The Mitsunobu inversion reaction of sterically hindered 17-hydroxy steroids

Article abstract of DOI:10.1007/s00706-004-0203-9

The Mitsunobu inversion reaction of 3-methoxyestra-1,3,5(10)-trien- 17β-ol is dramatically influenced by the acidic component. There appears to be a relationship between the dissociation constant of the electron-withdrawing substituent on the aryl acid and the overall effectiveness of the reaction, with more acidic species generally providing a higher yield of inverted product. Springer-Verlag 2004.

Full text of DOI:10.1007/s00706-004-0203-9

Monatshefte f€ur Chemie 135, 1129–1136 (2004)
DOI 10.1007/s00706-004-0203-9
The Mitsunobu Inversion Reaction
of Sterically Hindered 17-Hydroxy
Steroids
´
Pal Tapolcsanyi, Janos Wolfling, Erzsebet Mernyak,
and Gyula Schneider
´
´
¨
´
´
ꢀ
Department of Organic Chemistry, University of Szeged, H-6720 Szeged, Hungary
Received March 4, 2004; accepted March 29, 2004
Published online June 30, 2004 # Springer-Verlag 2004
Summary. The Mitsunobu inversion reaction of 3-methoxyestra-1,3,5(10)-trien-17ꢀ-ol is dramatically
influenced by the acidic component. There appears to be a relationship between the dissociation
constant of the electron-withdrawing substituent on the aryl acid and the overall effectiveness of the
reaction, with more acidic species generally providing a higher yield of inverted product.
Keywords. Steroids; Estrone derivatives; Mitsunobu reaction; Carboxylic acids; Isomers.
Introduction
The Mitsunobu reaction is widely employed for the inversion of configuration in
secondary alcohol derivatives [1, 2]. In general, this method proves efficacious for
a variety of substrates, furnishing useful yields of inverted products under mild,
essentially neutral reaction conditions. One of the most frequent goals of the
application of the Mitsunobu reaction is the epimerization of optically active sec-
ondary alcohols, and most of the published papers have reported and discussed
such transformations [3]. In these procedures, the optically active alcohol is reacted
with different carboxylic acids under Mitsunobu conditions. In most cases, the acid
strength is not the determining factor in the reaction, but for sterically hindered
alcohols, application of a strong acid is recommended [4]. Early references to this
observation included primary and secondary substrates such as nucleosides, carbo-
hydrates, and alkaloids [5–7]. 4-Nitrobenzoic acid has been shown to be a par-
ticularly effective coupling reagent for sterically encumbered alcohols [8, 9], a
finding which has led to practical modifications of the Mitsunobu reaction
ꢀ
Dedicated to Professor Sandor Antus on the occasion of his 60th birthday
Corresponding author. E-mail: schneider@chem.u-szeged.hu
´

´
P. Tapolcsanyi et al.
1130
[10, 11]. Despite these observations, the origins of this interesting phenomenon
have received little attention [12].
Dodge et al. have recently investigated the role of the carboxylic acid in the
Mitsunobu inversion of menthol [13], and other sterically hindered hydroxyl func-
tions on different sterane skeletons [14]. They found that, with different acids as
nucleophilic reagents, the relatively acidic 4-nitrobenzoic acid proved to be the
most convenient coupling partner. We have studied the Mitsunobu inversion of
the 17ꢀ- and 17ꢁ-hydroxy groups of 3-methoxyestra-1,3,5(10)-trien-17-ol (1, 2),
using different carboxylic acids and solvents. Moreover, we have investigated
the reactions of 16ꢀ-methyl- and 16ꢁ-methyl-3-methoxyestra-1,3,5(10)-trien-
17ꢀ-ol (4, 5) in order to establish whether the Mitsunobu reaction is suit-
able for the inversion of strongly hindered alcohols which have two adjacent
substituents.
Results and Discussion
Since literature reveals that both the carboxylic acid and the solvent used for the
Mitsunobu inversion process exert a considerable influence on the outcome of
the reaction, we performed the Mitsunobu inversion of 3-methoxyestra-1,3,5(10)-
trien-17ꢀ-ol (1) with various carboxylic acids in three different solvents: toluene,
chlorobenzene, and hexamethylphosphorus triamide (HMPT). The results demon-
strated that the more acidic aromatic carboxylic acids gave better yields in the
Mitsunobu reaction. The traditionally used relatively weak benzoic acid (pKa ¼
4.16) is one of the least attractive coupling partners. 4-Nitrobenzoic acid (pKa ¼
3.41) gave a similar result to that reported by Dodge and Lugar [14]. Despite
being more acidic, 3,5-dinitrobenzoic acid (pKa ¼ 2.82) led to a yield similar to
that with 4-nitrobenzoic acid. The best result was achieved with 2,4-dinitroben-
zoic acid, which has the highest dissociation constant (pKa ¼ 1.42). Among the
solvents, toluene and chlorobenzene proved convenient. HMPT is suitable for the
SN2 exchange reactions of steroid-oxyphosphonium species with nucleophilic aryl
acids. In the present systems, however, HMPT furnished only lower yields and the
work-up of the reaction mixture was more difficult, too. In each case, besides the
desired product we isolated an unsaturated side-product, 3-methoxy-17ꢀ-methyl-
estra-1,3,5(10),13-tetraene (3). Evidence of the structure of compound 3 was
provided by the exchange reaction of 17ꢁ-chloro-3-methoxyestra-1,3,5(10)-triene
with sodium acetate [15] and by the acetolysis of 17ꢀ-p-tolylsulfonyloxy-3-
methoxyestra-1,3,5(10)-triene with potassium acetate [16]. The formation of this
compound can be explained by the Wagner-Meerwein rearrangement of steroid-
oxyphosphonium species in parallel with the nucleophilic exchange reaction. The
first Wagner-Meerwein rearrangement under Mitsunobu reaction conditions had
been found for the rigid bicyclo[3.1.1.]heptanol system by Evans et al. [17]
(Scheme 1).
To compare the reactivities of the 17ꢀ- and 17ꢁ-hydroxy groups, we carried out
the Mitsunobu reaction of 3-methoxyestra-1,3,5(10)-trien-17ꢁ-ol (2) with 4-nitro-
benzoic acid in toluene and in chlorobenzene. 3-Methoxyestra-1,3,5(10)-trien-17ꢀ-
yl (4-nitrobenzoate) (1a) was obtained in only 10% and 12% yield. The main
product in both cases was 3. The pseudo axial 17ꢁ-hydroxy group exhibits low

The Mitsunobu Inversion Reaction
1131
Scheme 1
reactivity, in agreement with the general behaviour of the axial hydroxy group of
steroids [3].
We performed the Mitsunobu inversion of the very encumbered 16ꢀ-methyl
and 16ꢁ-methyl-3-methoxyestra-1,3,5(10)-trien-17ꢀ-ols (4, 5). The two methyl
functionalities cause strong steric hindrance about the adjacent alcohol moiety.
For compounds unsubstituted at position C-16 (1, 2), the most acidic 2,4-
dinitrobenzoic acid has proved to be the most effective. In the case of 16ꢀ-
methyl-3-methoxyestra-1,3,5(10)-trien-17ꢀ-ol (4), 2,4-dinitrobenzoic acid and
4-nitrobenzoic acid gave similar yields. For the Mitsunobu reaction we chose the
less bulky 4-nitrobenzoic acid, which resulted in 17% 16ꢀ-methyl-3-methoxyestra-
1,3,5(10)-trien-17ꢁ-yl (4-nitrobenzoate) (6a), 15% 16-methyl-3-methoxyestra-
1,3,5(10),16-tetraene (7), and 65% unreacted starting compound 4. A surprisingly
high yield was achieved in the Mitsunobu inversion of 16ꢁ-methyl-3-methoxyestra-
1,3,5(10)-trien-17ꢀ-ol (5). 16ꢁ-Methyl-3-methoxyestra-1,3,5(10)-trien-17ꢁ-yl (4-
nitrobenzoate) (8a) was obtained in 60%, 7 in 5%, and unreacted 5 in 31% yield
(Scheme 2).
The great difference in reactivity of the two isomers 4 and 5 can be explained
by steric reasons. The formation of the oxyphosphonium salt at the ꢀ-hydroxy
group is probably much more hindered by the two adjacent methyl groups, also
in the ꢀ-position, than in the case of 5, where the methyl group is situated far from
the reaction center. The formation of 16-methyl-3-methoxyestra-1,3,5(10),16-tetra-
ene (7) can easily be explained by the elimination of the oxyphosphonium species
preceding the last step of the Mitsunobu reaction.

´
P. Tapolcsanyi et al.
1132
Scheme 2
Experimental
Melting points were determined on a Kofler block. Optical rotations were measured in chloroform
(c ¼ 1) on a Polamat-A (Zeiss, Jena) polarimeter at 23ꢁC, and [ꢁ]D values are given in 10ꢂ1 ꢁcm2 gꢂ1
.
NMR spectra were recorded on a Bruker AM 400 instrument. Chemical shifts (ꢂ) are given in ppm, and
coupling constants (J) in Hz. For the determination of multiplicities, the J-MOD pulse sequence was
used. Elemental analysis data were determined with a Perkin-Elmer CHN analyser model 2400. For all
new compounds satisfactory elemental analyses were obtained. Thin-layer chromatography: silica gel
60, layer thickness 0.2 mm (Merck); solvent system (ss): (A) chloroform, (B) ethyl acetate:chloroform
(5:95, v=v); detection with iodine or UV (365 nm) or by spraying with 50% phosphoric acid and
heating at 100–120ꢁC for 10 min. Flash chromatography: silica gel 60, 40–63mm.
General Procedure
3-Methoxyestra-1,3,5(10)-trien-17-ol (1, 2; 1 mmol, 286 mg) or 3-methoxy-16-methylestra-1,3,5(10)-
trien-17-ol (4, 5; 1 mmol, 300 mg), 656 mg of triphenylphosphine (2.5mmol), and 305 mg of benzoic
acid (2.5mmol) or substituted benzoic acid (2.5mmol) were suspended in 15cm3 of dry toluene in a
50cm3, three-necked, round-bottomed flask equipped with a thermometer, a dropping funnel, and a
reflux condenser with a CaCl2-tube. The flask was placed on a heating magnetic stirrer and stirred for
2 min, and 436 mg of diethyl azodicarboxylate (2.5 mmol) were then added dropwise at room tem-
perature. The suspension cleared to give a yellow-orange solution and became slightly warm. The
reaction mixture was kept at 80ꢁC for 1.5 h, the solvent was then removed under vacuum, and the
residue was subjected to chromatography and eluted with CHCl3. With HMPT as solvent, the work-up

The Mitsunobu Inversion Reaction
1133
was as follows: at the end of the reaction, the reaction mixture was poured into water and allowed to
stand overnight. A dense oil formed, which was then separated from H2O and dissolved in CHCl3.
After drying and evaporation of the solvent, the residue was chromatographed on a silica gel column
with CHCl3.
Mitsunobu Esterification of 1 with Benzoic Acid
In toluene: A mixture of 286 mg of 1 (1mmol) and 305 mg of benzoic acid (2.5mmol) was treated in
15cm3 of toluene as described in the general procedure. The resulting crude product was chromato-
graphed on a silica gel column with CHCl3. The separation resulted in 83mg of 3 (31%), mp 108–
20
110ꢁC (Ref. [15] 107–109ꢁC), Rf ¼ 0.85 (ss A); ½ꢁꢃD ¼ ꢂ54 10ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3); 1H NMR
(400 MHz, CDCl3, Me4Si): ꢂ ¼ 1.01 (d, J ¼ 7.0 Hz, 17-CH3), 1.30–2.65 (overlapping multiplets, 13H),
2.90 (m, 6-H2), 3.78 (s, 3-OCH3), 6.66 (d, J ¼ 2.6 Hz, 4-H), 6.72 (dd, J ¼ 8.6, 2.6Hz, 2-H), 7.26 (d,
J ¼ 8.6 Hz, 1-H) ppm; 13C NMR (100 MHz, CDCl3, Me4Si): ꢂ ¼ 19.5 (17-CH3), 24.1, 26.9, 27.2, 30.1,
31.3, 32.1, 39.8 (C-8), 41.3 (C-9), 41.7 (C-17), 55.2 (3-OCH3), 111.2 (C-2), 113.9 (C-4), 125.9 (C-1),
133.0 (C-10), 136.9 (C-13), 138.3 (C-14), 138.9 (C-5), 157.5 (C-3) ppm. Continued elution resulted
20
in 148 mg of 2a (38%), mp 98–101ꢁC, Rf ¼ 0.60 (ss A); ½ꢁꢃD ¼ ꢂ1710ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3);
1H NMR (400MHz, CDCl3, Me4Si): ꢂ ¼ 0.86 (s, 18-H3), 1.30–2.40 (overlapping multiplets, 13H),
2.87 (m, 6-H2), 3.77 (s, 3-OCH3), 5.11 (d, J ¼ 6.0Hz, 17-H), 6.64 (d, J ¼ 2.6 Hz, 4-H), 6.71 (dd,
J ¼ 8.6, 2.6Hz, 2-H), 7.20 (d, J ¼ 8.6 Hz, 1-H), 7.45 (m, 30- and 50-H), 7.55 (t, J ¼ 7.3 Hz, 40-H), 8.06
(d, J ¼ 7.9 Hz, 20- and 60-H) ppm; 13C NMR (400MHz, CDCl3, Me4Si): ꢂ ¼ 16.8 (C-18), 24.5, 26.2,
28.1, 29.9, 30.3, 32.2, 39.1 (C-8), 43.7 (C-9), 45.4 (C-13), 49.5 (C-14), 55.2 (3-OCH3), 82.6 (C-17),
111.5 (C-2), 113.8 (C-4), 126.4 (C-1), 128.3 (2C, C-30 and C-50), 129.5 (2C, C-20 and C-60), 130.9
(C-10), 132.6 (C-10), 132.7 (C-40), 137.9 (C-5), 157.5 (C-3), 166.1 (C¼O) ppm. Further elution yielded
72mg of unreacted 1 (25%).
In chlorobenzene: A mixture of 286 mg of 1 (1mmol) and 305 mg of benzoic acid (2.5mmol) was
treated in 15cm3 of chlorobenzene as described in the general procedure. The resulting crude product
was chromatographed on a silica gel column with CHCl3. The separation resulted in 94 mg of 3 (35%),
117 mg of 2a (30%), and 92 mg of 1 (32%).
In HMPT: A mixture of 286 mg of 1 (1mmol) and 305 mg of benzoic acid (2.5mmol) was treated
in 15 cm3 of HMPT as described in the general procedure. The resulting crude product was chromato-
graphed on a silica gel column with CHCl3. The separation resulted in 32 mg of 3 (12%), 141 mg of 2a
(36%), and 137 mg of 1 (48%).
Mitsunobu Esterification of 1 with 4-Nitrobenzoic Acid
In toluene: A mixture of 286 mg of 1 (1mmol) and 418 mg of 4-nitrobenzoic acid (2.5mmol) was
treated in 15cm3 of toluene as described in the general procedure. The resulting crude product was
chromatographed on a silica gel column with CHCl3. The separation resulted in 16 mg of 3 (6%).
20
Continued elution resulted in 357 mg of 2b (82%), mp 154–155ꢁC, Rf ¼ 0.55 (ss A); ½ꢁꢃD
¼
1
ꢂ31 10ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3); H NMR (400 MHz, CDCl3, Me4Si): ꢂ ¼ 0.88 (s, 18-H3), 1.35–
1.85 (overlapping multiplets, 8H), 1.95 (m, 2H), 2.38 (m, 3H), 2.88 (m, 6-H2), 3.78 (s, 3-OCH3), 5.15
(d, J ¼ 6.1 Hz, 17-H), 6.64 (d, J ¼ 2.7 Hz, 4-H), 6.71 (dd, J ¼ 8.6, 2.7 Hz, 2-H), 7.20 (d, J ¼ 8.6 Hz,
1-H), 8.22 (d, J ¼ 8.8 Hz, 20- and 60-H), 8.30 (d, J ¼ 8.8 Hz, 30- and 50-H) ppm; 13C NMR (100MHz,
CDCl3, Me4Si): ꢂ ¼ 16.8 (C-18), 24.4, 26.1, 28.1, 29.9, 30.2, 32.2, 39.1 (C-8), 43.7 (C-9), 45.5 (C-13),
49.6 (C-14), 55.2 (3-OCH3), 83.9 (C-17), 111.5 (C-2), 113.8 (C-4), 123.5 (2C, C-30 and C-50), 126.3
(C-1), 130.6 (2C, C-20 and C-60), 132.3 (C-10), 136.2 (C-10), 137.9 (C-5), 150.5 (C-40), 157.5 (C-3),
164.2 (C¼O) ppm. Further elution yielded 28mg of unreacted 1 (10%).
In chlorobenzene: A mixture of 286 mg of 1 (1 mmol) and 418 mg of 4-nitrobenzoic acid
(2.5mmol) was treated in 15 cm3 of chlorobenzene as described in the general procedure. The resulting

´
P. Tapolcsanyi et al.
1134
crude product was chromatographed on a silica gel column with CHCl3. The separation resulted in
27mg of 3 (10%), 340 mg of 2b (78%), and 28mg of 1 (10%).
In HMPT: A mixture of 286 mg of 1 (1mmol) and 418 mg of 4-nitrobenzoic acid (2.5mmol) was
treated in 15cm3 of HMPT as described in the general procedure. The resulting crude product was
chromatographed on a silica gel column with CHCl3. The separation resulted in 13mg of 3 (5%),
270 mg of 2b (62%), and 74 mg of 1 (26%).
Mitsunobu Esterification of 1 with 3,5-Dinitrobenzoic Acid
In toluene: A mixture of 286 mg of 1 (1mmol) and 530 mg of 3,5-dinitrobenzoic acid (2.5mmol) was
treated in 15 cm3 of toluene as described in the general procedure. The resulting crude product
was chromatographed on a silica gel column with CHCl3. The separation resulted in 27mg of 3
20
(10%). Continued elution yielded 389 mg of 2c (81%), mp 174–175ꢁC, Rf ¼ 0.70 (ss A); ½ꢁꢃD
¼
1
ꢂ5 10ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3); H NMR (400MHz, CDCl3, Me4Si): ꢂ ¼ 0.90 (s, 18-H3), 1.30–
2.50 (overlapping multiplets, 13H), 2.88 (m, 6-H2), 3.77 (s, 3-OCH3), 5.23 (d, J ¼ 6.2 Hz, 17-H), 6.63
(d, J ¼ 2.7 Hz, 4-H), 6.69 (dd, J ¼ 8.6, 2.7 Hz, 2-H), 7.18 (d, J ¼ 8.6Hz, 1-H), 9.14 (d, J ¼ 2.3 Hz,
20- and 60-H), 9.22 (t, J ¼ 2.3 Hz, 40-H) ppm; 13C NMR (100MHz, CDCl3, Me4Si): ꢂ ¼ 16.7 (C-18),
24.5, 26.1, 28.0, 29.8, 30.1, 32.4, 39.1 (C-8), 43.6 (C-9), 45.7 (C-13), 49.6 (C-14), 55.2 (3-OCH3), 85.2
(C-17), 111.5 (C-2), 113.8 (C-4), 122.2 (C-40), 126.3 (C-1), 129.3 (2C, C-20 and C-60), 132.1 (C-10),
134.4 (C-10), 137.8 (C-5), 148.7 (2C, C-30 and C-50), 157.5 (C-3), 162.1 (C¼O) ppm. Further elution
yielded 14 mg of 1 (5%).
In chlorobenzene: A mixture of 286 mg of 1 (1mmol) and 530 mg of 3,5-dinitrobenzoic acid
(2.5mmol) was treated in 15 cm3 of chlorobenzene as described in the general procedure. The resulting
crude product was chromatographed on a silica gel column with CHCl3. The separation resulted in
27mg of 3 (10%) and 404 mg of 2c (84%).
Mitsunobu Esterification of 1 with 2,4-Dinitrobenzoic Acid
In toluene: A mixture of 286 mg of 1 (1mmol) and 530 mg of 2,4-dinitrobenzoic acid (2.5mmol) was
treated in 15 cm3 of toluene as described in the general procedure. The resulting crude product
was chromatographed on a silica gel column with CHCl3. The separation resulted in 8 mg of 3
20
(3%). Continued elution yielded 461 mg of 2d (96%), mp 86–89ꢁC, Rf ¼ 0.70 (ss A); ½ꢁꢃD
¼
1
ꢂ77 10ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3); H NMR (400 MHz, CDCl3, Me4Si): ꢂ ¼ 0.85 (s, 18-H3), 1.35–
2.40 (overlapping multiplets, 13H), 2.84 (m, 6-H2), 3.77 (s, 3-OCH3), 5.12 (d, J ¼ 5.9 Hz, 17-H), 6.62
(d, J ¼ 2.5 Hz, 4-H), 6.69 (dd, J ¼ 8.6, 2.5 Hz, 2-H), 7.18 (d, J ¼ 8.6Hz, 1-H), 7.98 (d, J ¼ 8.6 Hz,
60-H), 8.52 (dd, J ¼ 8.6, 2.2 Hz, 50-H), 8.73 (d, J ¼ 2.2 Hz, 30-H) ppm; 13C NMR (100 MHz, CDCl3,
Me4Si): ꢂ ¼ 16.6 (C-18), 24.2, 26.1, 28.0, 29.7, 29.8, 31.8, 39.0 (C-8), 43.4 (C-9), 45.3 (C-13), 49.1
(C-14), 55.2 (3-OCH3), 86.0 (C-17), 111.5 (C-2), 113.8 (C-4), 119.5 (C-30), 126.3 (C-1), 127.2 (C-50),
131.5 (C-60), 132.3 (C-10), 133.0 (C-10), 137.9 (C-5), 148.4 and 148.9 (2C, C-20 and C-40), 157.5 (C-3),
163.1 (C¼O) ppm.
In chlorobenzene: A mixture of 286 mg of 1 (1mmol) and 530 mg of 2,4-dinitrobenzoic acid
(2.5mmol) was treated in 15 cm3 of chlorobenzene as described in the general procedure. The resulting
crude product was chromatographed on a silica gel column with CHCl3, yielding 471 mg of 2d (98%).
In HMPT: A mixture of 286 mg of 1 (1mmol) and 530 mg of 2,4-dinitrobenzoic acid (2.5 mmol) was
treated in 15cm3 of HMPT as described in the general procedure. The resulting crude product was
chromatographed on a silica gel column with CHCl3, yielding 355 mg of 2d (74%) and 34mg of 1 (12%).
Mitsunobu Esterification of 2 with 4-Nitrobenzoic Acid
In toluene: A mixture of 286 mg of 2 (1mmol) and 418 mg of 4-nitrobenzoic acid (2.5mmol) was
treated in 15cm3 of toluene as described in the general procedure. The resulting crude product was

The Mitsunobu Inversion Reaction
1135
chromatographed on a silica gel column with CHCl3. The separation resulted in 190 mg of 3
20
(71%). Continued elution yielded 52 mg of 1a (12%), mp 159–162ꢁC, Rf ¼ 0.65 (ss A); ½ꢁꢃD
¼
1
þ86 10ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3); H NMR (400 MHz, CDCl3, Me4Si): ꢂ ¼ 0.99 (s, 18-H3), 1.30–
2.50 (overlapping multiplets, 13H), 2.89 (m, 6-H2), 3.78 (s, 3-OCH3), 4.97 (m, 17-H), 6.64 (d,
J ¼ 2.7 Hz, 4-H), 6.71 (dd, J ¼ 8.6, 2.7 Hz, 2-H), 7.20 (d, J ¼ 8.6Hz, 1-H), 8.21 (d, J ¼ 8.8 Hz,
20- and 60-H), 8.29 (d, J ¼ 8.8 Hz, 30- and 50-H) ppm; 13C NMR (100MHz, CDCl3, Me4Si): ꢂ ¼ 12.4
(C-18), 23.4, 26.2, 27.2, 27.7, 29.8, 37.0, 38.6 (C-8), 43.4 (C-13), 43.8 (C-9), 49.8 (C-14), 55.2
(3-OCH3), 84.3 (C-17), 111.5 (C-2), 113.8 (C-4), 123.5 (2C, C-30 and C-50), 126.3 (C-1), 130.6
(2C, C-20 and C-60), 132.3 (C-10), 136.1 (C-10), 137.8 (C-5), 150.4 (C-40), 157.5 (C-3), 164.6
(C¼O) ppm. Further elution yielded 43 mg of unreacted 2 (15%).
In chlorobenzene: A mixture of 286 mg of 2 (1 mmol) and 418 mg of 4-nitrobenzoic acid
(2.5mmol) was treated in 15 cm3 of chlorobenzene as described in the general procedure. The resulting
crude product was chromatographed on a silica gel column with CHCl3, yielding 204 mg of 3 (76%),
52mg of 1a (12%), and 25 mg of 2 (9%).
Mitsunobu Reaction of 4 with 4-Nitrobenzoic Acid
A mixture of 300 mg of 4 (1mmol) and 418mg of 4-nitrobenzoic acid (2.5mmol) was treated in
15cm3 of toluene as described in the general procedure. The resulting crude product was chromato-
graphed on a silica gel column with CHCl3. The separation resulted in 42 mg of 7 (15%), mp 95–98ꢁC,
20
1
Rf ¼ 0.85 (ss A); ½ꢁꢃD ¼ þ86 10ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3); H NMR (400MHz, CDCl3, Me4Si):
ꢂ ¼ 0.78 (s, 18-H3), 1.73 (s, 16-CH3), 1.40–2.35 (overlapping multiplets, 11H), 2.87 (m, 6-H2), 3.77
(s, 3-OCH3), 5.48 (s, 17-H), 6.63 (d, J ¼ 2.6 Hz, 4-H), 6.70 (dd, J ¼ 8.6, 2.6 Hz, 2-H), 7.17 (d,
J ¼ 8.6 Hz, 1-H) ppm; 13C NMR (100 MHz, CDCl3, Me4Si): ꢂ ¼ 17.3 and 17.6 (C-18 and 16-CH3),
26.6, 28.0, 29.8, 36.3 (2C), 37.4 (C-8), 44.6 (C-9), 46.1 (C-13), 55.2 (3-OCH3), 56.0 (C-14), 111.3 (C-
2), 113.8 (C-4), 126.0 (C-1), 133.2 (C-10), 137.3 (C-17), 138.0 (C-5), 139.4 (C-16), 157.4 (C-3) ppm.
20
Further elution yielded 76 mg of 6a (17%), mp 145–147ꢁC, Rf ¼ 0.65 (ss A); ½ꢁꢃD ¼ ꢂ4 10ꢂ1 ꢁcm2 gꢂ1
(c ¼ 1, CHCl3); 1H NMR (400 MHz, CDCl3, Me4Si): ꢂ ¼ 0.94 (s, 18-H3), 1.32 (d, J ¼ 7.0 Hz, 16-CH3),
1.10–2.40 (overlapping multiplets, 12H), 2.88 (m, 6-H2), 3.78 (s, 3-OCH3), 4.77 (s, 17-H), 6.64 (d,
J ¼ 2.7 Hz, 4-H), 6.71 (dd, J ¼ 8.6, 2.7 Hz, 2-H), 7.19 (d, J ¼ 8.6 Hz, 1-H), 8.22 (d, J ¼ 8.8 Hz, 20- and
60-H), 8.30 (d, J ¼ 8.8 Hz, 30- and 50-H) ppm; 13C NMR (100MHz, CDCl3, Me4Si): ꢂ ¼ 17.4 (C-18),
20.8 (16-CH3), 25.9, 28.0, 29.9, 32.7, 34.4, 38.8 (C-8), 41.2 (C-16), 43.5 (C-9), 45.3 (C-13), 51.0 (C-
14), 55.2 (3-OCH3), 90.4 (C-17), 111.5 (C-2), 113.8 (C-4), 123.6 (2C, C-30 and C-50), 126.3 (C-1),
130.6 (2C, C-20 and C-60), 132.3 (C-10), 136.2 (C-10), 137.9 (C-5), 150.4 (C-40), 157.5 (C-3), 164.4
(C¼O) ppm. Further elution resulted in 195mg of 4 (65%).
3-Methoxy-16ꢀ-methylestra-1,3,5(10)-trien-17ꢁ-ol (6)
Compound 6a (45 mg, 0.1 mmol) was dissolved in 5 cm3 of methanol containing 10mg of NaOCH3
(0.184 mmol) and the solution was refluxed for 1 h, than diluted with 20cm3 of H2O and extracted with
3ꢄ10 cm3 of CH2Cl2. The CH2Cl2 solution was washed with H2O, dried, and evaporated. The residual
oil was chromatographed on a silica gel column with a mixture of ethyl acetate=CHCl3 (5=95). The
product was crystallized from CHCl3=petrolether. This yielded 31mg of 6 (70%), mp 52–54ꢁC (Ref.
20
[18] 53–56ꢁC), Rf ¼ 0.40 (ss B); ½ꢁꢃD ¼ þ6010ꢂ1 ꢁcm2 gꢂ1 (c ¼ 1, CHCl3).
Mitsunobu Reaction of 5 with 4-Nitrobenzoic Acid
A mixture of 300 mg of 5 (1mmol) and 418mg of 4-nitrobenzoic acid (2.5mmol) was treated in
15cm3 of toluene as described in the general procedure. The resulting crude product was chromato-
graphed on a silica gel column with CHCl3. The separation resulted in 42mg of 7 (15%). Continued

´
P. Tapolcsanyi et al.: The Mitsunobu Inversion Reaction
1136
20
elution yielded 270 mg of 8a (60%), mp 161–163ꢁC, Rf ¼ 0.70 (ss A); ½ꢁꢃD ¼ þ1310ꢂ1 ꢁcm2 gꢂ1
(c ¼ 1, CHCl3); 1H NMR (400 MHz, CDCl3, Me4Si): ꢂ ¼ 0.94 (s, 18-H3), 1.02 (d, J ¼ 7.2 Hz, 16-CH3),
1.40–2.40 (overlapping multiplets, 11H), 2.71 (m, 1H), 2.88 (m, 6-H2), 3.77 (s, 3-OCH3), 5.22 (d,
J ¼ 5.8 Hz, 17-H), 6.64 (d, J ¼ 2.6Hz, 4-H), 6.70 (dd, J ¼ 8.6, 2.6 Hz, 2-H), 7.17 (d, J ¼ 8.6 Hz, 1-H),
8.24 (d, J ¼ 8.8 Hz, 20- and 60-H), 8.31 (d, J ¼ 8.8 Hz, 30- and 50-H) ppm; 13C NMR (100MHz, CDCl3,
Me4Si): ꢂ ¼ 16.0 (C-18), 17.2 (16-CH3), 26.0, 28.0, 29.9, 32.3, 33.5, 34.3 (C-16), 39.0 (C-8), 43.7 (C-
9), 46.7 (C-13), 48.8 (C-14), 55.2 (3-OCH3), 85.3 (C-17), 111.5 (C-2), 113.8 (C-4), 123.6 (2C, C-30
and C-50), 126.3 (C-1), 130.6 (2C, C-20 and C-60), 132.3 (C-10), 135.9 (C-10), 137.8 (C-5), 150.5 (C-
40), 157.5 (C-3), 164.3 (C¼O) ppm. Further elution resulted in 63mg of unreacted 5 (21%).
3-Methoxy-16ꢁ-methylestra-1,3,5(10)-trien-17ꢁ-ol (8)
Compound 8a (90 mg, 0.2mmol) was dissolved in 10cm3 of MeOH containing 15mg of NaOCH3
(0.276 mmol) and the solution was refluxed for 1 h, then diluted with 30cm3 of H2O and extracted with
3ꢄ15 cm3 of CH2Cl2. The CH2Cl2 solution was washed with H2O, dried, and evaporated. The resulting
crude product was chromatographed on a silica gel column with a mixture of ethyl acetate=CHCl3
(5=95). The product was crystallized from CHCl3=petrolether, yielding 44mg of 8 (73%), mp 126–
127ꢁC (Ref. [18] 125–127ꢁC), Rf ¼ 0.55 (ss B).
Acknowledgements
We thank the Hungarian Scientific Research Fund (OTKAT042673) for financial support of this work.
´
We also thank Dr. M. Rꢀozsa-Tarjani (University of Szeged, Hungary) for the NMR spectra.
References
[1] Mitsunobu O, Yamada M (1967) Bull Chem Soc Japan 40: 2380
[2] Mitsunobu O, Eguchi M (1971) Bull Chem Soc Japan 44: 3427
[3] Mitsunobu O (1981) Synthesis 1
[4] Brandstetter HH, Zbiral E (1978) Helv Chim Acta 61: 1832
[5] Mitsunobu O, Kimura J, Fujisawa Y (1972) Bull Chem Soc Japan 45: 245
[6] Jarosz S, Glodek J, Zamojski A (1987) Carbohydr Res 163: 289
[7] Simon C, Hosztafi S, Makleit S (1997) J Heterocyclic Chem 34: 349
[8] Eaton PE, Jobe PG, Reingold ID (1984) J Am Chem Soc 106: 6437
[9] Kocienski PJ, Street SDA, Yeates C, Campbell SF (1987) J Chem Soc Perkin Trans 1: 2171
[10] Mori K, Ikunama I (1984) Tetrahedron 40: 3471
[11] Mori K, Otsuka T, Oda M (1984) Tetrahedron 40: 2929
[12] Martin SF, Dodge JA (1991) Tetrahedron 32: 3017
[13] Dodge JA, Trujillo JI, Presnell M (1994) J Org Chem 59: 234
[14] Dodge JA, Lugar CW (1996) Bioorg Med Chem Lett 6: 1
[15] Johns WF (1961) J Org Chem 26: 4583
´
[16] Schneider Gy, Hackler L, Sohar P (1988) Liebigs Ann Chem 679
[17] Evans PA, Nelson JD, Rheinhold AL (1997) Tetrahedron Lett 38: 2235
ꢀ
[18] Mesko E, Schneider Gy, Dombi Gy, Zeigan D (1990) Liebigs Ann Chem 419