H); C NMR δ 11.90 (q), 18.80 (q), 20.13 (q), 20.81 (t), 21.50
q), 22.51 (q), 22.79 (q), 23.81 (t), 26.39 (t), 26.56 (t), 27.96 (d),
8.32 (q), 28.54 (t), 34.18 (t), 35.76 (d), 36.13 (t), 37.94 (s), 38.82
t), 39.43 (t), 43.34 (s), 45.21 (d), 45.86 (s), 50.70 (d), 51.56 (d),
4.77 (d), 65.00 (t), 65.30 (t), 112.20 (s), 124.88 (d), 175.78 (s),
02.79 (s); MS m/ z (relative intensity) 470 (M , 2), 455 (1),
49 (1), 100 (4), 99 (100), 69 (2), 57 (1), 55 (3), 43 (1); HRMS
calcd for C31
Calcd for C31
(
2
(
0.64-1.94 (m, 26 H), 0.82 (s, 3 H), 0.85 (s, 3 H), 0.86 (d, J )
7.7 Hz, 6 H), 0.87 (d, J ) 7.7 Hz, 3 H), 0.90 (s, 3 H), 1.23 (s, 3
H), 1.98 (m, 1 H), 2.20 (m, 1 H), 2.99 (s, 3 H), 4.47 (ddd, J )
5.5, 10.6, 10.6 Hz, 1 H); 13C NMR δ 11.98 (q), 14.54 (q), 18.75
(q), 18.89 (q), 20.80 (t), 22.53 (q), 22.78 (q), 23.08 (q), 23.72 (t),
24.26 (q), 25.97 (t), 27.94 (d), 28.50 (t), 29.87 (t), 33.97 (t), 35.52
(d), 35.91 (t), 36.04 (t), 36.18 (s), 39.34 (t), 39.40 (t), 40.00 (q),
40.30 (d), 40.77 (s), 43.43 (s), 49.37 (d), 54.58 (d), 54.77 (d),
55.07 (d), 74.86 (s), 86.54 (d); MS m/ z (relative intensity) 428
5
2
1
+
+
H
50
O
3
(M ) 470.3760, found 470.3757. Anal.
H
50
O
3
: C, 79.10; H, 10.71. Found: C, 79.29; H,
1
0.90.
5′r,7′â)-4′,4′-Dim eth ylspir o[1,3-dioxolan e-2,3′-ch olestan -
′-ol] (10). To a stirred solution of 1.60 g (69.6 mmol) of Na
was added dropwise a
+
(
(M - 96, 47), 411 (30), 410 (71), 395 (27), 358 (54), 357 (100),
7
52
274 (22), 149 (26), 135 (21), 95 (25); HRMS calcd for C30H O
+
in 200 mL of refluxing liquid NH
3
(M - 96) 428.4018, found 428.4019. Anal. Calcd for
S: C, 70.94; H, 10.76. Found: C, 70.88; H, 11.18.
(5′r)-4′,4′-Dim eth ylsp ir o[1,3-d ioxola n e-2,3′-ch olesta n -
solution of 2.18 g (4.64 mmol) of 9 and 1.03 g (13.9 mmol) of
t-BuOH in 50 mL of THF. After the addition was complete,
the reaction mixture was stirred at reflux temperature for 1
31 56 4
C H O
h, and then 10.0 g of solid NH
to evaporate overnight, and 100 mL of H
added to the residue. The aqueous phase was extracted with
4
Cl was added. NH
3
was allowed
(
38) The NMR spectrum of this crude product revealed the presence
2
O was cautiously
of unreacted (5R,7â)-7-hydroxy-4,4-dimethylcholestan-3-one (∼30%)
and, probably, the C(3) epimer of 11 (∼10%).

Seven-Center Fragmentation
′-on e] (12). A solution of 2.58 g (5.48 mmol) of 9 and 0.425
g (5.74 mmol) of t-BuOH in 50 mL of THF was added dropwise
to a stirred solution of 0.515 g (74.6 mmol) of Li in 200 mL of
J . Org. Chem., Vol. 61, No. 3, 1996 865
7
(3r,5r,7r)-4,4,7-Tr im eth ylch olesta n e-3,7-d iol 3-Meth -
a n esu lfon a te (4). To a solution of 0.608 g (1.37 mmol) of 13
in 25 mL of dry THF was added dropwise 4.1 mL (4.1 mmol)
of L-Selectride (1 M in THF) at -78 °C. The solution was
stirred at -78 °C for 2 h and allowed to come to rt overnight.
The reaction mixture was cooled again to 0 °C and cautiously
quenched with 3 mL of EtOH followed by the addition of 3
mL of 4 M NaOH and 3 mL of 35% H
O , and stirring was
2 2
continued at rt for 6 h. The reaction mixture was then
concentrated under reduced pressure, taken up in 50 mL of
liquid NH
complete, stirring was continued for 10 min, and 10.0 g of solid
NH Cl was added. Workup as described for the synthesis of
0 was followed by flash chromatography (10:1 petroleum
3
at reflux temperature. After the addition was
4
1
1
ether (bp 40-60 °C)/EtOAc) to yield 2.33 g (90%) of 12:
NMR δ 0.61 (s, 3 H), 0.78 (s, 3 H), 0.80-2.44 (m, 27 H), 0.83
d, J ) 6.5 Hz, 6 H), 0.87 (d, J ) 7.7 Hz, 3 H), 0.96 (s, 3 H),
H
(
1
.12 (s, 3 H), 3.80-3.98 (m, 4 H); 13C NMR δ 11.99 (q), 13.66
2 2 2
H O, and extracted with four 25 mL portions of CH Cl . The
(
(
q), 18.75 (q), 19.53 (q), 21.12 (t), 22.51 (q), 22.57 (q), 22.75
q), 23.72 (t), 25.09 (t), 27.04 (t), 27.94 (d), 28.39 (t), 34.89 (t),
combined organic layers were dried and evaporated. The
remaining residue was treated with MsCl as described for the
synthesis of 2. After workup, the crude product was purified
by flash chromatography (5:1 petroleum ether (bp 40-60 °C)/
3
5.61 (d), 36.10 (t), 36.73 (s), 38.57 (t), 39.43 (t), 39.82 (t), 42.36
(s), 42.60 (s), 48.90 (d), 49.56 (d), 53.89 (d), 54.88 (d), 55.66
(
d), 64.92 (2t), 112.66 (s), 212.71 (s); MS m/ z (relative
EtOAc) to yield 0.605 g (84%) of 4: 1H NMR (C
6 6
D ) δ 0.70 (s,
+
intensity) 472 (M , 1), 457 (1), 169 (1), 131 (1), 121 (1), 119
6 H), 0.71 (s, 3 H), 0.85-2.06 (m, 28 H), 0.99 (d, J ) 6.8 Hz,
(
2), 107 (1), 100 (10), 99 (100), 69 (3); HRMS calcd for C31
H
3
52
O
: C,
3
6 H), 1.01 (s, 3 H), 1.06 (d, J ) 6.6 Hz, 3 H), 1.23 (s, 3 H), 2.36
+
13
(
M ) 472.3916, found 472.3915. Anal. Calcd for C31
8.76; H, 11.09. Found: C, 78.96; H, 11.32.
5r,7r)-7-h yd r oxy-4,4,7-tr im eth ylch olesta n -3-on e (13).
H
52
O
6 6
(s, 3 H), 4.50 (br s, W1/2 ) 6.3 Hz, 1 H); C NMR (C D ) δ
7
12.12 (q), 13.14 (q), 18.90 (q), 20.86 (t), 21.41 (q), 22.45 (q),
22.69 (q), 24.06 (t), 24.26 (t), 27.99 (t), 28.05 (q), 28.08 (d), 28.61
(t), 32.19 (t), 32.30 (q), 35.81 (d), 36.03 (t), 36.30 (s), 36.62 (s),
37.88 (q), 38.72 (t), 39.61 (t), 39.68 (t), 42.64 (d), 43.71 (s), 44.19
(d), 51.13 (d), 51.25 (d), 55.25 (d), 71.54 (s), 87.09 (d); MS m/ z
(
To a stirred solution of 2.31 g (4.89 mmol) of 12 in 50 mL of
dry ether was added 7.0 mL (18.2 mmol) of MeMgI (2.6 M in
ether) at 0 °C. The reaction mixture was stirred at 0 °C for 1
h and at rt for an additional 1 h. The excess of MeMgI was
+
(relative intensity) 524 (M , 0.3), 509 (1), 428 (7), 413 (11),
then cautiously destroyed with saturated aqueous NH
4
Cl.
411 (28), 410 (100), 395 (45), 121 (10), 109 (9), 95 (11); HRMS
+
After dilution with 100 mL of H O, the two-phase mixture was
2
calcd for C31
H
56
56
O
4
S (M ) 524.3899, found 524.3897. Anal.
separated, and the aqueous layer was extracted with three 100
mL portions of EtOAc. The combined organic layers were
washed with 100 mL of brine, dried, and evaporated. The
crude product was treated with HCl in acetone as described
for the synthesis of 11. After workup, the crude product was
purified by flash chromatography (10:1 petroleum ether (bp
Calcd for C31
11.05.
H
O
4
S: C, 70.94; H, 10.76. Found: C, 71.14; H,
(3â,5r,7r)-4,4,7-T r im e t h y l-7-[(t r im e t h y ls ily l)o x y ]-
ch olesta n -3-ol 3-Meth a n esu lfon a te (5). To a stirred solu-
tion of 0.217 g (0.41 mmol) of mesylate 3 in 10 mL of dry
pyridine were added 0.330 g (2.0 mmol) of hexamethyldisila-
zane (HMDS) and 0.650 g (6.0 mmol) of TMSCl. The reaction
mixture was stirred at 50 °C for 20 h and then neutralized
4
0-60 °C)/ EtOAc) to give 1.75 g (81%) of 13: 1H NMR δ 0.67
(
0
1
s, 3 H), 0.84 (d, J ) 6.5 Hz, 6 H), 0.91 (d, J ) 6.6 Hz, 3 H),
.95-2.05 (m, 26 H), 1.00 (s, 3 H), 1.02 (s, 3 H), 1.05 (s, 3 H),
3
with 10 mL of saturated aqueous NaHCO . The reaction
.30 (s, 3 H), 2.31 (m, 1 H), 2.62 (m, 1 H); 13C NMR δ 12.25
mixture was concentrated under reduced pressure, and the
concentrate was taken up in 50 mL of EtOAc. The organic
layer was washed with 25 mL of brine, dried, and evaporated.
The remaining residue was flash chromatographed (10:1
petroleum ether (bp 40-60 °C)/EtOAc) to give 0.233 g (95%)
of 5: 1H NMR δ 0.09 (s, 9 H), 0.60 (s, 3 H), 0.75-2.03 (m, 27
H), 0.83 (s, 6 H), 0.85 (d, J ) 6.5 Hz, 6 H), 0.89 (d, J ) 6.5 Hz,
3 H), 0.97 (s, 3 H), 1.30 (s, 3 H), 3.00 (s, 3 H), 4.35 (dd, J )
(
(
q), 13.16 (q), 18.86 (q), 21.34 (t), 21.72 (q), 22.57 (q), 22.81
q), 23.70 (t), 25.32 (q), 29.97 (t), 28.01 (d), 28.57 (t), 32.44 (q),
3
(
5
4.75 (t), 35.68 (d), 36.12 (t), 36.26 (s), 38.46 (t), 39.50 (t), 39.58
t), 39.70 (t), 43.98 (s), 44.00 (d), 47.04 (s), 48.63 (d), 51.11 (d),
1.36 (d), 55.03 (d), 72.46 (s), 216.99 (s); MS m/ z (relative
+
intensity) 444 (M , 56), 430 (29), 429 (100), 426 (45), 411 (21),
41 (48), 340 (21), 125 (28), 43 (20); HRMS calcd for C30
3
52 2
H O
+
13
(M ) 444.3967, found 444.3967. Anal. Calcd for C30
H
52
O
2
: C,
5.1, 11.0 Hz, 1 H); C NMR δ 2.68 (3q), 12.26 (q), 13.54 (q),
8
1.02; H, 11.79. Found: C, 80.88; H, 12.06.
16.32 (q), 18.90 (q), 20.84 (t), 22.51 (q), 22.80 (q), 23.97 (t),
25.67 (t), 27.74 (t), 27.94 (d), 28.38 (q), 28.50 (t), 31.71 (q), 35.73
(d), 36.06 (s), 36.15 (t), 37.37 (t), 38.06 (s), 38.79 (q), 39.18 (t),
39.47 (t), 39.50 (t), 43.55 (s), 45.68 (d), 47.74 (d), 50.37 (d), 50.98
(d), 55.02 (d), 75.85 (s), 90.78 (d); MS m/ z (relative intensity)
(
3â,5r,7r)-4,4,7-Tr im eth ylch olesta n e-3,7-d iol 3-Meth -
a n esu lfon a te (3). To a stirred solution of 0.645 g (1.45 mmol)
of 13 in 50 mL of dry THF was added 0.080 g (2.11 mmol) of
LAH at -78 °C. The reaction mixture was stirred at -78 °C
for 2 h and at rt for an additional 30 min. The reaction
mixture was then carefully quenched with a small amount of
+
596 (M , 9), 500 (35), 485 (13), 419 (32), 418 (100), 410 (69),
395 (17), 210 (12), 143 (34), 73 (18); HRMS calcd for C34
SSi (M ) 596.4295, found 596.4298. Anal. Calcd for C34
H
H
64
O
O
4
-
-
+
saturated aqueous Na
Cl , the organic layer was dried and evaporated. The remain-
ing residue was flash chromatographed (5:1 petroleum ether
2
SO
4
. After addition of 50 mL of CH
2
-
64
4
SSi: C, 68.41; H, 10.81. Found: C, 68.14; H, 10.92.
2
(3r,5r,7r)-4,4,7-Tr im e t h yl-7-[(t r im e t h ylsilyl)oxy]-
ch olesta n -3-ol 3-Meth a n esu lfon a te (6). The mesylate 4
(0.300 g, 0.57 mmol) was treated with HMDS and TMSCl for
24 h as described above for the synthesis of 5. Workup and
flash chromatography (10:1 petroleum ether (bp 40-60 °C)/
1
(
bp 40-60 °C)/EtOAc) to give 0.643 g (99%) of a diol ( H NMR
δ 0.63 (s, 3 H), 0.73-2.00 (m, 29 H), 0.76 (s, 3 H), 0.80 (s, 3
H), 0.84 (d, J ) 6.5 Hz, 6 H), 0.89 (d, J ) 6.5 Hz, 3 H), 0.94 (s,
3
H), 1.26 (s, 3 H), 3.22 (m, 1 H)). This diol (0.528 g, 1.18
1
mmol) was treated with MsCl as described for the synthesis
EtOAc) yielded 0.307 g (90%) of 6: H NMR δ 0.09 (s, 9 H),
of 2. After workup, the crude product was purified by flash
chromatography (5:1 petroleum ether (bp 40-60 °C)/EtOAc)
0.61 (s, 3 H), 0.80-2.00 (m, 27 H), 0.82 (s, 3 H), 0.84 (d, J )
6.8 Hz, 6 H), 0.88 (d, J ) 6.5 Hz, 3 H), 0.89 (s, 3 H), 0.96 (s, 3
H), 1.29 (s, 3 H), 2.99 (s, 3 H), 4.52 (br s, W1/2 ) 4.0 Hz, 1 H);
1
to yield 0.566 g (91%) of 3: H NMR (C
0
6
6
D
6
) δ 0.70 (s, 3 H),
1
3
.71 (s, 3 H), 0.76-2.10 (m, 28 H), 0.84 (s, 3 H), 0.95 (d, J )
C NMR δ 2.61 (3q), 12.26 (q), 13.54 (q), 18.90 (q), 20.61 (t),
.8 Hz, 6 H), 1.03 (s, 3 H), 1.08 (d, J ) 6.4 Hz, 3 H), 1.23 (s, 3
21.79 (q), 22.53 (q), 22.81 (q), 24.00 (t), 24.39 (t), 27.79 (t), 27.94
(d), 28.54 (q), 28.54 (t), 31.84 (q), 32.15 (t), 35.78 (d), 36.06 (s),
36.16 (t), 36.91 (s), 38.63 (q), 38.88 (t), 39.47 (2t), 42.06 (d),
43.49 (s), 45.89 (d), 50.37 (d), 51.16 (d), 55.08 (d), 75.84 (s),
1
3
H), 2.38 (s, 3 H), 4.36 (dd, J ) 4.5, 11.9 Hz, 1 H); C NMR
(C
6
D
6
) δ 12.11 (q), 13.16 (q), 16.09 (q), 18.93 (q), 21.02 (t), 22.48
q), 22.75 (q), 24.05 (t), 25.63 (t), 27.72 (q), 28.00 (t), 28.11 (d),
8.69 (t), 32.21 (q), 35.86 (d), 36.33 (t), 37.12 (s), 37.12 (t), 37.85
s), 37.94 (q), 38.83 (t), 39.62 (t), 39.90 (t), 43.76 (d), 43.97 (s),
7.64 (d), 51.01 (d), 51.27 (d), 55.26 (d), 71.67 (s), 89.21 (d);
(
2
(
+
88.17 (d); MS m/ z (relative intensity) 596 (M , 5), 500 (11),
485 (10), 419 (33), 418 (100), 410 (64), 395 (28), 210 (8), 143
+
4
(17), 73 (10); HRMS calcd for C34
H
64
O
4
SSi (M ) 596.4295, found
SSi: C, 68.41; H, 10.81.
+
MS m/ z (relative intensity) 524 (M , 1), 509 (30), 428 (26),
13 (20), 411 (37), 410 (100), 395 (51), 367 (16), 121 (23), 95
596.4297. Anal. Calcd for C34
Found: C, 68.38; H, 11.00.
H
64
O
4
4
+
(20); HRMS calcd for C31
H
O
56
O
4
S (M ) 524.3899, found 524.3899.
Rea ction s of Mesyla tes 1-6 w ith Sod iu m ter t-Am yla te.
Gen er a l P r oced u r e. All reactions were carried out at a
concentration of ca. 0.1 M mesylate in dry toluene. The
Anal. Calcd for C31
H
56
4
S: C, 70.94; H, 10.76. Found: C,
7
0.98; H, 10.98.

8
66 J . Org. Chem., Vol. 61, No. 3, 1996
Bastiaansen et al.
solutions were degassed and refluxed under an Ar atmosphere.
Approximately 5 equiv of sodium tert-amylate (3.2 M in
toluene) was added at once, via syringe, to the refluxing
solution of the mesylate. The reaction mixture was heated at
reflux temperature for 10 min, quenched with precooled
112.09 (t), 124.84 (d), 131.08 (s), 147.66 (d), 213.82 (s); MS m/ z
(relative intensity) 428 (M , 11), 411 (31), 410 (100), 396 (17),
+
395 (51), 137 (46), 121 (14), 109 (11), 95 (19), 43 (13); HRMS
+
calcd for C30
H
52O (M ) 428.4018, found 428.4014.
1
20: H NMR δ 0.68 (s, 3 H), 0.72 (s, 3 H), 0.85 (d, J ) 6.6
Hz, 6 H), 0.82-2.03 (m, 28 H), 0.91 (d, J ) 6.5 Hz, 3 H), 1.27
(s, 3 H), 1.75 (s, 3 H), 2.71 (m, 1 H), 4.82 (d, J ) 7.7 Hz, 2 H);
4
saturated aqueous NH Cl, and then quickly cooled to 0 °C.
The mixture was vigorously stirred for 20 min, followed by
extraction with five portions of EtOAc. The combined organic
layers were dried and evaporated under reduced pressure to
afford the crude reaction products. Unless noted otherwise,
product ratios, yields, and pure compounds were obtained by
chromatographical techniques.
1
3
C NMR δ 12.37 (q), 13.47 (q), 18.90 (q), 22.52 (q), 22.75 (q),
23.76 (2t), 25.24 (q), 27.75 (t), 27.76 (t), 27.96 (d), 28.45 (t),
32.30 (q), 35.64 (d), 36.12 (t), 39.48 (t), 39.82 (s), 39.82 (2t),
41.31 (t), 44.50 (d), 44.50 (s), 46.04 (d), 47.46 (d), 51.28 (d),
51.50 (d), 55.10 (d), 73.71 (s), 110.62 (t), 148.13 (s); MS m/ z
+
a . The general procedure was employed by using 0.093 g
(relative intensity) 428 (M , 73), 413 (25), 411 (29), 410 (100),
(0.19 mmol) of 1. Workup and flash chromatography (100:1
395 (54), 161 (25), 121 (38), 109 (32), 95 (35), 43 (38); HRMS
+
to 2:1 petroleum ether (bp 40-60 °C)/EtOAc) afforded 0.005 g
calcd for C30
H
52O (M ) 428.4018, found 428.4013.
3
9
1
(
7%) of an inseparable 3:2 mixture of 14 and 15, respectively,
21: H NMR δ 0.60 (s, 3 H), 0.68 (s, 3 H), 0.80-2.33 (m, 28
and 0.070 g (75%) of unreacted 1.
H), 0.85 (d, J ) 6.9 Hz, 6 H), 0.91 (d, J ) 6.4 Hz, 3 H), 1.28 (s,
1
3 H), 1.58 (br s, 3 H), 1.69 (br s, 3 H); 13C NMR δ 12.35 (q),
(
5r,7r)-7-Met h ylch olest -2-en -7-ol (14): H NMR (main
peak) δ 5.59 (br s, W1/2 ≈ 5 Hz, 2 H); HRMS (3:2 mixture) calcd
13.30 (q), 18.90 (q), 19.84 (q), 22.53 (q), 22.76 (q), 22.91 (q),
23.68 (t), 23.75 (t), 27.74 (t), 27.97 (d), 28.48 (t), 29.22 (t), 32.30
(q), 35.65 (d), 36.12 (t), 37.43 (t), 39.49 (t), 39.86 (t), 43.36 (t),
+
for C28
H
48O (M ) 400.3705, found 400.3703.
1
(
5r,7r)-7-Met h ylch olest -3-en -7-ol (15): H NMR (main
4
5
4.05 (d), 44.27 (s), 44.41 (s), 49.41 (d), 49.79 (d), 51.19 (d),
peaks) δ 5.20 (br dd, J ) 2.2, 10.0 Hz, 1 H), 5.56 (m, 1 H).
5.13 (d), 73.33 (s), 121.47 (s), 135.42 (s); MS m/ z (relative
b. The general procedure was employed by using 0.336 g
+
intensity) 428 (M , 8), 411 (29), 410 (100), 395 (40), 367 (32),
(
2
(
0.64 mmol) of 2. Workup and flash chromatography (50:1 to
:1 petroleum ether (bp 40-60 °C)/EtOAc) afforded 0.008 g
3%) of 16, 0.104 g (38%) of an inseparable 1:5 mixture39 of 17
1
63 (17), 137 (19), 109 (18), 95 (22), 43 (22); HRMS calcd for
+
30
C H52O (M ) 428.4018, found 428.4016.
1
2
2: H NMR δ 0.68 (s, 3 H), 0.80-2.33 (m, 27 H), 0.85 (d, J
and 18, respectively, and 0.172 g (51%) of unreacted 2. The
)
6.3 Hz, 6 H), 0.89 (s, 3 H), 0.91 (d, J ) 6.7 Hz, 3 H), 0.93 (d,
spectroscopic data of 16, 17, and 18 are shown below.
1
J ) 6.8 Hz, 3 H), 0.99 (d, J ) 6.8 Hz, 3 H), 1.29 (s, 3 H), 2.65
1
6: H NMR δ 0.67 (s, 3 H), 0.85 (d, J ) 6.7 Hz, 6 H), 0.84-
(
(
qq, J ) 6.8, 6.8 Hz, 1 H); 13C NMR δ 12.38 (q), 18.10 (q), 18.99
q), 21.26 (q), 22.21 (q), 22.56 (q), 22.79 (q), 23.34 (t), 23.83 (t),
2
.08 (m, 21 H), 0.88 (d, J ) 6.7 Hz, 3 H), 1.00 (s, 3 H), 1.56 (d,
J ) 1.1 Hz, 3 H), 1.60 (d, J ) 1.2 Hz, 3 H), 2.11 (s, 3 H), 2.15-
2
(
6.23 (d), 27.82 (t), 28.01 (d), 28.37 (t), 28.57 (t), 30.48 (q), 35.62
d), 36.17 (t), 37.72 (t), 39.54 (t), 39.82 (t), 40.37 (t), 44.42 (s),
45.63 (d), 49.57 (s), 50.77 (d), 51.39 (d), 55.06 (d), 73.06 (s),
2
1
.50 (m, 3 H), 4.66 (dd, J ) 2.3, 9.9 Hz, 1 H), 4.69 (dd, J ) 2.3,
6.9 Hz, 1 H), 4.79 (br s, 1 H), 5.36 (dt, J ) 9.9, 16.9 Hz, 1 H);
C NMR δ 11.96 (q), 18.57 (q), 19.26 (q), 21.81 (q), 22.54 (t),
2.71 (q), 22.80 (q), 23.78 (t), 26.08 (t), 27.63 (t), 27.98 (q), 27.98
1
3
+
1
1
1
4
36.14 (s), 140.88 (s); MS m/ z (relative intensity) 428 (M ,
2
1), 411 (35), 410 (96), 395 (78), 367 (48), 189 (22), 161 (25),
(d), 30.05 (q), 33.60 (t), 35.84 (d), 36.00 (t), 39.46 (t), 39.72 (t),
+
37 (100), 121 (20), 43 (31); HRMS calcd for C30
28.4018, found 428.4017.
H
52O (M )
3
5
9.99 (t), 41.19 (s), 41.96 (s), 45.97 (d), 52.28 (d), 54.13 (d),
6.54 (d), 110.04 (t), 128.96 (s), 135.47 (d), 144.48 (d), 210.24
+
d . The general procedure was employed by using 0.213 g
0.41 mmol) of 4. Workup and flash chromatography (250:1
to 2:1 petroleum ether (bp 40-60 °C)/EtOAc) afforded 0.084 g
(s); MS m/ z (relative intensity) 428 (M , 6), 357 (35), 354 (7),
(
3
01 (7), 179 (9), 153 (100), 149 (25), 135 (85), 95 (43), 43 (37);
+
HRMS calcd for C30
H52O (M ) 428.4018, found 428.4014.
3
9
(
48%) of 24, 0.048 g (27%) of an inseparable 1:3.5 mixture of
1
(
3r,5r)-3,4,4-Tr im eth ylch olest-6-en -3-ol (17): H NMR
2
3 and 25, respectively, and 0.005 g (2%) of unreacted 4. The
1
3
(
main peaks) δ 0.66 (s, 3 H), 1.19 (s, 3 H), 5.56 (br s, 2 H);
NMR (main peaks) δ 12.11 (q), 14.17 (q), 34.25 (t), 34.25 (t),
4.68 (t), 37.00 (d), 53.00 (d), 54.49 (d), 126.88 (d), 130.40 (d);
C
4
0
40
spectroscopic data of 23, 24, and 25 are shown below.
1
(
5r,7r)-4,4,7-Tr im eth ylch olest-2-en -7-ol (23): H NMR
3
(main peaks) δ 0.66 (s, 3 H), 0.84 (d, J ) 6.5 Hz, 6 H), 1.29 (s,
+
MS m/ z (relative intensity) 410 (M - 18, 2), 297 (1), 147 (5),
1
3
1
5
H), 5.33-5.52 (m, 2 H); 13C NMR (main peaks) δ 12.14 (q),
2.91 (q), 31.46 (q), 39.73 (t), 42.55 (t), 44.04 (d), 45.01 (d),
36 (8), 122 (20), 121 (18), 105 (18), 69 (20), 55 (46), 43 (100);
+
HRMS (1:5 mixture) calcd for C30
28.4017.
H
52O (M ) 428.4018, found
0.04 (d), 51.41 (d), 121.64 (d), 137.64 (d); MS m/ z (relative
4
+
intensity) 410 (M - 18, 5), 395 (5), 161 (15), 147 (14), 137
1
(
3r,5r)-3,4,4-Tr im eth ylch olest-7-en -3-ol (18): H NMR
(63), 121 (74), 119 (43), 109 (50), 107 (61), 69 (100); HRMS
(
3
main peaks) δ 0.50 (s, 3 H), 0.83 (d, J ) 6.7 Hz, 6 H), 0.84 (s,
+
(1:3.5 mixture) calcd for C30H52O (M ) 428.4018, found 428.4020.
H), 0.85 (s, 3 H), 0.98 (s, 3 H), 0.90 (d, J ) 6.3 Hz, 3 H), 1.18
1
(
3â,7r)-3,4,7-Tr im eth ylch olest-4-en -7-ol (24): H NMR
13
(
(
s, 3 H), 5.20 (m, 1 H); C NMR (main peaks) δ 11.77 (q), 15.05
(
C D ) δ 0.70 (s, 3 H), 0.80-2.15 (m, 28 H), 0.92 (d, J ) 6.1
6 6
q), 18.82 (q), 20.96 (t), 23.52 (t), 23.90 (t), 24.41 (q), 27.96 (d),
Hz, 6 H), 0.95 (s, 3 H), 0.95 (d, J ) 6.4 Hz, 3 H), 1.00 (d, J )
3
7
4
5
6.10 (t), 36.18 (d), 48.39 (d), 52.46 (d), 54.86 (d), 56.05 (d),
6
1
2
3
4
7
4
1
4
.3 Hz, 3 H), 1.32 (s, 3 H), 1.63 (s, 3 H); 13C NMR (C
6 6
D ) δ
5.20 (s), 118.18 (d), 138.79 (s); MS m/ z (relative intensity)
2.35 (q), 16.63 (q), 18.93 (q), 19.91 (2q), 22.29 (t), 22.48 (q),
2.75 (q), 24.08 (t), 27.82 (t), 28.11 (d), 28.79 (2t), 30.61 (q),
5.87 (2d), 36.33 (t), 36.50 (t), 37.71 (s), 39.62 (t), 40.16 (t),
2.90 (t), 43.90 (s), 45.24 (d), 50.48 (d), 51.79 (d), 55.31 (d),
3.18 (s), 131.69 (s), 134.10 (s); MS m/ z (relative intensity)
+
28 (M , 0.4), 161 (4), 149 (4), 119 (9), 95 (13), 71 (38), 69 (16),
7 (31), 55 (26), 43 (100).
c. The general procedure was employed by using 0.252 g
(0.48 mmol) of 3. Workup and flash chromatography (100:1
to 2:1 petroleum ether (bp 40-60 °C)/EtOAc) afforded 0.040 g
+
28 (M , 11), 411 (33), 410 (100), 396 (22), 395 (59), 187 (13),
(
(
20%) of 19, 0.026 g (13%) of 20, 0.045 g (22%) of 21, 0.012 g
6%) of 22, and 0.043 g (17%) of unreacted 3. The spectroscopic
+
63 (14), 137 (47), 95 (28); HRMS calcd for C30
28.4018, found 428.4018.
(
H
52O (M )
data of 19-22 are shown below.
1
5r,7r)-3,4,7-Tr im eth ylch olest-3-en -7-ol (25): H NMR
1
1
9: H NMR δ 0.63 (s, 3 H), 0.84 (d, J ) 6.6 Hz, 6 H), 0.80-
(
3
main peaks) δ 0.67 (s, 3 H), 0.84 (d, J ) 6.5 Hz, 6 H), 1.27 (s,
2
.16 (m, 23 H), 0.89 (d, J ) 6.5 Hz, 3 H), 0.96 (s, 3 H), 1.54 (br
13
H), 1.53 (s, 3 H), 1.58 (s, 3 H); C NMR (main peaks) δ 11.90
s, 3 H), 1.64 (br s, 3 H), 2.04 (s, 3 H), 2.44 (m, 1 H), 4.86 (br d,
(
(
q), 12.35 (q), 15.45 (q), 18.84 (q), 19.38 (d), 21.54 (t), 22.54
q), 22.78 (q), 27.98 (d), 29.58 (t), 32.25 (q), 35.72 (d), 36.10 (t),
J ) 10.7 Hz, 1 H), 4.89 (dd, J ) 1.5, 17.6 Hz, 1 H); 5.01 (br t,
1
3
J ) 7.0 Hz, 1 H), 5.61 (dd, J ) 10.7, 17.6 Hz, 1 H); C NMR
δ 11.34 (q), 16.60 (q), 17.55 (q), 18.59 (q), 21.52 (t), 22.49 (q),
4
1
1
3.10 (d), 43.76 (d), 49.29 (d), 51.74 (d), 55.07 (d), 72.58 (s),
+
25.11 (s), 126.08 (s); MS m/ z (relative intensity) 410 (M
-
2
(
4
2.66 (t), 22.74 (q), 23.73 (t), 25.04 (t), 25.63 (q), 27.94 (t), 27.94
d), 34.61 (q), 35.76 (d), 36.01 (t), 38.97 (t), 39.42 (t), 40.63 (t),
2.14 (s), 43.07 (s), 45.32 (d), 51.42 (d), 53.61 (d), 54.95 (d),
8, 2), 395 (2), 135 (7), 121 (10), 119 (6), 109 (14), 107 (10), 95
(21), 57 (31), 43 (100).
(
40) The mass spectral data of this compound were obtained by GC-
(
39) This ratio was determined by 1H NMR analysis.
MS analysis.

Seven-Center Fragmentation
J . Org. Chem., Vol. 61, No. 3, 1996 867
e. The general procedure was employed by using 0.168 g
(5r,7r)-3,4,7-Tr im eth yl-7-[(tr im eth ylsilyl)oxy]ch olest-
1
(
0.28 mmol) of 5. Workup and flash chromatography (100:1
to 50:1 petroleum ether (bp 40-60 °C)/EtOAc) gave 0.024 g
17%) of a complex mixture of at least five apolar compounds
and 0.111 g (66%) of unreacted 5.
f. The general procedure was employed by using 0.191 g
3-en e (27): H NMR (main peaks) δ 0.10 (s, 9 H), 0.63 (s, 3
H), 0.83 (d, J ) 6.5 Hz, 6 H), 1.26 (s, 3 H), 1.54 (br s, 3 H),
1.60 (br s, 3 H); 13C NMR (main peaks) δ 2.65 (3q), 19.35 (q),
21.48 (t), 27.70 (t), 29.74 (t), 31.80 (q), 45.62 (d), 48.50 (d), 51.72
(d), 75.91 (s), 124.82 (s), 126.63 (s); MS m/ z (relative intensity)
(
+
(
0.32 mmol) of 6. Workup and flash chromatography (100:1
410 (M - 90, 22), 396 (30), 201 (21), 179 (16), 161 (17), 109
to 2:1 petroleum ether (bp 40-60 °C)/EtOAc) afforded 0.044 g
(51), 75 (69), 73 (79), 57 (100), 43 (75).
3
9
(
27%) of an inseparable ca. 1:1 mixture of 26 and 27 and
0
.111 g (58%) of unreacted 6. The spectroscopic data of 26
Ack n ow led gm en t. We thank A. van Veldhuizen for
4
1
and 27 are shown below.
1
13
recording H NMR and C NMR spectra and C. J .
Teunis, H. J ongejan, and R. van Dijk for mass spectral
data and elemental analyses. In addition, we thank Dr.
H. Kooijman (from the Bijvoet Center for Biomolecular
Research, Utrecht, The Netherlands) for carrying out
and interpreting the X-ray diffraction.
(
5r,7r)-4,4,7-Tr im eth yl-7-[(tr im eth ylsilyl)oxy]ch olest-
1
2
-en e (26): H NMR (main peaks) δ 0.10 (s, 9 H), 0.65 (s, 3
H), 0.86 (d, J ) 6.3 Hz, 6 H), 0.94 (s, 3 H), 1.32 (s, 3 H), 5.35-
5
2
1
3
.55 (m, 2 H); C NMR (main peaks) δ 2.65 (3q), 18.91 (q),
0.96 (t), 24.00 (t), 28.60 (t), 32.07 (q), 40.19 (t), 40.98 (t), 44.58
(
(
(
d), 46.01 (d), 49.12 (d), 51.29 (d), 75.70 (s), 121.93 (d), 137.79
d); MS m/ z (relative intensity) 418 (M - 82, 5), 210 (4), 175
6), 143 (9), 95 (18), 81 (17), 75 (31), 73 (60), 57 (32), 43 (100);
+
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
compounds 16, 19-22, and 24 (6 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
+
HRMS (1:1 mixture) calcd for C33
00.4416.
H
60OSi (M ) 500.4414, found
5
(41) Careful column chromatography afforded a small sample of a
ca. 1:2 mixture of 26 and 27, respectively. With this mixture the
structures of 26 and 27 could be established by spectroscopic analysis
1
13
(
H and C NMR and GC-MS).
J O951655Z