Stereospecific and stereodivergent construction of quaternary carbon centers through switchable directed/nondirected allylic substitution

Article abstract of DOI:10.1002/anie.200453991

Selectivity at the flick of a switch: Through 1,3-chirality transfer, a directing/nondirecting leaving group facilitates the stereospecific and stereodivergent construction of quaternary carbon centers by copper-mediated allylic substitution with Grignard

Full text of DOI:10.1002/anie.200453991

Angewandte
Chemie
Communications
Oxidation of the directing leaving group in an SN2’ reaction with
Grignard reagents causes a switch in the sense of the 1,3-chirality
transfer from syn to anti substitution.The syn-directing effect is used for
building up oligo(deoxypropionate)s diastereoselectively.For more
information, see the Communications by B.Breit and co-workers on
the following pages.
Angew. Chem. Int. Ed. 2004, 43, 3785
DOI: 10.1002/anie.200453991
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3785

Communications
Asymmetric Synthesis
substitution with Grignard reagents. Furthermore, we show
that the stereochemical outcome of the title reaction can be
reversed through an oxidative on/off switch with regard to the
directing power of the o-DPPB group. Thus, both enantio-
mers of the substitution product can be prepared from a single
enantiomer of the substrate (Scheme 1).
Stereospecific and Stereodivergent Construction
of Quaternary Carbon Centers through
Switchable Directed/Nondirected Allylic
Substitution**
Bernhard Breit,* Peter Demel, and Christopher Studte
The enantioselective construction of quaternary stereogenic
carbon centers is a challenge in organic synthesis. Much
progress has been made with enantioselective catalytic
methods, but still no general solution to this synthetic
problem exists.[1] A reliable alternative strategy for the
predictable installation of a quaternary stereocenter is 1,3-
chirality transfer from a more readily accessible stereogenic
center by a sigmatropic process, such as the Claisen rear-
rangement.[2] However, the nature of the sigmatropic process
sets structural limitations on the type of carbon nucleophile
that can be transferred. Allylic substitution with organo-
metallic carbon nucleophiles could provide a more general
solution if all aspects of selectivity could be controlled. Of
particular interest is copper-mediated allylic substitution, as it
allows the introduction of hard nucleophiles, such as alkyl,
alkenyl, and aryl groups, into an existing carbon skeleton.[3]
These reactions generally proceed by anti attack of the
nucleophile with respect to the leaving group.[3,4] However,
the simultaneous control of regio- and stereochemistry is a
difficult problem to solve, and only a few successful examples
are known.[5]
Scheme 1. Concept of stereodivergent allylic substitution with organo-
copper reagents for the stereospecific construction of quaternary
carbon centers by employing a switchable directing/nondirecting
leaving group.
Progress has been made in this area with substrate-
directed reactions in which reagent-directing leaving groups,
such as carbamates[6] and benzothiazoles,[7] are used. How-
ever, control over the geometry of the double bond in the
product is often unsatisfactory, so that the chirality transfer is
incomplete.[3] Furthermore, in many cases an excess of the
organometallic reagent is required, which is particularly
undesirable if valuable organic residues are to be transferred.
We recently found a solution to these problems by employing
the ortho-diphenylphosphanylbenzoate group (o-DPPB) as a
new reagent-directing leaving group. Copper-mediated and
copper-catalyzed allylic substitution with only a stoichiomet-
ric amount of a Grignard reagent could be carried out with
excellent chemo-, regio-, and stereoselectivity and with
recovery of the directing group.[8]
In a first series of experiments we studied the regioselec-
tivity of the directed allylic substitution with respect to the
formation of a quaternary carbon center. Thus, treatment of
the secondary trisubstituted allylic o-DPPB derivative 1[9]
with CuBr·SMe2 (0.5 equiv)[10] and a methyl or n-butyl
Grignard reagent (1.1 equiv)[11] provided the SN2’ substitution
products 2 and 3, respectively, with complete control of
regioselectivity and double-bond geometry (Table 1, entries 1
and 2). Furthermore, the o-DPPB esters (E)-4 and (Z)-4,
derived from the primary trisubstituted allylic alcohols
geraniol and nerol, respectively, underwent analogous trans-
formations to give the products 5–7 of SN2’ substitution in
excellent yields and with perfect regioselectivity, independent
of the double-bond geometry in the starting material (Table 1,
entries 3–6).
We report herein on the regioselective, stereospecific, and
stereodivergent construction of quaternary carbon stereo-
centers through o-DPPB-directed, copper-mediated allylic
To study chirality transfer we chose the o-DPPB esters
(À)-8a and (À)-8b, which are readily available from d-
mannitol. The corresponding products of allylic substitution
are equipped with appropriate functionalities for the flexible
incorporation of the quaternary stereocenter into a desired
carbon skeleton. The reaction of the p-methoxybenzyl (PMB)
ether derivative (À)-8a with a range of Grignard reagents in
the presence of CuBr·SMe2 (0.5 equiv) proceeded smoothly
with excellent regio- and stereoselectivity (Scheme 2; Table 2,
[*] Prof. Dr. B. Breit, Dr. P. Demel, Dipl.-Chem. C. Studte
Institut für Organische Chemie und Biochemie
Albert-Ludwigs-Universität Freiburg
Albertstrasse 21, 79104 Freiburg (Germany)
Fax: (+49)761-203-8715
E-mail: bernhard.breit@orgmail.chemie.uni-freiburg.de
[**] This work was supported by the DFG, the Fonds der Chemischen
Industrie, and the Krupp Foundation (Alfried Krupp Award for
Young Professors to B.B.).
3786
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200453991
Angew. Chem. Int. Ed. 2004, 43, 3786 –3789

Angewandte
Chemie
Table 1: Regioselective formation of quaternary carbon centers through o-DPPB-directed allylic substitution.[a]
Entry
1
o-DPPBester [b]
RMgX (equiv)
MeMgI (1.1)
Product
SN2’/SN2[c]
>99:1
Yield [%][d]
68[e]
1
1
2
3
2
nBuMgBr (1.1)
99:1
99
3
4
(E)-4
(E)-4
MeMgI (1.2)
EtMgBr (1.2)
5
6
95:5
91
80
>98:2
5
6
(Z)-4
(E)-4
EtMgBr (1.2)
6
7
>98:2
>99:1
95
87
nBuMgBr (1.2)
[a] Reactions were performed in diethyl ether, c(o-DPPBester) =0.05m. The Grignard reagent (0.51–1.23m in diethyl ether) was added to the reaction
mixture with a syringe pump over a period of 30 min. [b] Prepared from the corresponding allylic alcohol by an esterification protocol reported
previously.[9] [c] Determined by GC (CPSil5CB, 30 m, 0.32 mm ID, Chrompack). [d] Yield of isolated product after distillation (entry 1) or
chromatographic purification (entries 2–6). [e] The low yield is due to the volatility of the product.
Table 2: Stereospecific and stereodivergent formation of quaternary carbon
centers: acyclic substrates.[a]
Entry Substrate[b] R2
SN2’/SN2[c] E/Z[c]
Product
CT[d] Yield[e]
(ee [%])
(ee [%])[c]
[%]
[%]
1
2
3
4
8a (94)
8a (94)
8a (93)
8a (93)
8a (93)
8a (92)
Me >95:5
Et >99:1
nBu >99:1
>95:5
9a
–
72
86
99
89
53
>99:1
>99:1
>99:1
>99:1
(À)-9b (94)
(À)-9c (93)
(À)-9d (93)
(À)-9e (85)
100
100
100
91
iPr
>99:1
>99:1
5[f]
6
Bn
tBu
14:86[g]
80:20[g]
9 f (n.d.[h]) n.d.[h] 90
7
8
9
10
11
12[i]
8b (>99) Et
8b (>99) nBu >99:1
8b (>99) iPr
>99:1
>99:1
>99:1
>99:1
>99:1
>98:2
>95:5
(À)-9g (98)
(À)-9h (97)
(À)-9i (97)
(+)-9g (99)
(+)-9h (99)
98
98
98
100
100
84
87
84
85
87
94
>99:1
>99:1
11 (>99)
11 (>99)
11 (>99)
Et
nBu >98:2
iPr
97:3[g]
(+)-9i (>99) 100
[a] All reactions were performed in diethyl ether, c(o-DPPBester) =0.05m. The
Grignard reagent (0.76–1.23m in diethyl ether) was added to the reaction mixture
with
a syringe pump over a period of 15–20 min. [b] Prepared from the
corresponding allylic alcohol[12] by an esterification protocol reported previously.[9]
The enantiomeric excess was determined by HPLC analysis of the corresponding
allylic alcohol (Chiralpak AD (8a), Chiralcel OD-H (8b)). [c] Determined by
1H NMR spectroscopy (entry 1) or HPLC analysis (entries 2, 3, 5, 7, 9: Chiral-
cel OD-H; entry 4: Chiralpak AD after removal of the TBDMS group; entries 9, 12:
Chiralcel OD-H after removal of the TBDPS group). [d] The chirality transfer (CT)
was calculated as CT=(ee(9)/ee(8 or 11))”100. [e] Yield of isolated product after
chromatographic purification. [f] c(8a)=0.01m in diethyl ether; the Grignard
reagent (0.07m in diethyl ether) was added over a period of 90 min. [g] Product
ratios were determined by 1H NMR spectroscopy. [h] n.d.=not determined. [i] With
1-methyl-2-pyrrolidinone (NMP) as a cosolvent (one third of the total solvent
volume).
Scheme 2. Stereospecific and stereodivergent construction of quater-
nary carbon centers through the switchable directed/nondirected allylic
substitution of the acyclic substrates (À)-8a/b and (À)-11 (see
Table 2). PG=protecting group, PMB=p-methoxybenzyl,
TBDMS=tert-butyldimethylsilyl, TBDPS=tert-butyldiphenylsilyl.
entries 1–4). Perfect 1,3-chirality transfer was observed in the
construction of quaternary carbon centers (Table 2, entries 2–
4). However, when a benzyl Grignard reagent was used the
degree of chirality transfer was slightly lower (91%; Table 2,
entry 5). The absolute configuration of the substitution
products was determined upon the conversion of (À)-9b
into the known ester (À)-10 [Eq. (1); (DDQ = 2,3-dichloro-
5,6-dicyano-1,4-benzoquinone)].[5b]
Thus, as expected, the reaction proceeds through a syn
substitution pathway with respect to the leaving group. The o-
DPPB group acts as a directing leaving group through
Angew. Chem. Int. Ed. 2004, 43, 3786 –3789
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3787

Communications
phosphane coordination of the organocopper reagent. A
transition state that accounts for the observed stereochemis-
try is depicted in Scheme 1.
Next, allylic substitution of the enoate (À)-8b was
examined. In this case the problem of thermodynamically
unfavorable enoate deconjugation combines with a chemo-
selectivity issue arising from reactivity of the ethyl ester
functionality toward the nucleophilic Grignard reagent.[3,5a,b]
Interestingly, even in this case the directed allylic substitution
of occurred with complete chemo-, regio-, and stereoselec-
tivity and with excellent chirality transfer to give the
substitution products (À)-9g–i (Table 2, entries 7–9).
To access the opposite enantiomers of the substitution
products of the same allylic o-DPPB ester substrates the
stereochemical course of the allylic substitution must be
reversed. Instead of proceeding through a directed syn
substitution pathway, the reaction should proceed through
the nondirected anti attack of the nucleophile with respect to
the leaving group. Thus, the directing power of the o-DPPB
functionality would have to be turned off, for example, by
oxidation of the phosphane group of the o-DPPB enoate (À)-
8b to give the phosphane oxide (À)-11. The ability of the o-
DPPB group to coordinate the copper reagent should thus be
suppressed and the leaving-group ability of the benzoate
group increased significantly.
Treatment of the allylic o-DPPB oxide ester (À)-11 with
dialkyl zinc reagents in the presence of CuCN·2LiCl in THF
furnished the anti SN2’ substitution products (+)-9g–i in good
yields with perfect regioselectivity and excellent 1,3-anti
chirality transfer.[13] However, when the o-DPPB ester (À)-8b
was subjected to reaction conditions identical to those used in
entries 10–12 of Table 2, allylic substitution did not occur
even after warming the reaction mixture to room temperature
and prolonged reaction time (24 h). Evidently the oxidation
of the phosphane functionality to the phosphane oxide serves
two purposes. First, it switches off the directing power of the
phosphane group and thus suppresses the directed syn
substitution pathway. Second, it enhances the leaving-group
ability of the benzoate group to make nondirected anti
substitution possible. This methodology is also applicable to
cyclic substrates with similar efficiency. For example, stereo-
divergent directed/nondirected substitution of the six-mem-
bered-ring systems 12a,b and 13a,b is possible (Scheme 3;
Table 3).
Scheme 3. Stereospecific and stereodivergent construction of quater-
nary carbon centers through the switchable directed/nondirected allylic
substitution of the cyclic substrates (À)-12a/b and (À)-13a/b (see
Table 3).
Table 3: Stereospecific and stereodivergent formation of quaternary
carbon centers: cyclic substrates.
Entry Substrate[a]
R
SN2’/SN2[b] Product
(ee [%])[b]
CT[c] Yield[d]
[%] [%]
(ee [%])
1[e]
12a (97)
12a (97)
12a (97)
12a (97)
13a (94)
13a (94)
13a (94)
13a (94)
12b (97)
12b (97)
12b (97)
13b (94)
13b (94)
13b (94)
Me
99:1
(+)-14a (96)
99 >95
99 >95
99 >95
95 >95
97 >95
2[e]
nBu >99:1
(+)-14b (96)
(À)-14c (96)
(+)-14d (92)
(À)-14a (91)
3[e]
iPr
Ph
Et
nB
iPr
Ph
Me
98:2
40:60
96:4
4[e]
5[f]
6[f]
u
98:2 +)-15( a (94) 100 >95
7[f]
96:4
41:59
99:1
(À)-15b (91)
(+)-15c (82)
(À)-14a (93)
(À)-14b (96)
(+)-14c (94)
(+)-14a (93)
97 >95
87 >95
96 >95
99 >95
97 >95
99 >95
8[f]
9[g]
10[g]
11[g]
12[g]
13[g]
14[g]
nBu >99:1
iPr
Et
nB
iPr
97:3
>99:1
u
99:1 À)-15( a (94) 100 >95
99:1
(+)-15b (93) 99 >95
[a] Prepared from the corresponding allylic alcohol following an ester-
ification protocol reported previously.[9] The enantiomeric excess was
determined by HPLC analysis (Chiralpak AD-H (12a), Chiralcel OD-H
(13a)). [b] Determined by 1H NMR spectroscopy and GC analysis
(Supelco Beta Dex 110 (14a,14d,15c), C.E.I. G-TA (14b,14c,15a,15b)).
[c] The chirality transfer (CT) was calculated as CT=(ee(14)/ee(12))”
100. [d] Yield was determined by GC. [e] The Cu-complexed o-DPPB
esters were added in diethyl ether/dichloromethane (95:5, c=0.01m) to
the Grignard reagent (0.05m in diethyl ether) by using a syringe pump
(6 mLhÀ1). [f] As for [e], but diethyl ether/dichloromethane (4:1, c=
0.01m). [g] The oxidized o-DPPBesters ( c=0.07m in THF) were added
In conclusion, we have shown that the o-DPPB/o-DPPB
oxide system can be used as a switchable directing/non-
directing leaving group in a copper-mediated allylic substitu-
tion reaction with Grignard and organozinc reagents for the
regioselective, stereospecific, and stereodivergent construc-
tion of quaternary carbon centers. Thus, both enantiomers of
the substitution product are readily available from one
enantiomer of the substrate.
to the zinc–copper reagents (c=1.00m in THF) at a rate of 12 mLhÀ1
.
Experimental Section
Synthesis of (À)-9h (syn substitution): CuBr·SMe2 (11.9 mg,
0.058 mmol) was added to a solution of freshly purified (À)-8b
(82 mg, 0.117 mmol, > 99% ee) in diethyl ether (2.3mL), and the
mixture was stirred for a further 5 min at room temperature.
nBuMgBr (0.11 mL, 0.140 mmol, 1.23m solution in diethyl ether)
3788
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3786 –3789

Angewandte
Chemie
was added to the resulting yellow solution over a period of 20 min by
using a syringe pump. When the addition of the Grignard reagent was
complete, the reaction mixture was quenched by adding a saturated
aqueous solution of NH4Cl (2.5 mL), followed by diethyl ether
(10 mL) and an aqueous solution of NH3 (12.5%, 1 mL). This mixture
was extracted with diethyl ether (3” 20 mL). The combined ethereal
phases were washed with water (10 mL), then brine (10 mL), and
dried (MgSO4). The solvent was then removed in vacuo. Flash
chromatography of the residue on silica with petroleum ether (60–70)/
diethyl ether (95:5) yielded analytically pure (À)-9h (45 mg, 85%,
SN2’/SN2 > 99:1, E/Z > 99:1, > 97% ee) as a colorless oil. HPLC
(Chiralcel OD-H, 158C, n-heptane, 0.8 mLminÀ1, 227 nm): tR ((À)-
9h): 20.01 min (98.7%), tR ((+)-9h): 21.90 min (1.3%); [a]2D2 = À3.7
(c = 1.75, CHCl3); 1H NMR (400.136 MHz, CDCl3, TMS): d = 0.89 (t,
3H, J = 7.3Hz, CH 3), 1.06 (s, 9H, 3” CH 3), 1.23(t, 3H, J = 7.3Hz,
the Wittig rearrangement, see: T. Nakai, K. Mikami, Chem. Rev.
1986, 86, 885 – 902.
[3] Y. Yamamoto, Methods of Organic Chemistry (Houben-Weyl),
Vol. E21, Thieme, Stuttgart, 1995, pp. 2011 – 2040; b) B. Breit, P.
Demel in Modern Organocopper Chemistry (Ed.: N. Krause),
Wiley-VCH, Weinheim, 2002, pp. 188 – 223.
[4] E. J. Corey, N. W. Boaz, Tetrahedron Lett. 1984, 25, 3063 – 3066.
[5] a) T. Ibuka, M. Tanaka, S. Nishii, Y. Yamamoto, J. Chem. Soc.
Chem. Commun. 1987, 1596 – 1598; b) T. Ibuka, N. Akimoto, M.
Tanaka, S. Nishii, Y. Yamamoto, J. Org. Chem. 1989, 54, 4055 –
4061; c) N. Harrington-Frost, H. Leuser, M. I. Calaza, F. F.
Kneisel, P. Knochel, Org. Lett. 2003, 5, 2111 – 2114.
[6] a) C. Gallina, Tetrahedron Lett. 1982, 23, 3094 – 3096; b) H. L.
Goering, S. S. Kantner, C. C. Tseng, J. Org. Chem. 1983, 48, 715 –
721; c) J. H. Smitrovich, K. A. Woerpel, J. Am. Chem. Soc. 1998,
120, 12998 – 12999; d) J. H. Smitrovich, K. A. Woerpel, J. Org.
Chem. 2000, 65, 1601 – 1614.
CH3), 1.24 (s, 3H, CH3), 1.28 (m, 4H, 2 ” CH2), 1.53(m, 1H, CH ),
2
1.67 (m, 1H, CH2), 4.10 (m, 1H, CH2), 4.13(m, 1H, CH ), 4.22 (dd,
2
2H, J = 5.5, 1.7 Hz, CH2), 5.58 (m, 1H, CH), 5.90 (dt, 1H, J = 15.9,
1.7 Hz, CH), 7.39 (m, 6H, Ar-H), 7.68 ppm (m, 4H, Ar-H); 13C NMR
(100.624 MHz, CDCl3): d = 14.0, 14.2, 19.2, 21.0, 23.1, 26.8 (3 ” C),
26.8, 39.2, 47.6, 60.5, 64.5, 127.6 (4 ” C), 127.6, 129.6 (2 ” C), 133.9,
134.2 (2 ” C), 135.6 (4 ” C), 176.0 ppm; elemental analysis (%) calcd
for C28H40O3Si (452.70): C 74.29, H 9.01; found: C 74.38, H 8.91.
[7] a) P. Barsanti, V. Calꢀ, L. Lopez, G. Marchese, F. Naso, G. Pesce,
J. Chem. Soc. Chem. Commun. 1978, 1085 – 1086; b) V. Calꢀ, L.
Lopez, W. F. Carlucci, J. Chem. Soc. Perkin Trans. 1 1983, 2953–
2956; c) S. Valverde, M. Bernabꢁ, S. Garcia-Ochoa, A. M.
Gꢂmez, J. Org. Chem. 1990, 55, 2294 – 2298.
[8] B. Breit, P. Demel, Adv. Synth. Catal. 2001, 343, 429 – 432.
[9] The o-DPPB esters described in this work were prepared
according to methodology described previously: B. Breit, G.
Heckmann, S. K. Zahn, Chem. Eur. J. 2003, 9, 425 – 434.
[10] CuBr·SMe2 (99% purity) was purchased from Aldrich and used
without further purification.
[11] All Grignard reagents were freshly prepared under standard
conditions by using activated magnesium turnings in diethyl
ether according to the dry-stir activation protocol: K. V. Baker,
J. M. Brown, N. Hughes, A. J. Skarmilis, A. Sexton, J. Org. Chem.
1991, 56, 698 – 703.
Synthesis of (+)-9h (anti substitution):
A freshly prepared
solution of CuCN·2LiCl in THF (52 mL, 0.052 mmol, 1m; a mixture
of the two salts was dried for 2 h at 1208C in vacuo (oil pump) prior to
use) was cooled to À308C and diluted with additional THF (0.5 mL).
A solution of nBu2Zn in heptane (0.1 mL, 0.1 mmol, 1.0m) was added
dropwise, and the resulting colorless solution was stirred for a further
10 min at À308C.
A
solution of (À)-11 (31 mg, 0.043 mmol,
> 99% ee) in THF (1 mL) was added over a period of 5 min. The
resulting colorless solution was warmed to 08C over 2.5 h, during
which time a beige suspension formed (quantitative conversion by
TLC). The reaction was quenched by adding a saturated aqueous
solution of NH4Cl (2 mL), followed by diethyl ether (20 mL) and an
aqueous solution of NH3 (12.5%, 0.5 mL). This mixture was extracted
with diethyl ether (3” 10 mL). The combined ethereal phases were
washed with water (10 mL), then brine (10 mL), and dried (MgSO4).
The solvent was removed in vacuo. Flash chromatography of the
residue on silica with petroleum ether (60–70)/ethyl acetate (95:5)
yielded pure (+)-9h (17 mg, 87%, SN2’/SN2 > 99:1, E/Z > 98:2,
> 99% ee) as a colorless oil. HPLC (Chiralcel OD-H, 158C, n-
heptane, 0.5 mLminÀ1, 227 nm): tR ((À)-9h): 20.01 min (0.5%), tR
((+)-9h): 21.90 min (99.5%); [a]2D0 = +9.1 (c = 0.87, CHCl3).
[12] J. H. Hong, M.-Y. Gao, Y. Choi, Y.-C. Cheng, R. F. Schinazi,
C. K. Chu, Carbohydr. Res. 2000, 328, 37 – 48.
[13] Experiments with magnesium cuprates also lead to excellent 1,3-
chirality transfer through anti substitution. However, the regio-
selectivity for the SN2’ substitution product was incomplete. An
excess of the zinc reagent was required for quantitative
conversion.
Received: February 11, 2004 [Z53991]
Published Online: May 26, 2004
Keywords: allylation · asymmetric synthesis · enantioselectivity ·
.
Grignard reaction · organocopper reagents
[1] a) S. F. Martin, Tetrahedron 1980, 36, 419 – 460; b) K. Fuji, Chem.
Rev. 1993, 93, 2037 – 2066; c) E. J. Corey, A. Guzman-Perez,
Angew. Chem. 1998, 110, 402 – 415; Angew. Chem. Int. Ed. 1998,
37, 388 – 401; d) C. Spino, C. Beaulieu, Angew. Chem. 2000, 112,
2006 – 2008; Angew. Chem. Int. Ed. 2000, 39, 1930 – 1932; e) J.
Christoffers, A. Mann, Angew. Chem. 2001, 113, 4725 – 4732;
Angew. Chem. Int. Ed. 2001, 40, 4591 – 4597; f) J. Christoffers, A.
Baro, Angew. Chem. 2003, 115, 1726 – 1728; Angew. Chem. Int.
Ed. 2003, 42, 1688 – 1690.
[2] a) For [3,3’] sigmatropic processes, see: H. Frauenrath, Methods
of Organic Chemistry (Houben-Weyl), Vol. E21, Thieme, Stutt-
gart, 1995, pp. 3301 – 3756; b) for [2,3’] sigmatropic processes,
see: J. Kallmerten, Methods of Organic Chemistry (Houben-
Weyl), Vol. E21, Thieme, Stuttgart, 1995, pp. 3757 – 3809; c) for
Angew. Chem. Int. Ed. 2004, 43, 3786 –3789
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3789