Diastereoselective free radical halogenation, azidation, and rearrangement of β-silyl Barton esters

Article abstract of DOI:10.1021/ol0268182

(equation presented) Barton esters of β-silylcarboxylic acids were decomposed by photolysis alone in organic solvents or in the presence of ethanesulfonyl azide or bromotrichloromethane. Products of the reaction, β-silylthiopyridyl ethers, β-silyl azides, or alkenes, were formed with significant control of stereochemistry.

Full text of DOI:10.1021/ol0268182

ORGANIC
LETTERS
2
002
Vol. 4, No. 24
253-4256
Diastereoselective Free Radical
Halogenation, Azidation, and
4
Rearrangement of â-Silyl Barton Esters
Douglas S. Masterson and Ned A. Porter*
Department of Chemistry, Vanderbilt UniVersity, NashVille, Tennessee 37235
n.porter@Vanderbilt.edu
Received August 29, 2002
ABSTRACT
Barton esters of â-silylcarboxylic acids were decomposed by photolysis alone in organic solvents or in the presence of ethanesulfonyl azide
or bromotrichloromethane. Products of the reaction, â-silylthiopyridyl ethers, â-silyl azides, or alkenes, were formed with significant control
of stereochemistry.
The attention to free radical chemistry over the past decade
has been due in part to the functional group tolerance of
free radical reactions and to the ability to control stereo-
chemistry in radical transformations.1,2 Recently we reported
on the stereoselective formation and elimination of â-bro-
mosilanes in Lewis acid promoted atom transfer additions
to branched allylsilanes.3 The resulting â-bromosilanes
underwent a nonradical elimination process that gave the
thermodynamically less stable cis olefins in good yield. The
configuration of the product olefins in these reactions allowed
for the analysis of the stereochemical outcome in the atom
are typically initiated by photolysis, thus providing a mild
source of the desired alkyl radical species. We report here
that â-silyl Barton esters can be readily prepared and used
in free radical halogenation, azidation, and thiopyridyl
rearrangement reactions with moderate to excellent diaste-
reocontrol and yield (Scheme 1).
Scheme 1
4
transfer reaction. Anti-elimination of bromosilane yields cis
olefin products when the â-bromosilane has a threo config-
uration, a result of stereochemical control in the bromine
atom transfer.
Thiohydroxamate esters, 1, (Barton esters) are widely used
as sources of alkyl radicals5 because they are easily prepared
and provide ready entry to a variety of radical precursors
from activated carboxylic acids. Reactions of Barton esters
,6
(
1) Baguley, P. A.; Walton, J. C. Angew. Chem., Int. Ed. 1998, 37, 3072-
082.
2) Curran, D. P.; Porter, N. A.; Giese, B. Stereochemistry of Radical
Reactions; VCH: Weinheim, 1996.
3) Porter, N. A.; Zhang, G.; Reed, A. D. Tetrahedron Lett. 2000, 41,
773-5777.
4) Jarvie, A. W. P.; Holt, A.; Thompson, J. J. Chem. Soc. B 1969, 852-
55.
3
(
(
5
8
The thiohydroxamate esters were readily prepared by
reaction of activated and appropriately substituted â-silyl
carboxylic acids with the sodium salt of 2-mercaptopyridine-
(
(
5) Barton, D. H. R.; Zard, S. Z. Pure Appl. Chem. 1986, 58, 675-684.
(6) Barton, D. H. R.; Bridon, D.; Fernandez-Picot, I.; Zard, S. Z.
7
Tetrahedron 1987, 43, 2733-2740.
0.1021/ol0268182 CCC: $22.00 © 2002 American Chemical Society
N-oxide. The carboxylic acids were prepared using Flem-
1
Published on Web 10/30/2002

Scheme 2
Table 1. Formation of Thiopyridyls from Barton Esters
entry
substratea
4 drb,c
alkened E/Ze
1
2
3
4
5
6
3a
3b
3c
3d
3d
3e
80:20
80:20
72:28
87:13
83:17g
91:9h
24:76
16:84
ndf
90:10
90:10
79:21
a
b
Reaction of 3 at -10 °C in CH2Cl2 initiated by light. Diastereomeric
1
c
product ratio determined by H NMR of the crude product mixture. Isolated
d
yields by preparative TLC, 40% for reactions at -10 °C. Prepared by
TBAF/THF treatment of the diastereomeric product mixture 4 at room
temperature. e Determined by gas chromatography by comparison with
f
authentic standards. Not determined because of the volatility of alkene
g
products. Reaction at room temperature, isolated yield by preparative TLC,
8
0%. h Isolated yield, 20% by TLC.
8
ing’s procedure as outlined in Scheme 2. Acids 2a-e were
transformed to the corresponding thiohydroxamate esters by
conversion to the acid chloride followed by reaction with
the 2-mercaptopyridine-N-oxide salt in methylene chloride
to be the major geometrical isomer present (Table 1, entries
and 2). Formation of the Z isomer can be rationalized
assuming an E2 elimination pathway for the TBAF-induced
elimination process from threo-4a (Scheme 3b).
1
(reaction, workup, and purification were performed in the
dark) (Scheme 2). Compounds 3a-e were obtained in good
crude yields and were isolated by recrystallization or column
When the R group in 3 was a simple alkyl group (entries
2
and 3) or cycloalkyl (entry 1), the diastereomeric ratio (dr)
chromatography (SiO
NMR.
We chose the thiopyridyl rearrangement reaction of 3a-
e, as illustrated in Scheme 3 for 3a, as an initial test of
2
) diastereomerically pure as judged by
of the thiopyridyl products was on the order of 80:20. Isolated
yields were generally poor to moderate for this transforma-
tion, 30-40%, for the reactions carried out at -10 °C. When
R in 3 was a phenyl group (3d and 3e in Table 1, entries
9
4
-6), the dr observed for 4 is greater than that observed
when R is alkyl or cycloalkyl. When 3d was irradiated at
room temperature, an isolated yield of 4d of 80% was
obtained for the thiopyridyl product. Stereoselectivity for this
transformation dropped from 87:13 at -10 °C to 83:17 at
room temperature. The highest level of diastereoselectivity
was achieved with 3e, which gave a dr of 91:9 at -10 °C
Scheme 3
(entry 6) but a poor yield of 20%.
When the thiopyridyl products 4d and 4e were subjected
to TBAF elimination, the E isomer was the major olefin
formed. This suggests that either the stereochemical course
for reaction of 3d and 3e differed from the other compounds
studied or that the TBAF elimination did not proceed by a
1
0
strict E2 elimination process. To address this point, the
major and minor diastereomers of 4d were isolated by
chromatography. The isolated isomers were then separately
subjected to TBAF elimination, and the olefin products were
analyzed by GC and compared with authentic samples. The
products derived from the minor isomer of 4d were a >99:1
E/Z olefin mixture, and the olefins obtained from the major
diastereocontrol. A methylene chloride solution of the Barton
ester was irradiated until the solution became practically
colorless (usually 3-4 h irradiation time). The solvent was
then removed under reduced pressure, the thiopyridyl
products were isolated by preparative TLC, and NMR
analysis was conducted to determine the diastereomeric ratio
of the resulting products. In most cases, the diastereomeric
ratio was >80:20 (Table 1). When 4a was treated with a
solution of TBAF, elimination took place, which smoothly
produced olefins, whose E/Z ratio (GC) was in excellent
agreement with the diastereomeric ratio of the thiopyridyl
products as determined by NMR. The olefin mixture was
compared to authentic samples and the Z isomer was found
(
threo) isomer of 4d were an 82:18 E/Z mixture. This result
11
can be rationalized by an E1cb mechanism for the TBAF-
induced elimination from 4d and 4e, a conclusion consistent
(
7) Crich, D. Aldrichimica Acta 1987, 20, 35-42.
(8) Fleming, I.; Hill, J. H. M.; Parker, D.; Waterson, D. J. Chem. Soc.,
Chem. Commun. 1985, 318-321.
(9) Barton, D. H. R.; Jaszberenyi, J. C.; Theodorakis, E. A.; Reibenspies,
J. H. J. Am. Chem. Soc. 1993, 115, 8050-8059.
(10) Bernhard, W.; Fleming, I. J. Organomet. Chem. 1984, 271, 281-
88.
2
(11) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry; Harper & Row: New York, 1987.
4254
Org. Lett., Vol. 4, No. 24, 2002

with observations reported by Fleming.10 We conclude that
the results from all of the transformations are consistent; threo
is the major product formed in every case.
Attempts were made to promote an E2 elimination process
from 4d and 4e. The use of a nonpolar solvent (hexane) and
lower reaction temperatures (-10 and -78 °C) in the
elimination reaction gave an increase in Z olefin compared
to reactions carried out at room temperature in THF, but the
olefin dr never approached the ratio expected for an E2
process.
Table 2. Formation of Azides from Barton Esters
entry
substratea
6 drb,c
olefind E/Ze
1
2
3
4
5
3a
3b
3c
3d
3e
60:40f
76:24g
67:33
92:8i
42:58
24:76
ndh
81:19
89:11
91:9
a
b
Reaction of 3 at -10 °C in CH2Cl2 initiated by light. Diastereomeric
1
c
product ratio determined by H NMR of the crude product mixture. Isolated
Free radical azidation of alkyl iodides takes place readily
d
yields by preparative TLC, 30-45% for reactions at -10 °C. Prepared
by TBAF treatment of the diastereomeric product mixture 6. Determined
e
in the presence of ethanesulfonyl azide (5) to give good
f
yields of alkyl azides.1
2,13
by gas chromatography by comparison with authentic standards. Yield of
To date, there have been no studies
1
4
4% determined by H NMR of crude product compared to an internal
g
of acyclic stereocontrol in these free radical azidation
processes, and we used 3a-e to explore the formation of
alkyl azides where stereochemistry was an issue (Scheme
standard. Ratio determined by TBAF elimination because of overlap of
signals in the NMR. Not determined because of volatility of alkene
products. Yield of 56% determined by H NMR of crude product compared
to an internal standard. Yield of 66% by the slow infusion method described
in Supporting Information.
h
i
1
4).
thiopyridyl rearrangement product, 4, was produced in the
reactions with ethanesulfonyl azide. For 3d, NMR indicated
a 56% yield of azide 6d with the thiopyridyl products 4d
being formed in 40%. Significant quantities of thiopyridyl
products are formed in all of the azidation reactions, and
product accountability is generally in excess of 80%.
It is likely that at lower reaction temperatures, the extrusion
Scheme 4
2
of SO from the ethylsulfonyl radical is slow. This assump-
The azidation of 3a-e was performed in a manner as
tion is supported by the observation that NMR and GC-
MS analysis showed that no 2-ethylthiopyridine was pro-
duced, as expected if ethyl radical chain propagation had
occurred. Because both azide and thiopyridyl products are
observed, we suggest a mechanism for the transformation
as shown in Scheme 5.
reported for alkyl iodides1 except that no external initiator
was employed. A methylene chloride solution of the thio-
hydroxamate ester was irradiated in the presence of 3-5
equiv of 5 under an inert atmosphere until the solution
became practically colorless. The resulting mixture was
concentrated, and the residue was purified by preparative
TLC. Table 2 shows that the course of the azidation of 3a-e
appears to be highly substrate-dependent. Thus, 3a gave a
mixture of alkyl azides 6a with a dr of 60:40 (Table 2 entry
2,13
Scheme 5
1), and when this product diastereomer mixture was subjected
to treatment with TBAF, elimination of the elements of silyl
azide took place with the resulting alkenes being produced
in an E/Z ratio of 42:58. The excellent agreement between
the GC analysis of the olefins and the NMR analysis of the
azide mixture 6a allowed the TBAF elimination to be used
as a diagnostic tool for the determination of relative
configuration of the product alkyl azides. Diastereoselec-
tivities for the reaction improved substantially for 3d and
3
3
e as illustrated in entries 4 and 5 of Table 2. Azidation of
d gave an alkyl azide mixture with a dr of 92:8, and 3e
provided an alkyl azide mixture with a dr of 91:9. As was
the case with the thiopyridyl products 4d and 4e, the azides
6
d and 6e did not produce olefins whose isomeric composi-
NMR and mass spectral analysis of the crude reaction
mixtures indicated the presence of ethyl disulfone 7 (Scheme
tion was consistent with that of the parent azides. The TBAF
elimination results for substrates 6d and 6e suggest an E1cb
elimination mechanism, a result that is consistent with the
elimination reaction of thiopyridyls 4d and 4e.
5
), a product that is formed in yields comparable to that of
the azide. The presence of 7 and the absence of the
-ethylthiopyridine in the product mixtures indicates that
ethyl radicals are not available for chain propagation. When
the azidation of 3d was initiated with Et B/O , both azide
2
1
H NMR analysis of the crude reaction mixtures showed
that, in addition to azide products 6, a substantial amount of
3
2
and thiopyridyl products were observed in a ca. 1:1 ratio
along with substantial amounts of 2-ethylthiopyridine.
(
(
12) Ollivier, C.; Renaud, P. J. Am. Chem. Soc. 2000, 122, 6496-6497.
13) Ollivier, C.; Renaud, P. J. Am. Chem. Soc. 2001, 123, 4717-4727.
Org. Lett., Vol. 4, No. 24, 2002
4255

substrate 3b. Substrate 3d and 3e gave olefin with high E
selectivity (E/Z ) 92:8 and 95:5) with excellent product
yields (80%). We suspect that substrates 3d and 3e when
halogenated may undergo a stepwise elimination of PhMe -
2
SiBr, accounting for the E selectivity in these reactions.
Table 3. Formation of Alkenes from Reaction of Barton Esters
with BrCCl
3
entry
substratea,b
olefin E/Zc
1
2
3
4
3a
3b
3d
3e
38:62
33:67
92:8
The results obtained for halogenation, azidation, and
thiopyridyl formation from â-silyl radicals show that con-
siderable levels of diastereocontrol can be achieved for a
variety of substrates. A preliminary discussion of stereocon-
trol in radical reactions similar to the ones described here is
given in the accompanying Letter by Chabaud et al. The
transformations described may prove to be useful in a variety
of synthetic transformations.
95:5
a
Reaction of 3 at -10 °C in CCl3Br/CH2Cl2 initiated by light. b Yields
were 60-80% by isolation or determined by gas chromatography with
p-xylene internal standard. c Determined by gas chromatography and
comparison with authentic samples.
â-Halosilanes undergo anti-elimination of silylhalide to
3,4
Acknowledgment. This research was supported by grants
from the NIH (HL 17921) and the NSF (CHE- 0107697).
We thank Professor Philippe Renaud for helpful discussions
and for informing us of his results prior to publication.
produce olefins in moderate to excellent yields, and we
sought entry to these halosilanes via 3a-e. Free radical
halogenation was performed by irradiating the thiohydrox-
2 2 3
amate ester in a 1:1 mixture of CH Cl /CBrCl until the
mixture became practically colorless, and the crude reaction
mixture was immediately analyzed by GC. The results from
the halogenation reaction are presented in Table 3. The
halogenation and elimination of 3a (Table 3 entry 1) resulted
in an olefin E/Z ratio of 38:62, and a similar E/Z ratio (33:
Supporting Information Available: Detailed descrip-
tions of experimental procedures and spectral information
and analyses for new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
67) was obtained in the halogenation and elimination of
OL0268182
4256
Org. Lett., Vol. 4, No. 24, 2002