A new zincate-mediated rearrangement reaction of 2-(1-hydroxyalkyl)-1- alkylcyclopropanol

Article abstract of DOI:10.1002/chem.200901054

A novel rearrangement of 2-(1-hydroxyalkyl)-1-alkylcyclopropanol has been found. It proceeds in the presence of a catalytic amount of organozinc ate complex to give vic-diols. The rearrangement can be applied to various types of 2-(1-hydroxyalkyl)-1-alkyl-cyclopropanol, which can be easily prepared from the corresponding α,β-ep-oxyketones and bis(iodozincio)methane. When bicyclo[13.1.0]-pentadecane-1, 15-diol was treated with the organozinc ate complex, the corresponding 14-membered cyclic vic-diol was obtained. Thus, this rearrangement is also useful for changing the ring size of cyclic substrates.

Full text of DOI:10.1002/chem.200901054

FULL PAPER
DOI: 10.1002/chem.200901054
A New Zincate-Mediated Rearrangement Reaction of 2-(1-Hydroxyalkyl)-1-
alkylcyclopropanol
Kenichi Nomura and Seijiro Matsubara*[a]
Abstract: A novel rearrangement of 2-
(1-hydroxyalkyl)-1-alkylcyclopropanol
has been found. It proceeds in the pres-
ence of a catalytic amount of organo-
zinc ate complex to give vic-diols. The
rearrangement can be applied to vari-
ous types of 2-(1-hydroxyalkyl)-1-alkyl-
cyclopropanol, which can be easily pre-
pared from the corresponding a,b-ep-
oxyketones and bis(iodozincio)me-
ACHTUNGTRENNpUNG entadecane-1,15-diol was treated with
the organozinc ate complex, the corre-
sponding 14-membered cyclic vic-diol
was obtained. Thus, this rearrangement
is also useful for changing the ring size
of cyclic substrates.
thane.
When
bicycloACTHNGUTRENNU[G 13.1.0]-
Keywords: cyclopropanol · homo-
geneous catalysis · organozinc ate
complexes · rearrangement · zinc
Introduction
Cyclopropanes have been rec-
ognized as variants of C-3
open-chain units in organic
synthesis,[1] and a large number
of cyclopropane ring-opening
reactions have been reported.[2]
Scheme 1. Julia-type transformation of 2-(1-hydroxyalkyl)-1-alkylcyclopropane and 2-(1-hydroxyalkyl)-1-alkyl-
cyclopropanol.
Triggered by the formation of
cation or radical at
carbon adjacent to the cyclo-
a
a
a
propane ring, most of the reactions proceed to release the
strain of the three-membered ring.[3] In 1960, Julia et al. re-
ported an acid-catalyzed ring-opening isomerization of cy-
clopropyl alkyl carbinol (Scheme 1, top).[4] The reaction af-
fords an (E)-homoallylic alcohol via the formation of a car-
bocation on the cyclopropyl-substituted carbon. We have re-
ported a preparation of 2-(1-hydroxyalkyl)-1-alkylcyclopro-
(TFA) gives the corresponding b,g-unsaturated ketone 4, as
shown in Scheme 1, bottom.[6,7]
We also tried to convert the zinc alkoxide of 2-(1-hydrox-
yalkyl)-1-alkylcyclopropanol (3a’) directly into the corre-
sponding b,g-unsaturated ketone by making use of the
Lewis acidity of the zinc species in the reaction mixture.[8]
a,b-Epoxyketone 2a (0.5 mmol) was treated with an excess
amount of bis(iodozincio)methane (1, 1.25 mmol) at 258C,
then the reaction mixture was heated at reflux for 2 h to
give desired b,g-unsaturated ketone 4a in good yield
(Scheme 2). However, treatment of 2a with 1.1 equivalents
of 1 gave 2,3-dihydroxy-3-phenyl-4-pentene (5a) as the
panol
3 from a,b-epoxyketone 2 and bis(iodozincio)-
methane 1.[5] When the Julia-type reaction was applied to 2-
(1-hydroxyalkyl)-1-alkylcyclopropanol 3, the carbocation-
mediated ring-opening reaction with trifluoroacetic acid
[a] K. Nomura, Prof. S. Matsubara
Department of Material Chemistry
Graduate School of Engineering, Kyoto University
Kyoutodaigaku-katsura
Nishikyo-ku, Kyoto 615-8510 (Japan)
Fax : (+81)75-383-2463
E-mail: matsubar@orgrxn.mbox.media.kyoto-u.ac.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901054.
Scheme 2. Tandem transformation of 2a into 4a by a [2+1] cycloaddition
of bis(iodozincio)methane (1).
Chem. Eur. J. 2010, 16, 703 – 708
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
703

alkoxy and butyl groups. The
reaction became sluggish if the
proportion of butyllithium to
zinc(II) chloride was decreased
(Table 1, entries 5 and 6) be-
cause the conditions would not
allow formation of a zincate
complex,
(RO)2BuZnLi. A stoichiomet-
ric amount of zincate
(tBu3ZnLi),[12,13] which can be
such
as
Scheme 3. A novel rearrangement reaction of 3a’ or [D2]-3a’.
major product instead of b,g-unsaturated ketone 4a
Table 1. The rearrangement of 3b into 5b.
(Scheme 3a).[9,10] When the latter reaction was examined by
using the labeled reagent CD2ACHTNUGTRNEUNG(ZnI)2 ([D2]-1), 5,5-dideuter-
io-3-phenylpent-4-ene-2,3-diol ([D2]-5a) was obtained
(Scheme 3b). We investigated the details of this novel rear-
rangement.
Entry
Reagent
CH2A(ZnI)2 (1 equiv)
Me2Zn (2 equiv)
BuLi (3 equiv)
BuLi (3 equiv)/ZnCl2 (1 equiv)
BuLi (3 equiv)/ZnCl2 (2 equiv)
BuLi (2 equiv)/ZnCl2 (1 equiv)
tBu3ZnLi (1 equiv)
tBu3ZnLi (0.2 equiv)/BuLi (2 equiv)
tBu3ZnLi (0.1 equiv)/BuLi (2 equiv)
t [h]
Yield[a] [%]
1
2
3
4
5
6
7
8
9
G
3
3
3
3
3
3
3
6
18
0[b]
0[b]
88
53
0[c]
70
88
65
Results and Discussion
We speculated that the reaction proceeded via a zincate in-
termediate. The working hypothesis of this rearrangement is
shown in Scheme 4. Zinc alkoxide 3a’’, which was formed
12
[a] The yield was determined by using 1H NMR spectroscopy with bro-
moform as the internal standard. [b] The starting material was recovered
completely. [c] A complex mixture was obtained.
prepared easily from ZnCl2 and tBuLi, was also effective in
this rearrangement (Table 1, entry 7). Treatment of the pre-
pared lithium alkoxide of 3b with a catalytic amount of
tBu3ZnLi also gave 5b (Table 1, entries 8 and 9).
We traced the rearrangement of 3b with reactIR
(Figure 1).[14] The reaction proceeded gradually, and was
almost complete after 4.5 h.
To determine the generality of the rearrangement, we
treated various lithium alkoxides of 2-(1-hydoxyalkyl)-1-al-
kylcyclopropanol 3 with the zincate complex as a catalyst
(Table 2). Substrates 3c and 3d gave the corresponding 1,2-
diols in good yields (Table 2, entries 1 and 2). Cyclopropyl
alkyl carbinols were also converted into the corresponding
1,2-diols in good yields with a low diastereoselectivity
Scheme 4. A working hypothesis of the rearrangement.
from 3a’ under heating, would coordinate with the remain-
ing 1. Zincate A may fragment into aldehyde and carbanion
fragments (B) due to the strong electron-donating effect of
the zincate. The aldehyde and carbanion fragments then
form a tight pair through zinc (C) and give 5a’ through nu-
cleophilic addition.[11]
In considering the role of the zincate complex, we opti-
mized the rearrangement of 2-hydroxymethyl-1-phenylcyclo-
propanol (3b; Table 1). Treatment of 3b with two equiva-
lents of dimethylzinc (Table 1, entry 2) or three equivalents
of butyllithium (Table 1, entry 3) resulted in complete recov-
ery of the starting material. One equivalent of zinc(II) chlo-
ride and three equivalents of butyllithium, however, gave
the desired product in 88% yield (Table 1, entry 4). The
combination was expected to form a zincate complex with
Figure 1. ReactIR results for the conversion of 3b into 5b.
704
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 703 – 708

Rearrangement of 2-(1-Hydroxyalkyl)-1-alkylcyclopropanol
FULL PAPER
Table 2. Examples of the rearrangement of 3 into 5.
Entry
1
Substrate
Product
t [h]
Yield[a] [%]
89
24
Scheme 5. The reaction of a mono-protected substrate under the rear-
rangement conditions.
2
3
24
24
89
of the diols. The rearrangement did not proceed with sub-
strates 6 or 7, which have ether groups. In the case of 6, the
corresponding b,g-unsaturated ketone 8 was obtained quan-
titatively via cyclopropane ring-opening followed by elimi-
nation of methoxy group. Treatment of 7 resulted in com-
plete recovery of the starting material. These results demon-
strate the importance of zinc dialkoxide formation (3a’’ in
Scheme 4) in the rearrangement.
78 d.r. (3:2)[b,c]
4
24
98 d.r. (2:1)[b,d]
When the rearrangement is applied to the lithium alkox-
ide of a bicyclic compound, a ring-contracting reaction
occurs. In the case of bicycloACHTNUTRGNE[UNG 13.1.0]hexadecane-1,14-diol
5
24
1
86 d.r. (3:1)[b,e]
(3i),[15] the zincate-catalyzed reaction gave 14-membered
diol 5i (1-vinylcyclotetradecan-1,2-diol; Scheme 6). The ste-
reochemistry of the major product is (1R*,2S*). With the
same substrate, the Julia-type reaction with trifluoroacetic
acid gave the corresponding ring-expanded product 9.
Meanwhile, treatment of the lithium alkoxide of spirocyclic
diol 3j with tBu3ZnLi gave the corresponding ring-expanded
product 5j in good yield (Scheme 7).
6[f]
57[g]
7[h]
1
n.d.[i]
[a] Isolated yields. [b] The diastereomeric ratios were determined by
using 1H NMR spectroscopy. [c] The stereochemistry of the major prod-
uct was (2S*,3R*). [d] The stereochemistry of major product was
(2S*,3S*). [e] The stereochemistry of the major product was (2S*,3R*).
[f] The reaction was carried out at 478C. [g] Propiophenone was also ob-
tained in 30% yield. [h] The reaction was carried out at 608C. [i] Propio-
phenone and cyclohexanone were obtained in 68 and 52% yield, respec-
tively.
The substrate, 2-(1-hydroxyalkyl)-1-alkylcyclopropanol 3,
is prepared from a,b-epoxyketone 2 and dizinc 1.[5] We ap-
plied this method to a,b-aziridinylketone 10 and obtained 2-
(1-aminoalkyl)-1-alkylcyclopropanol 11. Product 11 was also
available as a substrate of the rearrangement and, in this
case, the corresponding vic-amino alcohol 12 was obtained
as a single isomer (Scheme 8).
Enantioenriched 3 can be easily obtained from optically
active 2, which can be prepared by a Katsuki–Sharpless
asymmetric epoxidation followed by a Swern oxidation. If
the rearrangement proceeds in a stereospecific manner, this
(Table 2, entries 3–5). In the case of cyclopropyl dimethyl
carbinol 3g, the corresponding diol was obtained in 57%
yield (Table 2, entry 6). It is worth noting in this case that
propiophenone was also ob-
tained in 30% yield. More-
over, when substrate 3h was
treated under these conditions
at 608C, 1-phenyl-1-propanone
and cyclohexanone were ob-
tained (Table 2, entry 7). In
this case, the rearranged prod-
uct was not obtained. The ke-
tones may come from the reac-
tion
intermediate
(B
in
Scheme 4).[11]
As shown in Scheme 5, we
attempted the rearrangement
with the mono-protected form
Scheme 6. The rearrangement of 3i to 5i.
Chem. Eur. J. 2010, 16, 703 – 708
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
705

S. Matsubara and K. Nomura
Scheme 7. The rearrangement of 3j to 5j.
Scheme 8. The transformation of 10 to 12 by a [2+1] cyclization with bi-
s(iodozincio)methane.
can be a useful method for constructing enantioenriched
vic-diols. Therefore, we studied the stereochemistry of the
rearrangement in detail.
~
&
Figure 2. Measurement of the ee of (1R,2R)-3b ( ) and (S)-5b ( ) during
the reaction.
The reaction with optically pure (1R,2R)-3b gave (S)-
5b[16] with a loss of enantiomeric purity (Scheme 9). To ex-
Figure 2. A decrease in the enantiomeric purity of 3b was
observed, and that of 5b also was gradually lost at the same
time. The result in Figure 2 implies that the enantiomeric
purity of product 5b depends on that of starting material
3b. The rearrangement may proceed in a stereospecific
manner, but the racemization of the substrate may deter-
mine the enantiomeric purity of the product.
Scheme 9. The rearrangement of optically active 3b.
It should be noted that the diastereomer of 3b was not
observed during the reaction, even though the enantiomeric
purity was gradually lost. The substrate recovered at each
point had a cis configuration. We suppose that the racemiza-
tion of 3b proceeds as shown in Scheme 11.[17] Zinc alkoxide
(1R,2R)-3b’ is converted into b-carbanion D by cyclopro-
plain the loss of enantiomeric purity during the reaction, we
propose three possibilities: 1) A racemization of the product
proceeds under the reaction condition; 2) the rearrangement
is not stereospecific; or 3) a
racemization of the substrate
proceeds under the reaction
condition. To clarify the cause
of the loss of enantiomeric
purity, we conducted the fol-
lowing experiment.
If the obtained optically
Scheme 11. Proposed mechanism for the isomerization of 3b’ under the reaction conditions.
active product, (S)-5b (59%
ee), was treated under the
same conditions as in the rearrangement, no significant loss
of enantiomeric purity was detected (Scheme 10). We also
tracked the enantiomeric purity of substrate (1R,2R)-3b and
product (S)-5b during the reaction. The result is shown in
pane ring-opening with an electron-donating effect from the
electron-rich zinc atom. Racemization of carbanion D may
proceed under the THF reflux conditions and give E. For-
mation of a cyclopropane ring from E gives the opposite en-
antiomer of the substrate, (1S,2S)-3b’.[18]
Based on all these results, we attributed the loss of enan-
tiomeric purity of the product to the racemization of the
substrate.
Scheme 10. Behavior of partially optically active (S)-5b under the rear-
rangement conditions.
706
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 703 – 708

Rearrangement of 2-(1-Hydroxyalkyl)-1-alkylcyclopropanol
FULL PAPER
[4] a) M. Julia, S. Julia, R. Guegan, Bull. Soc. Chim. Fr. 1960, 1072;
b) M. Julia, S. Julia, S.-Y. Tchen, Bull. Soc. Chim. Fr. 1961, 1849;
c) K. Sakaguchi, M. Fujita, Y. Ohfune, Tetrahedron Lett. 1998, 39,
4313; d) K. Sakaguchi, M. Higashino, Y. Ohfune, Tetrahedron 2003,
59, 6647; e) M. Honda, Y. Yamamoto, H. Tsuchida, M. Segi, T. Na-
kajima, Tetrahedron Lett. 2005, 46, 6465; f) M. Honda, T. Mita, T.
Nishizawa, T. Sano, M. Segi, T. Nakajima, Tetrahedron Lett. 2006,
47, 5751; g) S. R. Wilson, A. E. Davey, M. E. Guazzaroni, J. Org.
Chem. 1992, 57, 2007; h) C. Singh, S. Pandey, G. Saxena, N. Srivasta-
va, M. Sharma, J. Org. Chem. 2006, 71, 9057.
Conclusions
We found some novel rearrangements of lithium alkoxides
of 2-(hydroxyalkyl)-1-alkylcyclopropanols that were mediat-
ed by a zincate complex. Various types of substrates are
available and the reaction can be applied for fascinating
transformations of molecular skeleton. And at the same
time, we have proposed the ability of the zincate complex to
donate electrons.[19] This electron-pushing effect may apply
to other reactions.
[5] K. Nomura, K. Oshima, S. Matsubara, Angew. Chem. 2005, 117,
6010; Angew. Chem. Int. Ed. 2005, 44, 5860.
[6] K. Nomura, S. Matsubara, Synlett 2008, 1412.
[7] a) T. Sato, T. Kikuchi, N. Sootome, E. Murayama, Tetrahedron Lett.
1985, 26, 2205; b) T. Sato, T. Kikuchi, H. Tsujita, A. Kaetsu, N. Soo-
tome, K. Nishida, K. Tachibana, E. Murayama, Tetrahedron 1991,
47, 3281; c) C. Brꢁkner, H. Reissig, J. Org. Chem. 1988, 53, 2440;
d) C. Toratsu, T. Fujii, T. Suzuki, K. Takai, Angew. Chem. 2000, 112,
2837; Angew. Chem. Int. Ed. 2000, 39, 2725.
[8] Preparation of bis(iodozincio)methane was accompanied by forma-
tion of zinc(II) iodide. The Lewis acidity of zinc(II) iodide seems to
be sufficient for b,g-unsaturated ketone formation. See: S. Matsu-
bara, H. Yoshino, Y. Yamamoto, K. Oshima, H. Matsuoka, K. Mat-
sumoto, K. Ishikawa, E. Matsubara, J. Organomet. Chem. 2005, 690,
4859.
[9] K. Nomura, S. Matsubara, Chem. Commun. 2009, 2212.
[10] Treatment of an a,b-unsaturated ketone with an aldehyde in the
presence of chromium(II) gives 2-(1-hydroxyalkyl)-1-alkylcyclopro-
panol. Meanwhile, the same reaction with the addition of chlorotri-
methylsilane affords cross-pinacol coupling product. See: a) K.
Takai, R. Morita, C. Toratsu, Angew. Chem. 2001, 113, 1150; Angew.
Chem. Int. Ed. 2001, 40, 1116; b) K. Takai, R. Morita, H. Matsushi-
ta, C. Toratsu, Chirality 2003, 15, 17.
Experimental Section
General procedure for the rearrangement: tBu3ZnLi was prepared ac-
cording the procedure reported in reference [13g]. tBuLi (1.6m in pen-
tane, 3 mmol) was added to a solution of 2-(1-hydroxyalkyl)-1-alkylcyclo-
propanol (3, 1 mmol) in THF (8 mL) at 08C and stirred for 10 min. The
prepared tBu3ZnLi (THF solution, 0.2 mmol) was added to the solution
at 08C and the mixture was heated to reflux. Then saturated aqueous
NH4Cl was added to quench the reaction and the mixture was extracted
with ethyl acetate. The combined organic layers were washed with brine
and dried over Na2SO4. The solvent was removed under reduced pres-
sure, and the residue was purified by using silica gel column chromatog-
raphy to give the corresponding vic-diol.
Acknowledgements
[11] Still and Macdonald have reported that a-alkoxyallyl carbanions un-
dergo preferential protonation and alkylation at the unsubstituted
terminus and react with carbonyl compounds at the alkoxy-substitut-
ed terminus. See, W. C. Still, T. L. Macdonald, J. Org. Chem. 1976,
41, 3620.
[12] M. Isobe, S. Kondo, N. Nagasawa, T. Goto, Chem. Lett. 1977, 679.
[13] For representative examples, see: a) T. Harada, T. Katsuhira, D.
Hara, Y. Kotani, K. Maejima, R. Kaji, A. Oku, J. Org. Chem. 1993,
58, 4897; b) T. Harada, T. Katsuhira, K. Hattori, A. Oku, Tetrahe-
dron 1994, 50, 7987; c) Y. Kondo, M. Fujinami, M. Uchiyama, T. Sa-
kamoto, J. Chem. Soc. Perkin Trans. 1 1997, 799; d) T. Harada, T.
Kaneko, T. Fujiwara, A. Oku, J. Org. Chem. 1997, 62, 8966; e) T.
Harada, T. Kaneko, T. Fujiwara, A. Oku, Tetrahedron 1998, 54,
9317; f) Y. Kondo, T. Komine, M. Fujinami, M. Uchiyama, T. Saka-
moto, J. Comb. Chem. 1999, 1, 123; g) M. Uchiyama, T. Furuyama,
M. Kobayashi, Y. Matsumoto, K. Tanaka, J. Am. Chem. Soc. 2006,
128, 8404.
[14] The reactIR measurement was conducted every 30 s. Figure 1 shows
the absorption intensity of 3b and 5b plotted every 30 min to simpli-
fy the figure. See the Supporting Information for more details.
[15] Although 3i was used as a single diastereomer form, we could not
determine the relative stereochemistry of 3i.
[16] Compared with the result of specific rotation in the literature, it was
found that 5b has (S)-configuration. See: M. E. Vargas-Diaz, L.
Chacon-Garcia, P. Valezquez, J. Tamariz, P. Joseph-Nathan, L. G.
Zepeda, Tetrahedron: Asymmetry 2003, 14, 3225.
This work was supported financially by the Japanese Ministry of Educa-
tion, Culture, Sports, Science and Technology and by Kyoto University,
International Innovation Center. The authors acknowledge support from
the Global COE Program “Integrated Materials Science” (#B-09). K.N.
thanks the Japan Society for the Promotion of Science for Young Scien-
tists for Research Fellowships.
[1] For recent examples, see: a) S. K. Jackson, A. Karadeolian, A. B.
Driega, M. A. Kerr, J. Am. Chem. Soc. 2008, 130, 4196; b) T. Ok, A.
Jeon, J. Lee, J. H. Lim, C. S. Hong, H.-S. Lee, J. Org. Chem. 2007,
72, 7390; c) M. Tanaka, M. Ubukata, T. Matsuo, K, Yasue, K. Mat-
sumoto, Y. Kajimoto, T. Ogo, T. Inaba, Org. Lett. 2007, 9, 3331; d) T.
Tada, Y. Ishida, K. Saigo, Org. Lett. 2007, 9, 2083; e) A. Srikrishna,
R. R. Babu, P. C. Ravikumar, Synlett 2007, 0655; f) H. Fujii, Y. Osa,
M. Ishihara, S. Hanamura, T. Nemoto, M. Nakajima, K. Hasebe, H.
Mochizuki, H. Nagase, Bioorg. Med. Chem. Lett. 2008, 18, 4978;
g) N. T. Tzvetkov, T. Arndt, J. Mattay, Tetrahedron 2007, 63, 10497.
[2] For representative examples, see: a) D. H. Gibson, C. H. DePuy,
Chem. Rev. 1974, 74, 605; b) O. G. Kulinkovich, Chem. Rev. 2003,
103, 2597; c) R. I. Khusnutdinov, U. M. Dzhemilev, J. Organomet.
Chem. 1994, 471, 1–18; d) S. C. Wang, D. J. Tantillo, J. Organomet.
Chem. 2006, 691, 4386; e) M. Rubin, M. Runina, V. Gevorgyan,
Chem. Rev. 2007, 107, 3117; f) J. E. Baldwin, Chem. Rev. 2003, 103,
1197; g) H.-U. Reissig, R. Zimmer, Chem. Rev. 2003, 103, 1151;
h) M. Yu, B. L. Pagenkoph, Tetrahedron 2005, 61, 321.
[3] For examples, see: a) H. Rinderhagen, P. A. Waske, J. Mattay, Tetra-
hedron 2006, 62, 6589; b) T. Irshaidat, Tetrahedron Lett. 2008, 49,
5894; c) A. E. Nadany, J. E. Mckendrick, Tetrahedron Lett. 2007, 48,
4071; d) A. D. Allen, M. F. Fenwick, H. H. Riyad, T. T. Tidwell, J.
Org. Chem. 2001, 66, 5759; e) A. L. J. Beckwith, V. W. Bowry, J.
Org. Chem. 1989, 54, 2681; f) A. L. J. Beckwith, V. W. Bowry, J. Am.
Chem. Soc. 1994, 116, 2710; g) B. S. J. Blagg, M. B. Jarstfer, D. H.
Rogers, C. D. Poulter, J. Am. Chem. Soc. 2002, 124, 8846; h) K. B.
Wiberg, G. Szeimies, J. Am. Chem. Soc. 1968, 90, 4195.
[17] a) D. Seyferth, L. G. Vaughan, J. Am. Chem. Soc. 1964, 86, 883;
b) Y. D. Wu, K. N. Houk, J. A. Marshall, J. Org. Chem. 1990, 55,
1421; c) E. J. Verner, T. Cohen, J. Am. Chem. Soc. 1992, 114, 375;
d) K. Tomooka, T. Igarashi, M. Watanabe, T. Nakai, Tetrahedron
Lett. 1992, 33, 5795; e) R. Hoffmann, T. Rꢁckert, R. Brꢁckner, Tet-
rahedron Lett. 1993, 34, 297; f) K. Tomooka, T. Igarashi, T. Nakai,
Tetrahedron Lett. 1993, 34, 8139; g) R. Gawley, Tetrahedron Lett.
1999, 40, 4297.
Chem. Eur. J. 2010, 16, 703 – 708
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
707

S. Matsubara and K. Nomura
[18] Although we attempted to trap D and E with allyl bromide and
methyl iodide, the corresponding b-allylated ketone or b-methylated
ketone was not obtained.
[19] Kondo and Uchiyama have reported the directed ortho metalation
of aromatic compounds, such as ethyl benzoate, with di-tert-butyl-
the zinc atom of the zincate has stronger basicity than that on con-
ventional zinc reagents, we are expecting that a ligand of zincate
complex, such as an alkoxy group, has an ability as electron donor.
See: Y. Kondo, M. Shilai, M. Uchiyama, T. Sakamoto, J. Am. Chem.
Soc. 1999, 121, 3539.
TMP zincate (LiZnACHTNUTRGNEUNG(tmp)tBu2). The reaction directly affords o-zin-
cated aromatic compounds. This reactivity is in contrast to the low
basicity found for conventional zinc reagents. Because the ligand on
Received: April 21, 2009
Published online: November 11, 2009
708
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 703 – 708