Molecular insight into surface organometallic chemistry through the combined use of 2D HETCOR solid-state NMR spectroscopy and silsesquioxane analogues

Article abstract of DOI:10.1002/1521-3773(20011203)40:23<4493::AID-ANIE4493>3.0.CO;2-X

Feher cage analogues, such as 1, provide easy access to models of surface structures such as 2, and give a better understanding of surface reactions. In addition, surface complexes such as 2 can be fully characterized by high-resolution 1D and 2D solid-state NMR spectroscopy.

Full text of DOI:10.1002/1521-3773(20011203)40:23<4493::AID-ANIE4493>3.0.CO;2-X

COMMUNICATIONS
1963, 2, 683; b) less basic organocerium reagents also exhibit aza-
enolization: D. Enders, E. Diez, R. Fernandez, E. Martin-Zamora,
J. M. Munoz, R. R. Pappalardo, J. M. Lassaleta, J. Org. Chem. 1999,
64, 6329.
Molecular Insight Into Surface Organometallic
Chemistry Through the Combined Use of 2D
HETCORSolid-State NMRSpectroscopy and
Silsesquioxane Analogues**
[3] For selected recent examples, see: a) review of asymmetric Strecker-
type reactions: L. Yet, Angew. Chem. 2001, 113, 900; Angew. Chem.
Int. Ed. 2001, 40, 875; b) for asymmetric Mannich reactions, see: W.
Notz, K. Sakthivel, T. Bui, G. Zhong, C. F. Barbas III, Tetrahedron
Lett. 2001, 42, 199; B. List, J. Am. Chem. Soc. 2000, 122, 9336; S. Saito,
K. Hatanaka, H. Yamamoto, Org. Lett. 2000, 2, 1891; K. Miura, K.
Tamaki, T. Nakagawa, A. Hosomi, Angew. Chem. 2000, 112, 2034;
Angew. Chem. Int. Ed. 2000, 39, 1958.
Mathieu Chabanas, Elsje Alessandra Quadrelli,
¬
Bernard Fenet, Christophe Coperet,*
JeanThivolle-Cazat, Jean-Marie Basset,*
Anne Lesage, and Lyndon Emsley*
[4] a) G. K. Friestad, J. Qin, J. Am. Chem. Soc. 2000, 122, 8329; b) G. K.
The structural characterizationof heterogeneous catalysts
remains difficult. In the late eighties, Feher and later others
showed that polyhedral oligomeric silsesquioxanes (POSS)
canbe used to model the surface structure of partially
dehydroxylated silica and therefore infer information about
the coordination sphere of metal complexes directly support-
Friestad, J. Qin, J. Am. Chem. Soc. 2001, 123, 9922.
ˆ
[5] For reviews of allyl organometallic addition to C N bonds, see: Y.
Yamamoto, N. Asao, Chem. Rev. 1993, 93, 2207; E. F. Kleinman, R. A.
Volkmann in Comprehensive Organic Synthesis, Vol. 2 (Eds.: B. M.
Trost, I. Fleming), Pergamon, New York, 1991, pp. 975 1006; see also
ref. [1]. For selected recent examples, see:a) allyltin: H. Nakamura, K.
Aoyagi, J.-G. Shim, Y. Yamamoto, J. Am. Chem. Soc. 2001, 123, 372;
F. M. Guerra, M. R. Mish, E. M. Carreira, Org. Lett. 2000, 2, 4265;
J. A. Marshall, K. Gill, B. M. Seletsky, Angew. Chem. 2000, 112, 983;
Angew. Chem. Int. Ed. 2000, 39, 953; b) allylboron: K. Watanabe, S.
Kuroda, A. Yokoi, K. Ito, S. Itsuno, J. Organomet. Chem. 1999, 581,
103; c) allyltitanium: X. Teng, Y. Takayama, S. Okamoto, F. Sato, J.
Am. Chem. Soc. 1999, 121, 11916; d) allylmagnesium: S. Maiorana, C.
Baldoliu, P. D. Buttero, E. Licandro, A. Papagni, M. Lanfranchi, A.
Tiripicchio, J. Organomet. Chem. 2000, 593 594, 380; A. G. Steinig,
D. M. Spero, J. Org. Chem. 1999, 64, 2406; e) allylsamarium: R.
Yanada, M. Okaniwa, A. Kaieda, T. Ibuka, Y. Takemoto, J. Org.
Chem. 2001, 66, 1283; f) allylindium: H. M. S. Kumar, S. Anjaneyulu,
E. J. Reddy, J. S. Yadav, Tetrahedron Lett. 2000, 41, 9311; T. H. Chan,
W. Lu, Tetrahedron Lett. 1998, 39, 8605; g) allylgermanium: T.
Akiyama, J. Iwai, Y. Onuma, H. Kagoshima, Chem. Commun. 1999,
2191.
[1]
ed onoxides such as silica. This approach canalso provide
mechanistic insights on surface chemistry processes. However,
no direct comparison between these models and their related
surface complexes has beenreported.
We have recently shown that the reaction of tris(neopen-
ˆ
tyl)neopentylidene tantalum, [Ta( CHtBu)(CH2tBu)3] (1)
ꢀ
ˆ
with SiO2-(500) gives a mixture of mono-[ SiO-Ta( CHtBu)-
(CH2tBu)2]
and
bisgrafted
surface
complexes
[2a]
ꢀ
ˆ
[( SiO)2Ta( CHtBu)(CH2tBu)], while its reactionwith a
SiO2-(700) provides the monografted species [ SiO-Ta-
( CHtBu)(CH2tBu)2] (2s) as the sole surface species. The
characterization was based on elemental analysis, the evolu-
tion of neopentane during grafting of 1, and solvolysis of 2s as
well as infrared (IR) spectroscopy.[2b]
ꢀ
ˆ
[6] a) X. Fang, M. Johannsen, S. Yao, N. Gathergood, R. G. Hazell, K. A.
Jorgensen, J. Org. Chem. 1999, 64, 4844; b) D.-K. Wang, Y.-G. Zhou,
Y. Tang, X.-L. Hou, L.-X. Dai, J. Org. Chem. 1999, 64, 4233; c) A. S.
Pilcher, P. DeShong, J. Org. Chem. 1996, 61, 6901.
[7] a) Sluggish reactivity of allylsilanes toward neutral imines has been
noted: see refs. [1] and [5a]; b) although allylsilane addition to
iminium ions is effective, a strong Lewis acid such as TiCl4 or BF3 ¥
Et2O is often required; for a review, see: I. Fleming, J. Dunogus, R.
Smithers, Org. React. 1989, 37, 57; recent examples: J. V. Schaus, N.
Jain, J. S. Panek, Tetrahedron 2000, 56, 10263; T. Akiyama, M. Sugano,
H. Kagoshima, Tetrahedron Lett. 2001, 42, 3889.
We report here the use of high-resolutionsolid-state one-
dimensional (1D) and two-dimensional (2D) NMR spectros-
copy to study 2s and the use of high-resolution solution-state
1D and 2D NMR spectroscopy of its molecular analogue
À À
ˆ
[(c-C5H9)7Si7O12Si O Ta( CHtBu)(CH2tBu)2] (2m) to define
the structure of 2s at a molecular level and to investigate the
reactionpathway leading to the grafted species.
[8] A. S. Pilcher, H. L. Ammon, P. DeShong, J. Am. Chem. Soc. 1995, 117,
5166; C. J. Handy, Y.-F. Lam, P. DeShong, J. Org. Chem. 2000, 65,
3542.
[9] The optical rotationof 4 ([a]2D0 ˆ À2.0 (c ˆ 3.4 inCHCl 3)) was within
experimental error of an authentic sample ([a]2D0 ˆ À2.2 (c ˆ 3.4 in
CHCl3)) prepared from the known homoallylic amine: T. Basile, A.
Bocoum, D. Savoia, A. Umani-Ronchi, J. Org. Chem. 1994, 59, 7766.
[10] Hydrogenation of 3g gives 6, identical to that from n-propyl radical
additionto 2g [Eq. (2)]. See ref. [4b].
¬
[*] Dr. C. Coperet, Dr. J.-M. Basset, M. Chabanas, Dr. J. Thivolle-Cazat
¬
Laboratoire de Chimie Organometallique de Surface
UMR-9986 CNRS ESCPE Lyon, 43 blvd du 11 Novembre 1918
69626 Villeurbanne Cedex (France)
Fax : (‡33)4-7243-1795
Dr. E. A. Quadrelli
Dipartimento di Chimica e Chimica Industriale
Via Risorgimento 35, 56100 Pisa (Italy)
B. Fenet
O
O
Centre de RMN UCB Lyon 1 ESCPE Lyon
43 bd du 11 Novembre 1918, 69626 Villeurbanne Cedex (France)
H2 (1 atm)
nPr•
N
2g
(2)
3g
HN
Et
Pd/C
Dr. L. Emsley, Dr. A. Lesage
Laboratoire de Stereochimie et des Interactions Moleculaires
radical
addition
CH2Ph
(100%)
¬ ¬
¬
nPr
¬
UMR-117 CNRS ENS Lyon, Ecole Normale Superieure de Lyon
69364 LyonCedex (France)
6
[**] We are also indebted to the CNRS, ENS Lyon, and ESCPE Lyon for
financial support. M.C. is grateful to the French ministry of education,
research, and technology (MENRT) for a pre-doctoral fellowship.
[11] The allylsilane addition fails in 10:1 CH2Cl2/water, which would not
likely interfere with allylindium addition (see ref. [5f]).
[12] Similar explanations have been advanced regarding the enhanced
reactivity of bis(allyl)silanes. A. Shibato, Y. Itagaki, E. Tayama, Y.
Hokke, N. Asao, K. Maruoka, Tetrahedron 2000, 56, 5373.
¡
E.A.Q. gratefully acknowledges Universita di Pisa and S.N.A.M. for
financial support. 2D HETCOR ˆ two-dimensional heteronuclear
correlation.
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
Angew. Chem. Int. Ed. 2001, 40, No. 23
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4023-4493 $ 17.50+.50/0
4493

COMMUNICATIONS
As previously described, the surface complex 2s, was
a contact time of 2 ms (Figure 1c), shows clear correlations
betweenthe signals at d ˆ 4.2 and 1.0 in the 1H dimension (F1)
with those at d ˆ 246 and 33.5, respectively, in the carbon
dimension (F2). The signal at d ˆ 96.5 also correlates with the
peak centered at d ˆ 1.0, strongly suggesting that the meth-
ylene protons are buried under this massive signal (the
chemical shifts of the corresponding protons for the
surface model 2m are indeed around 1 ppm, see Table 1 and
below).
ˆ
prepared by treating a solution of [Ta( CHtBu)(CH2tBu)3]
inpentane (1.1 equivalents with respect to surface silanols;
0.7 0.8 accessible OH nmÀ2) with SiO2-(700) [2b] After workup,
.
2s was obtained as a yellow powder typically containing
4.2%wt of Ta. Noteworthy is the rapid (within1 mi)n
coloration of silica and discoloration of the supernatant
solution upon addition of the pentane solution of 1 to
SiO2-(700). All the observations and chemical analyses are
consistent with the proposed structure of 2s [Eq. (1)], but
further corroborationat the molecular level has beensought
using both solid-state NMR spectroscopy of the grafted
species and solution NMR spectroscopy of molecular model
species.
A reactionintermediate canbe detected if the preparation
*
of 2s involves a rather short (5 min or less) or a cold (below
15 208C) drying step; under such conditions, an extra signal
appears inthe solid-state NMR spectrum at d ˆ 110 (Fig-
ure 1b), which readily disappears upontreating the sample
under vacuum or under heat. During this step, neopentane is
evolved. All these observations suggest that the intermediate
ˆ
ꢀ
ꢀ
À
ˆ
1 ‡ SiOH À! [ SiO Ta( CHtBu)(CH2tBu)2] ‡ NpH
(1)
SiO2-(700) 2s
observed inthe reactionof [Ta(
CHtBu)(CH2tBu)3] with
ꢀ
À
SiO2-(700), is the tetraalkyl derivative [ SiO Ta(CH2tBu)4] (3s)
which is formed under kinetic control. The intermediate 3s
rapidly releases neopentane, to finally yield 2s as the sole
surface complex.
To improve the signal-to-noise ratio in the solid-state NMR
spectra of 2, the complex was partially 13C enriched at the a-
ˆ
positionby reactionof [Ta( C*HtBu)(C*H2tBu)3] (1*) with
ꢀ
À
ˆ
SiO2-(700), which yielded [ SiO Ta( C*HtBu)(C*H2tBu)2]
To corroborate the assignments proposed above and to gain
further insight in the reaction mechanism suggested by the
detectionof intermediate 3s, we have prepared a molecular
model of 2m using the silsesquioxane (c-C5H9)7Si7O12SiOH (4)
a Feher-cage analogue,[1d,f] that canmimic anisolated silanol
of a silica surface partially dehydroxylated at 7008C. The
thermodynamic product of the reaction between the mono-
1
*
*
(2s ). The solid-state H NMR spectrum of 2s shows two
broad resonance signals at d ˆ 4.2 and 1.0 (see the Supporting
Information),[3] and its 13C solid-state cross polarization/magic
angle spinning (CP/MAS) NMR spectrum displays strong
signals at d ˆ 246, 96.5, and 33.5 (Figure 1a). Moreover, the
2D HETCOR solid-state NMR spectra provided anunequiv-
*
ˆ
ocal assignment for 2s (see Supporting Information for
silanol 4 and [Ta( CHtBu)(CH2tBu)3] is 2m (for characteristic
experimental details). The 2D NMR spectrum, recorded with
data, see Table 1) alongside one equivalent of neopentane.[4]
The NMR spectroscopy data
obtained for 2m are ingood
agreement with those of its sol-
id-state counterpart, 2s. Based on
NMR spectroscopy data, the re-
action is under kinetic control
and reaches completion after
48 h. The kinetic monitoring of
the reactionshows that a color-
less intermediate, [(c-C5H9)7-
Si7O12SiOTa(CH2tBu)4] (3m) is
formed immediately uponmix-
ing of the reagents. The long-
lived intermediate 3m (t1/2 ˆ 9 h)
has beencharacterized by 1D
and 2D solution NMR spectros-
copy, which yields data ingood
agreement with those obtained
for its solid counterpart 3s (Ta-
ble 1). The decompositionpath
of complex 3m consists of the
elimination a neopentane unit
by means of a-H abstractionto
yield the orange complex 2m.
The combination of the data
obtained in the solid state for the
synthesis of 2s and in solution for
2m allows a mechanistic pathway
to be proposed for the reaction:
1
13
Figure 1. a) CP/MAS 13C solid-state NMR spectra of 2s , b) 2s and 3s , c) 2D H- C HETCOR solid-state NMR
*
*
*
spectra of 2s (2 ms of contact time). d) Proton traces of the signals at d ˆ 246, 96.5, and 33.5 in the 2D 1H
*
13C NMR spectrum.
4494
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4023-4494 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 23

COMMUNICATIONS
Table 1. 1H an d13C chemical shifts of 1, 2m, an d3m inC D6.
end products, and determination of the grafting-reaction
mechanism. Moreover, we have shown that the Feher cage
analogue 4, is a good model for SiO2-(700), a silica partially
dehydroxylated at 7008C, which contains mainly isolated
silanols, and provides a good estimation of the chemical shifts
for surface complexes and can help in understanding elemen-
tary steps of surface reactions.
6
Com-
plex resonance[a] assignement
1H NMR [d]
13C NMR [d]
resonance
assignement
ˆ
1
0.81
1.14
1.45
1.89
CH2C(CH3)3
CH2C(CH3)3
34.7
34.95
35.05
47.0
113.75
249.9
22.6, 22.65
CHC(CH3)3
CH2C(CH3)3
ˆ
CHC(CH3)3
CHC(CH3)3
CH2C(CH3)3
ˆ
ˆ
CHC(CH3)3
CH2C(CH3)3
ˆ
CHC(CH3)3
(c-CH(CH2)4)
Received: April 2, 2001
Revised: September 18, 2001 [Z16881]
2m
1.1 2.1(m)[b] (c-CH(CH2)4)
27.4, 27.45, 27.85, 27.9 (c-CH(CH2)4)
CHaHbC(CH3)3 33.95
CHaHbC(CH3)3 34.4
1.03(d)[c]
1.20(d)[c]
1.20
1.41
4.77
CH2C(CH3)3
ˆ
CHC(CH3)3
CH2C(CH3)3
CH2C(CH3)3
34.75
45.55
95.6
245.2
22.7, 22.6,
22.55
ˆ
ˆ
CHC(CH3)3
CHC(CH3)3
CH2C(CH3)3
ˆ
CHC(CH3)3
[1] For some examples onthe preparationof Feher cages or analogues
(POSS) and their reactivity with organometallics, see: a) F. J. Feher,
D. N. Newman, J. F. Walzer, J. Am. Chem. Soc. 1989, 111, 1741; F. J.
Feher, D. N. Newman, J. Int. Biomed. Inf. Data 1990, 112, 1931; b) F. J.
Feher, K. J. Weller, J. J. Schwab, Organometallics 1995, 14, 2009; c) F. J.
Feher, T. A. Budzichowski, Polyhedron 1995, 14, 3239; d) R. Ducha-
teau, H. C. L. Abbendhuis, R. A. Van Santen, S. K.-H. Thiele, M. F. H.
VanTol, Organometallics 1998, 17, 5222; e) R. Duchateau, H. C. L.
Abbendhuis, R. A. Van Santen, A. Meetsma, S. K.-H. Thiele, M. F. H.
VanTol, Organometallics 1998, 17, 5663; f) R. Duchateau, U. Cremer,
R. J. Harmsen, S. I. Mohamud, H. C. L. Abbendhuis, R. A. Van Santen,
A. Meetsma, S. K.-H. Thiele, M. F. H. Van Tol, M. Kranenburg,
Organometallics 1999, 18, 5447; g) for a review onvarious surface
models other thanPOSS, see: R. Murugavel, A. Voigt, M. G.
Walawalkar, H. W. Roesky, Chem. Rev. 1996, 96, 2205.
ˆ
CHC(CH3)3
c-CH(CH2)4)
3m
1.1 2.1(m)[b] (c-CH(CH2)4)
1.34
1.87
CH2C(CH3)3
CH2C(CH3)3
27.45, 27.55, 27.75, 27.9 (c-CH(CH2)4)
[d]
CH2C(CH3)3
CH2C(CH3)3
CH2C(CH3)3
35.2
108.7[b]
[a] multiplicity in parenthesis; singlets are not specified. [b] broad signal.
[c] J ˆ 14.0 Hz. [d] Not detected.
addition of the silanol to the alkylidene moiety of 1 (1 !3s)
followed by an a-H abstraction( 3s !2s; Scheme 1).[5]
[2] a) V. Dufaud, G. P. Niccolai, J. Thivolle-Cazat, J.-M. Basset, J. Am.
Chem. Soc. 1995, 117, 4288 4294; b) L. Lefort, M. Chabanas, O. Maury,
The reactivities of 4 and SiO2-(700) are relatively similar
(immediate reactionof 1 with silanols and a subsequent
slower a-H abstractionstep), which is probably because of
similar pKa values[6] and steric crowding. Finally, 2D HET-
COR solid-state NMR spectroscopy is animportant tool to
¬
D. Meunier, C. Coperet, J. Thivolle-Cazat, J.-M. Basset, J. Organomet.
Chem. 2000, 593 594, 96.
[3] The natural abundance 13C NMR spectrum displays three signals at d ˆ
ˆ
À
33.5 (C(CH3)3), 96.5 (CH2C(CH3)3), and 246 ( CH C(CH3)3), albeit
with a poor signal/noise ratio, especially for the latter. Therefore, 2s was
1
unequivocally assign H an d13C chemical shifts for surface
partially 13C enriched at the a-positionto the metal, 2s , by reactionof
*
complexes, and will be important in the future development of
surface organometallic chemistry. In conclusion, these data
taken together show, besides the analogy between surface and
solutionchemistry, that the combined use of high-resolution
solid state NMR and solution NMR spectroscopy to inves-
tigate molecular analogues, is a powerful tool in surface
organometallic chemistry, and enables the identification of
ˆ
[Ta( C*HtBu)(C*H2tBu)3] (1*) with SiO2-(700)
[4] These values are different from those obtained by Wolczanski and co-
.
ˆ
workers for [(tBu)3SiOTa( CHtBu)(CH2tBu)2]:R. E. Lapointe, P. T.
Wolczanski, G. D. Van Duyne, Organometallics 1985, 4, 1810. This
illustrates the dramatic effects onthe NMR data of substituents onSi
and the necessity of the oxygen atoms around the Si centers.
[5] This was infact indirectly showninthe studies of the formationof the
surface complexes, see ref. [2a]. Noteworthy is the reactionof HCl on
Scheme 1. Comparisonof the reactionof silica partially dehydroxylated at 700 8C (SiO2(-700)) versus (c-C5H9)7Si7O12SiOH.
Angew. Chem. Int. Ed. 2001, 40, No. 23
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4023-4495 $ 17.50+.50/0
4495

COMMUNICATIONS
ˆ
À
[Ta( CHtBu)(CH2tBu)3] which yields Cl Ta(CH2tBu)4: R. R. Schrock,
J. D. Fellmann, J. Am. Chem. Soc. 1978, 100, 3359.
OBz
O
OBz
[6] a) For comprehensive information on the pKa values of silicas, see: The
Surface Properties of Silicas (Ed.: A. P. Legrand), Wiley, New York,
1998, chapter 4B; b) for related comments on 4, see ref. [1f].
Me
Me
HO
HO
H
OH
1
a)
+
2 (70%)
OBz
3 (10%)
O
O
OH
HO
O
O
OBz
a)
(23%)
Ph
O
H
Ph
HO
H
OH
O
4
5
Asymmetric Permanganate-Promoted
Oxidative Cyclization of 1,5-Dienes by Using
Chiral Phase-Transfer Catalysis**
Scheme 1. Oxidative cyclizationof dieens
(2 equiv), AcOH (4 equiv), Adogen464 (0.4 equiv)/Et O. Bz ˆ benzoyl.
1 and 4. a) 0.4m KMnO4
2
Table 1. Oxidative cyclizationof geranyl benzoate ( 1) (see Scheme 1).[a]
Richard C. D. Brown* and John F. Keily
Entry
Solvent
Phase-transfer
agent
Yield of 2
[%]
Yield of 3
[%]
2,5-Bis(hydroxymethyl)-substituted tetrahydrofurans (THF
diols) are an important structural unit found in many natural
and synthetic, biologically active molecules.[1] Anelegatn
approach to the synthesis of THF diols is by the permanga-
nate-promoted oxidative cyclization of 1,5-dienes, which leads
to the stereospecific formation of up to four new stereocenters
[b]
1
2
3
4
CH2Cl2
CH2Cl2
Et2O
TBAB
TBAB
TBAB
Adogen464
8[c]
not recorded
44[d]
21[d]
7[d]
40[d]
[d]
Et2O
70
10[d]
[a] Reactions were conducted on a 0.19 mmol scale. TBAB ˆ tetrabutyl-
ammonium bromide. [b] No AcOH added to the reaction mixture. [c] Yield
estimated by HPLC. [d] Yields represent analytically pure isolated
material.
6]
and the THF ring.[2 The same transformation has been
carried out using other metal oxo reagents and catalysts.[7, 8]
In the permanganate-promoted oxidative cyclization reaction,
control of absolute stereochemistry has been achieved with
chiral auxiliaries;[4a,b] however, preparation of enantiomeri-
cally enriched THF diols by oxidative cyclizations of achiral
1,5-dienes has not been reported. Herein we report the novel
high-yielding phase-transfer permanganate-promoted oxida-
tion of geranyl benzoate (1) to the corresponding THF diol 2
(Scheme 1), and a procedure for the asymmetric oxidative
cyclization of phenone dienes to produce enantiomerically
enriched THF diols.
The use of phase-transfer conditions in permanganate-
promoted oxidations is well known,[9] although we were not
aware of applicationto the oxidative cyclizationof 1,5-dienes.
Therefore before attempting to develop an asymmetric
process, we investigated the phase-transfer oxidative cycliza-
tionof a readily available substrate 1 using achiral tertiary
ammonium salts (Scheme 1).[10]
albeit inlower yield (Scheme 1). We thenreturned to our
primary goal of developing an asymmetric oxidative cycliza-
tion.
Studies on the oxidation of cinnamate esters by quaternary
ammonium permanganates led Lee et al. to conclude that the
transition state was electron-rich, and they proposed an
enolate-like structure 6 with stabilizationdue to interaction
with the ammonium ion (Scheme 2).[11] The similarity of 6 and
the initial adduct 7 formed inthe phase-transfer-catalyzed
nucleophilic epoxidation of enones suggested that chiral
catalysts developed for the epoxidationreactionmight also
be applied to the oxidative cyclization.[12]
R4N
O
O
Oxidationof dieen
1 with potassium permanganate
Ar
Ar
OMe
OMe
O
O
(2 equiv) indiethyl ether buffered with ethanoic acid gave
the best yield of THF diol 2 (70%) whenAdogen464 (0.1
1 equiv) was employed (entry 4, Table 1). Oxidation of enone
4 under the same conditions gave the desired THF diol 5,
O
O
Mn
Mn
O
O
O
R4N
O
6
O
R4N
O
[*] Dr. R. C. D. Brown, J. F. Keily
Department of Chemistry
O
R1
R2
R1
O
R2
University of Southampton
Highfield, SouthamptonSO17 4BJ (UK)
Fax : (‡44)23-80596805
Cl
7
Scheme 2. Possible reactionpathways for the permanganate-promoted
oxidation of electron-poor olefins (top), and the nucleophilic epoxidation
of a,b-enones with ClOÀ (bottom).
E-mail: rcb1@soton.ac.uk
[**] This work was supported by the EPSRC (J.F.K.) and the Royal Society
(R.C.D.B.). We thank Dr. Barry Lygo (University of Nottingham) for
helpful discussions, and Syngenta and Merck Sharp and Dohme for
unrestricted grants.
Oxidative cyclization of geranyl benzoate (1) inthe
presence of 8 in dichloromethane/water/ethanoic acid at
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
4496
¹ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4023-4496 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 23