New elements in the reactivity of α-cyclopropyl vinyl radicals

Article abstract of DOI:10.1055/s-2002-31915

The reactivity of the α-cyclopropyl vinyl radical has been examined through the use of propargylic bromomethyldimethylsilyl ethers bearing a cyclopropyl group on the acetylenic moiety. With unsubstituted precursors, allenes can be obtained. With appropriate trapping possibilities (5-exo cyclization or hydrogen transfer), the cyclopropyl moiety on the vinyl radical is retained.

Full text of DOI:10.1055/s-2002-31915

LETTER
923
New Elements in the Reactivity of -Cyclopropyl Vinyl Radicals
C
E
yclopropyl
V
i
m
nylRadicals ily Mainetti, Louis Fensterbank,* Max Malacria*
UMR 7611, Université Pierre et Marie Curie, 4, Place Jussieu, b. 229, 75005 Paris, France
Fax +33(1)44277360; E-mail: fensterb@ccr.jussieu.fr; E-mail: malacria@ccr.jussieu.fr
Received 22 March 2002
This required us to prepare a certain number of cyclopro-
pyl propargylic alcohols of type 2 whose synthesis is
scarcely described in the literature (Scheme 2). As cyclo-
propylacetylene is an expensive and not so readily avail-
Abstract: The reactivity of the -cyclopropyl vinyl radical has
been examined through the use of propargylic bromomethyldime-
thylsilyl ethers bearing a cyclopropyl group on the acetylenic moi-
ety. With unsubstituted precursors, allenes can be obtained. With
appropriate trapping possibilities (5-exo cyclization or hydrogen able starting material, we preferred using the two
transfer), the cyclopropyl moiety on the vinyl radical is retained.
following routes. Starting from commercially available
cyclopropanecarboxaldehyde, a Corey–Fuchs sequence,8
via dibromoolefin 1, provided good yields of alcohols
Key words: vinyl radicals, cyclizations, ring opening, hydrogen
transfer, cyclopropanes
2
a–c. Alternatively, an even more straightforward route
involved the dimetallation of the commercially available
chloropentyne, followed by trapping with a carbonyl de-
9
Thanks to its well-established kinetic and thermodynamic
rivative, and provided comparable yields of 2a and 2c.
Secondary alcohol 2d was obtained through a sequence
we already reported and which involves the formation of
the allyl ether 3 followed by a Wittig [2,3]-rearrange-
ment. For the OTBS-substituted cyclopropyl alcohol 2e,
we prepared the cyclopropyl acetylenic derivative 4 in its
Z form according to the method of de Meijere.11
parameters, the ring opening of the cyclopropylcarbinyl
radical is one of the most popular radical clocks.1,2 As it
provides ultrafast mechanistic probes of radical inter-
3
mediacy and lifetime, its uses span biological systems
10
4
to surface science. A related but much less assessed
system is the vinyl analog (Scheme 1), for which, to
the best of our knowledge, no physical data have been
reported.
7
-1
k
25 = 9.4 10 s
Br
R2
CBr4,
PPh3
1 . n-BuLi
O
H
Br
(2 equiv.)
R1
3
-1
k
25 = 10 s
2.
O
HO
1
, 78%
R1
R2
1
2
2
2
2
a, R = R = H, 86%
1
2
b, R = i-Pr, R = H, 69%
?
1
2
c, R = R = i-Pr , 86%
1
. n-BuLi
(
2 equiv.)
Scheme 1
2
a, 66%
Cl
2c, 71%
2
.
O
R1
R2
Inspection of the literature provided only sporadic reports.
One of the first investigators in that area was Crandall5
who showed that at high concentration almost no allene
was observed from the reduction of an -cyclopropyl vi-
nyliodide. Only under dilute reaction conditions a sub-
stantial amount of allene was formed. Another important
1
2
. NaH
.
Br
n-BuLi
2a
O
THF, ∆
HO
6
3, 93%
2d, 92%
contribution was given by Back, who studied the seleno-
sulfonation of cyclopropylacetylene, and found a mixture
where the allene is the minor component. Interestingly,
when run with vinylcyclopropane, the same reaction af-
forded only ring-opened adducts.
OTBS
OTBS
O
H
MgCl
1. n-BuLi
2. DMF
Z-4
5, 80%
3. KH2PO4
All these findings suggested that the ring opening of the
vinyl radical is probably slower. We wanted to examine
this, and for that purpose utilized our tool, the radical cy-
clization of propargylic bromomethyl-dimethylsilyl
OTBS
7
HO
ethers.
2e, 89%
2
dias. 1 : 1
Synlett 2002, No. 6, 04 06 2002. Article Identifier:
Scheme 2
1
437-2096,E;2002,0,06,0923,0926,ftx,en;G07502ST.pdf.
Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214
©
Scheme 2

9
24
E. Mainetti et al.
LETTER
Formylation12 and alkylation furnished a 1:1 mixture of originates from the fast opening of the cyclopropyl ring on
17
alcohol 2e. All these propargylic alcohols could be un- 15. Precursor 6e possesses a OTBS group which could
eventfully silylated with bromomethyldimethylchlorosi- be useful for further organic synthesis and should acceler-
lane in presence of triethylamine and DMAP to give the ate the ring opening process.18 Nevertheless, the radical
corresponding silyl ethers 6a–e.
cyclization of 6e furnished a complex mixture from which
no allene was detected. Only one diastereomer of 17, dis-
playing a cis relationship between the hydroxy group and
the methyl group as determined by NOE, was isolated, al-
beit in low yield.
Our preliminary attempts focused on the simplest precur-
sor 6a. Upon treatment with Bu SnH under slow addition
3
–4
–1
conditions (2 10 molh ) in order to favor the initial 5-
exo-dig cyclization, followed by Tamao oxidation, allene
diol 9 was obtained in a low yield (24%) which did not re-
flect the cleanliness of the reaction (Scheme 3). A methyl-
lithium treatment and further trapping by methyl
chloroformate resulted in a much better yield of function-
1
2
. Bu3SnH
. MeLi
1
3
alized allene 10. Thus under these conditions, no inter-
molecular reduction of the vinyl radical 7 seems to
intervene at the expense of the ring opening process which
generates homoallenyl radical 8.
O
O
HO
SiMe3
Si
6d
Si
Br
12
13, 57%
We performed the same transformation on precursor 6b
and showed that a stereogenic center at the propargylic
position does not control the diastereoselectivity of the
ring opening of the cyclopropyl ring, since a 1:1 diastere-
omeric mixture of allenes 11 was obtained. This reaction
constitutes one of the rare example of radical allene syn-
1
. Bu3SnH
2. MeLi
O
O
HO
16, 64%
SiMe3
Si
Si
14
15
Br
1
4
thesis, and gives an access to functionalized allene de-
1
5
rivatives.
Because we run these reactions under slow addition con-
ditions in a dilute medium to favor the initial 5-exo-dig cy-
clization, it will always be difficult to assess the reactivity
of these -cyclopropyl vinyl radicals by intermolecular
trapping. So, we concentrated on intramolecular competi-
tive trapping reactions. For instance, we set an unsatura-
tion for a further 5-exo-trig cyclization from the initial
vinyl radical as in precursors 6d and 6e. A clean reaction
OTBS
OTBS 1. Bu SnH
3
2. MeLi
O
HO
SiMe3
Si
Br
6e, 2 dias. 1 : 1
17, 30%
Scheme 4
1
6
was observed with 6d, giving cyclopentenol 13 as a sin-
1
0
gle diastereomer. This suggests that the 5-exo cycliza-
tion of 12 was faster than the ring-opening process
Another type of possible trapping is hydrogen transfer and
previous studies have established that the diisopropyl pre-
(Scheme 4). This result is especially significant when
1
9
cursor 6c is perfectly suited for that purpose. Here also,
two possibilities are open for the vinyl radical 18. The ma-
compared to the outcome of the cyclization of the allyl
precursor 14 that gave 1,6-diene 16 as sole adduct, which
HO
2. Tamao ox.
1. Bu3SnH, AIBN
HO
2.10-4 mol.h-1
)
9, 24%
(
O
O
O
Si
PhH, ∆
Si
Si
Br
MeO2CO
Me3Si
6a
7
8
2
. MeLi
3. ClCO2Me
10, 66%
1
2
3
. Bu3SnH, AIBN
. MeLi
. ClCO2Me
MeO2CO
Me3Si
O
Si
Br
6b
11, 57%, 1 : 1 ds.
Scheme 3
Synlett 2002, No. 6, 923–926 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Cyclopropyl Vinyl Radicals
925
jor pathway is the opening of the cyclopropyl ring which References
delivers allene 21. The minor one is a diastereoselective
(
1) For a leading review on radical clocks, see: Newcomb, M.
Tetrahedron 1993, 49, 1151.
2) Nonhebel, D. C. Chem. Soc. Rev. 1993, 347.
1
,5-H transfer, followed by a 5-endo-trig cyclization and
diastereoselective reduction to give a single diastereomer
(
of 22 (Scheme 5), whose stereochemistry was deduced by
analogy with our previous reports.
(3) For a recent contribution, see: Lis, L.; Koltun, E. S.; Liu, H.-
1
9
W.; Kass, S. R. J. Am. Chem. Soc. 2002, 124, 1276.
(
(
4) Kretzschmar, I.; Levinson, J. A.; Friend, C. M. J. Am. Chem.
Soc. 2000, 122, 12395.
5) Crandall, J. K.; Tindell, G. L.; Manmade, A. Tetrahedron
Lett. 1982, 23, 3769.
In conclusion, we have uncovered new and useful pieces
of information on the reactivity of -cyclopropyl vinyl
radicals. With unsubstituted precursors, allenes can be ob-
tained. With appropriate trapping possibilities (5-exo cy-
clization or hydrogen transfer), the cyclopropyl moiety on
the vinyl radical is retained. This allows the construction
of cyclic derivatives incorporating a versatile cyclopropyl
ring. Our results seem to confirm that the ring opening of
(6) Back, T. G.; Muralidharan, K. R. J. Org. Chem. 1989, 54,
21.
7) Fensterbank, L.; Malacria, M.; Sieburth, S. M. Synthesis
997, 813.
1
(
1
(
(
8) Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.
9) Corley, E. G.; Thompson, A. S.; Huntington, M. Org. Synth.
-
cyclopropyl vinyl radicals is a relatively slow process,
2000, 77, 231.
especially when compared to the alkyl series. The loss of
conjugation of the vinylcyclopropane moiety which re-
sults from a 90° flip in order to reach the C-C bond break-
ing conformation could constitute the retarding factor. To
pursue this work, two directions are now followed. The
first one will consist in the determination of kinetics,
which could be useful for the development of a new radi-
cal clock. The second one will focus in the application of
the -cyclopropyl vinyl radical as a new partner for radi-
cal cascades leading to the synthesis of biologically rele-
vant molecules.
(10) Journet, M.; Malacria, M. J. Org. Chem. 1992, 57, 3085.
(11) Militzer, H.-C.; Schömenauer, S.; Otte, C.; Puls, C.; Hain, J.;
Bräse, S.; de Meijere, A. Synthesis 1993, 998.
(
12) Journet, M.; Cai, D.; DiMichele, L. M.; Larsen, R. D.
Tetrahedron Lett. 1998, 39, 6427.
(
13) Spectroscopic data for 10. IR(neat): 2960, 1960, 1750, 1450,
–1
1
1
370, 1270 cm ; H NMR (200 MHz, CDCl ) 5.21 (m, 1
3
H, vinyl), 4.48 (t, J = 2.0 Hz, 2 H, CH O), 3.75 (s, 3 H,
2
CO CH ), 1.97 (quint, J = 7.4 Hz, 2 H, CH CH ), 1.32 (mAB,
2
3
2
3
2 H, CH Si), 0.97 (t, J = 7.4 Hz, 3 H, CH ), 0.02 (s, 9 H,
2
3
1
3
Si(CH ) ); C NMR (50 MHz, CDCl3) 201.8 (C,
3 3
C=C=CH), 155.8 (C, C=O), 97.4 (C, C=C=CH), 94.8 (CH,
C=C=CH), 70.1 (CH , CH O), 54.8 (CH , CO CH ), 22.5
2
2
3
2
3
(
CH , CH CH ), 18.2 (CH , CH Si), 13.6 (CH ), –1.3 (3
2 2 3 2 2 3
CH , Si(CH ) ). Anal. Calcd. for C H O Si: C, 59.46; H,
Acknowledgement
3
3 3
12 22
3
9.15. Found: C, 59.34; H, 9.35.
E. M. thanks the Ministère de l’Education Nationale for an AMX
scholarship.
(14) For a radical synthesis of allenes based on the -elimination
of vinylsulfoxides, see: (a) Delouvrié, B.; Lacôte, E.;
Fensterbank, L.; Malacria, M. Tetrahedron Lett. 1999, 40,
3565. (b) Mouriès, V.; Delouvrié, B.; Lacôte, E.;
Fensterbank, L.; Malacria, M. Eur. J. Org. Chem. 2002, in
press.
H
H
1. Bu3SnH
2. MeLi
HO
21, 46%
O
O
18
Me3Si
Si
Si
6c
Br
1,5-H
O
19
Si
5-endo-trig
H
H
2. MeLi
H
O
20
Si
HO
SiMe3
22, 24%
Scheme 5
Synlett 2002, No. 6, 923–926 ISSN 0936-5214 © Thieme Stuttgart · New York

9
26
E. Mainetti et al.
LETTER
(
15) For instance, see: (a) Lee, P. H.; Bang, K.; Lee, K.; Lee, C.-
H.; Chang, S. Tetrahedron Lett. 2000, 41, 7521. (b) Spino,
C.; Fréchette, S. Tetrahedron Lett. 2000, 41, 8033.
Si(CH ) ]. Anal. Calcd. for C H OSi: C, 69.58; H, 10.78.
Found: C, 69.55; H, 10.89.
(17) (a) We thank one of the referees for bringing to our attention
3
3
13 24
(
c) Stichler-Bonaparte, J.; Kruth, H.; Lunkwitz, R.;
that the rate constant of a 5-exo vinyl radical cyclization has
been estimated to be: 3.2 10 s at 60 °C: Beckwith, A. L.
8
–1
Tschierske, C. Liebigs Ann. 1996, 1375.
(
16) (a) The cis relative stereochemistry was assigned by analogy
J.; O’Shea, D. M. Tetrahedron Lett. 1986, 27, 4525.
1
0
5 –1
with our previous work, see ref. and: Devin, P.;
Fensterbank, L.; Malacria, M. J. Org. Chem. 1998, 63,
(b) Compared to a value of 2.3 10 s (20 °C) for the alkyl
analog: Chatgilialoglu, C.; Ingold, K. U.; Scaiano, J. C. J.
Am. Chem. Soc. 1981, 103, 7739. (c) Although the
Beckwith system cannot directly be compared to the system
studied herein (different substitution pattern), this data could
suggest an intrinsic faster cyclization of the vinyl radical,
thus contributing significantly to the observed effect.
(18) An alkoxy group reduces the activation barrier for the ring
opening of cyclopropylcarbinyl radicals, see: Martinez, F.
N.; Schlegel, H. B.; Newcomb, M. J. Org. Chem. 1998, 63,
3618.
6
764. (b) Spectral data for 13. IR(neat): 3350, 3080, 2950,
–
1 1
1640, 1250 cm ; H NMR (400 MHz, CDCl ) 4.40 (m, 1
3
H, CH-OH), 2.31 (m, 2 H, CH cyclopentenol + CHCH ),
2
3
1
.76 (d, J = 13.8 Hz, 1 H, CH Si), 1.64 (d, J = 13.8 Hz, 1 H,
2
CH Si), 1.34 (m, 1 H, CH cyclopropane), 1.26 (m, 1 H, CH
2
2
cyclopentenol), 1.11 (d, J = 6.6 Hz, 3 H, CH ), 0.70 (m, 2 H,
3
cyclopropane), 0.55 (m, 1 H, cyclopropane), 0.41 (m, 1 H,
1
3
cyclopropane), 0.04 [9 H, Si(CH ) ]; C NMR (100 MHz,
3
3
CDCl3) 141.2 (C, C=C), 137.1 (C, C=C), 79.9 (CH, CH-
OH), 41.9 (CH , cyclopentenol), 38.6 (CH, cyclopentenol),
(19) Bogen, S.; Gulea, M.; Fensterbank, L.; Malacria, M. J. Org.
Chem. 1999, 64, 4920.
2
2
3.0 (CH ), 16.4 (CH , CH Si), 9.7 (CH, cyclopropane), 5.8
3 2 2
(
CH , cyclopropane), 5.2 (CH , cyclopropane), –0.4 [3 CH ,
2
2
3
Synlett 2002, No. 6, 923–926 ISSN 0936-5214 © Thieme Stuttgart · New York