Alcohol inversion using cesium carboxylates and DMAP in toluene

Full text of DOI:10.1021/jo0007783

J . Org. Chem. 2000, 65, 8379-8380
8379
Ta ble 1. Rea ction of 3â-Ch olesta n yl Mesyla te (3) w ith 5
Equ iv of CsOAc in Tolu en e a t Reflu x for 4 Da ys a n d
Am in e Ca ta lysts
Alcoh ol In ver sion Usin g Cesiu m
Ca r boxyla tes a n d DMAP in Tolu en e
% acetate % alkenes % mesylate
Natalie A. Hawryluk and Barry B. Snider*
entry
amine
equiv
4
5
3
Department of Chemistry, Brandeis University,
Waltham, Massachusetts 02454-9110
1
2
3
4
5
6
18-crown-6
DMAP
DMAP
DMAP
DMAP
0.5
0.1
0.25
0.5
0.75
92a
15
8a
7
10
8a
10
0
78
13
14a
13
4
76
75a
76
snider@brandeis.edu
Received May 22, 2000
4-pyrrolidino- 0.5
pyridine
83
13
7
4-(4-methyl-
piperidino)-
pyridine
TMEDA
pyridine
0.5
85
11
4
Inversion of secondary alcohols by the SN2 reaction of
a cesium carboxylate with a secondary sulfonate in DMF
has been widely used in synthesis.1,2 Because removal of
DMF can be difficult, alternate procedures using benzene
or toluene as the solvent with 0.5 equiv of 18-crown-6 to
solubilize potassium and cesium carboxylates have also
been extensively used.3 During our synthesis of dysi-
herbaine,4 we found that that the SN2 reaction of CsOAc
(5 equiv) with crude nitrobenzenesulfonate 1 proceeded
cleanly in toluene at reflux for 12 h to give 75-85% of 2
without the use of 18-crown-6.4 Surprisingly, no reaction
occurred under the same conditions with pure 1. Since 1
was prepared from the alcohol with nitrobenzenesulfonyl
chloride, Et3N, and DMAP in CH2Cl2, we suspected that
residual DMAP present in crude 1 was necessary for the
SN2 reaction with CsOAc in toluene at reflux. This was
confirmed by heating a suspension of CsOAc (5 equiv) in
a toluene solution of pure nosylate 1 (0.13 M) and DMAP
(0.1 equiv) at reflux for 8 h to give 85% of acetate 2.
8
9
10
11
0.5
0.5
0.5
0.5
0
0
0
0
0
0
0
0
99
99
99
95
bipyridine
DABCO
a
Isolated yield. Other yields were determined by analysis of
NMR spectra.
DMAP, or other amine additives as shown in Table 1.
Use of 0.5 equiv of 18-crown-6 afforded 92% of 3-R-
cholestanyl acetate (4a ) and 8% of a mixture of 2- and
3-cholestenes (5) (entry 1).3j Use of 0.5 equiv of DMAP
afforded 76% of 3-R-acetate 4a , 8% of alkenes 5, and 13%
of recovered mesylate 3 (entry 4). Similar results were
obtained with 0.25 and 0.75 equiv of DMAP (entries 3
and 5), while reaction was very slow with only 0.1 equiv
of DMAP (entry 2). Heating mesylate 3 and 0.5 equiv of
DMAP without CsOAc gave 2% of alkenes 5, 5% of 3-R-
cholestanyl mesylate (6), 93% of recovered 3, and none
of the known DMAP alkylation product.5 Other 4-(di-
alkylamino)pyridines such as 4-pyrrolidino- and 4-(4-
methylpiperidino)pyridine are as effective as DMAP at
facilitating the SN2 reaction (entries 6, 7).
We decided to explore the generality of DMAP as an
alternative to 18-crown-6 for the SN2 reactions of cesium
carboxylates in toluene at reflux since DMAP is less
expensive than 18-crown-6 and is easy to remove and
recycle by extraction into dilute acid. The inversion of
3â-cholestanyl mesylate (3) was chosen for initial studies
since the starting material is readily available and both
substitution and elimination products can be easily
characterized. Mesylate 3 and 5 equiv of CsOAc were
heated in toluene at reflux for 4 d with 18-crown-6,
* Corresponding author. Phone: 781-736-2550. Fax: 781-736-21516.
(1) (a) Kruizinga, W. H.; Strijtveen, B.; Kellogg, R. M. J . Org. Chem.
1981, 46, 4321. (b) Huffman, J . W.; Desai, R. C. Synth. Commun. 1983,
13, 553.
(2) For discussions of the “cesium effect”, see: (a) Dijkstra, G.;
Kruizinga, W. H.; Kellogg, R. M. J . Org. Chem. 1987, 52, 4230. (b)
Salvatore, R. N.; Nagle, A. S.; Schmidt, S. E.; J ung. K. W. Org. Lett.
1999, 1, 1893.
We hypothesized that DMAP forms a toluene-soluble
complex with CsOAc leading to a DMAP-complexed
cesium cation and a reactive, uncomplexed acetate anion.
We therefore investigated TMEDA since it could solubi-
lize the cesium cation by chelation. No reaction occurred
with TMEDA (entry 8); pyridine, bipyridine, and DABCO
were equally ineffective (entries 9-11). These results
suggest that only the strongly basic but unhindered
pyridine nitrogen of 4-(dialkylamino)pyridines forms a
(3) (a) Liotta, C. L.; Harris, H. P.; McDermott, M.; Gonzalez, T.;
Smith, K. Tetrahedron Lett. 1974, 2417. (b) Torisawa, Y.; Okabe, H.;
Ikegami, S. Chem. Lett. 1984, 1555. (c) Willis, C. L. Tetrahedron Lett.
1987, 28, 6705. (d) Yan, J .; Bittman, R. J . Lipid Res. 1990, 31, 160. (e)
Akiyama, T., Takechi, N.; Ozaki, S.; Shiota, K. Bull. Chem. Soc. J pn.
1992, 65, 366. (f) Senanayake, C. H.; Singh, S. B.; Bill, T. J .; DiMichele,
L. M.; Liu, J .; Larsen, R. D.; Verhoeven, T. R. Tetrahedron Lett. 1993,
34, 2425. (g) Sato, K.-i.; Yoshitomo, A. Chem. Lett. 1995, 39. (h) Sato,
K.-i.; Suzuki, K.; Hashimoto, Y. Chem. Lett. 1995, 83. (i) Shimizu, T.;
Hiranuma, S.; Nakata, T. Tetrahedron Lett. 1996, 37, 6145. (j) Sato,
K.-i.; Yoshitomo, A.; Takai, Y. Bull. Chem. Soc. J pn. 1997, 70, 885.
(4) Snider, B. B.; Hawryluk, N. A. Org. Lett. 2000, 2, 635.
(5) Bennua-Skalmowski, B.; Krolikiewicz, K.; Vorbrueggen, H. Bull.
Soc. Chim. Belg. 1994, 103, 453.
10.1021/jo0007783 CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/25/2000

8380 J . Org. Chem., Vol. 65, No. 24, 2000
Notes
Ta ble 2. Rea ction of Secon d a r y Mesyla tes w ith 5 Equ iv
of CsOAc or CsOBz a n d 0.5 Equ iv of DMAP in Tolu en e a t
Reflu x
(entry 7) and 79% of benzoate 11b (entry 8). CsOBz is
not only more reactive than CsOAc, but also more
selective for substitution rather than elimination with
mesylates 3, 8, and 10, since the mass not accounted for
in entries 5-8 is volatile alkene that was lost on workup.
CsOCOR, time,
ester
(%)a
alkene mesylate
entry substrate
R )
d
(%)a
(%)a
1
2
3
4
5
6
7
8
3
3
6
6
8
8
10
10
Me
Ph
Me
Ph
Me
Ph
Me
Ph
4
1
1
1
8
6
2
2
4a (75) 5 (8)
4b (91) 5 (9)
7a (68) 5 (32)
7b (68) 5 (30)
14
0
0
0
9a (20)
9b (63)
b
b
b
b
20
20
0
11a (36)
11b (79)
0
a
Isolated yield. bThe unaccounted for material is the volatile
alkene.
soluble complex with CsOAc in toluene at reflux, as does
18-crown-6, permitting the SN2 reaction to occur in
excellent yield. We examined the use of DME as a polar
solvent that could chelate to CsOAc. Unfortunately,
heating 3 with 5 equiv of CsOAc in DME at reflux for 4
d afforded a complex mixture of products, presumably
resulting from SN1 reaction.
As expected, more hydrophobic carboxylates are more
reactive than acetate. Use of 5 equiv of CsOBz1 and 0.5
equiv of DMAP in toluene at reflux for 20 h afforded 91%
of 3-R-benzoate 4b and 9% of alkenes 5. No reaction
occurred without DMAP after 4 d. Use of cesium palmi-
tate1 with 0.5 equiv of DMAP for 20 h or without DMAP
for 4 d afforded 90% of 3-R-palmitate 4c and 9% of
alkenes 5. However, the palmitate forms intractable
soapy emulsions on workup. As expected, cesium formate
was less reactive, providing only 13% of 3-R-formate 4d ,
7% of alkenes 5, and 80% of recovered 3 after reflux for
4 d with DMAP.
In conclusion, we have shown that DMAP catalyzes the
displacement of secondary sulfonates by cesium carboxy-
lates in toluene at reflux offering a practical and inex-
pensive alternative to 18-crown-6. The novel use of
DMAP to solubilize a metal carboxylate in toluene may
be of general utility in organic synthesis.
The DMAP-catalyzed reaction of CsOAc and CsOBz
with other mesylates was investigated to determine the
scope of the reaction as shown in Table 2. 3-R-Cholestanyl
mesylate (6) was more reactive than the â-isomer 3
requiring only 1 d for complete reaction, even with the
less reactive CsOAc (entries 3 and 4), but giving more
alkenes 5 as expected for an axial mesylate, as was also
observed with 18-crown-6, which gave 82% of acetate 7
and 10% of alkenes 5.3i Hindered menthyl mesylate (8)
afforded only 20% of acetate 9a after 8 d at reflux (entry
5). CsOBz is much more effective giving 63% of benzoate
9b after 6 d at reflux (entry 6). Similarly, trans-tert-
butylcyclohexyl mesylate 10 afforded 36% of acetate 11a
Exp er im en ta l Section
Gen er a l P r oced u r e for In ver sion of Mesyla tes. A 0.3 M
solution of mesylate (1 equiv), cesium carboxylate (5 equiv), and
DMAP (0.5 equiv) in dry toluene was heated at reflux for the
indicated time. The mixture was concentrated and diluted with
EtOAc. The EtOAc solution was washed with water, 1 M HCl,
and brine, dried (Na2SO4), and concentrated to give crude ester.
Purification by flash chromatography on silica gel using gradient
elution with 10:1 hexane/EtOAc to 6:1 hexane/EtOAc gave
alkene, followed by ester and unreacted mesylate in the yields
indicated in Tables 1 and 2. The data for 1,4 2,4 3,3i 4a ,3i 4b,6
4d ,7 5,6 6,3i 7a ,3i 7b,6 8,3i 9a ,3i 9b,6 10,1b 11a ,1b and 11b6 are
identical to those of authentic samples.
Ack n ow led gm en t. We are grateful to the NIH (GM-
46470) for generous support of this research.
(6) Barrett, A. G. M.; Braddock, D. C.; J ames, R. A.; Koike, N.;
Procopiou, P. A. J . Org. Chem. 1998, 63, 6273.
(7) Bose, A. K.; Lal, B.; Hoffman, W. A., III; Manhas, M. S.
Tetrahedron Lett. 1973, 1619.
J O0007783