The stereochemistry of 1,2-additions of allylmagnesium, allylindium, and allylbismuth to cyclohexenones

Article abstract of DOI:10.1016/j.tetlet.2006.03.034

Allylmagnesium, allylindium, and allylbismuth generally showed a preference for axial addition to cyclohexenones. Allylmagnesium was the most stereoselective. Reactions with an α-methylated enone (carvone) were the most selective, except that allylbismuth was unreactive with this substrate.

Full text of DOI:10.1016/j.tetlet.2006.03.034

Tetrahedron Letters 47 (2006) 3291–3294
The stereochemistry of 1,2-additions of allylmagnesium,
allylindium, and allylbismuth to cyclohexenones
Liang Zhao and D. Jean Burnell*
Department of Chemistry, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4J3
Received 25 January 2006; revised 1 March 2006; accepted 6 March 2006
Available online 29 March 2006
Abstract—Allylmagnesium, allylindium, and allylbismuth generally showed a preference for axial addition to cyclohexenones. Allyl-
magnesium was the most stereoselective. Reactions with an a-methylated enone (carvone) were the most selective, except that allyl-
bismuth was unreactive with this substrate.
Ó 2006 Elsevier Ltd. All rights reserved.
The transition structure for the axial 1,2-addition of a
metalated nucleophile to a cyclohexenone is inherently
lower in energy than the transition structure for the
equatorial addition. Houk and co-workers1 ascribed this
difference to both torsional strain in the equatorial tran-
sition structure and maintenance of a nearly coplanar
geometry of the enone moiety in the axial transition
structure. However, the nature of the metal has been
shown to be important with simple cyclohexenones2
and with a-heterosubstituted cyclohexenones.3,4 The
former might be explained by differences in the geo-
metries about the metals;1 the latter might be complicated
by electrostatic, chelation3 and additional steric consi-
derations. Also, significant differences in the mechanism
have been discovered, depending on a number of para-
meters including the metal.5
indium additions was reported to be better in THF than
in water, although reaction times in THF were very
much longer than in water.9
The reaction of allylmagnesium bromide with 4-tert-
butylcyclohex-2-enone in ether at room temperature
gave a 5:1 mixture of the epimers 1 and 2,10 respectively,
in a combined yield of 83%. The Grignard reaction in
THF gave the same ratio of epimers.
O
OH
+
OH
1
2
OH
+
In connection with ongoing synthetic efforts, we wished
to assess the stereoselectivity of additions of allylindium
to cyclohexenones. This is because allylindium species
were reported to add to a conformationally restricted
ketone, such as 4-tert-butylcyclohexanone, to give epi-
meric alcohols in a ratio of better than 1:4, favoring
the equatorial-addition product, whereas allylmagne-
sium chloride added to the same, saturated ketone to
give the epimeric alcohols in a 1.2:1 ratio, very slightly
favoring the axial-addition product.6,7 In addition, allyl-
indium is tolerant of alcohol functions in the substrate.
Indeed, allylindium reactions can be conducted in an
aqueous medium,8,9 but the diastereoselectivity of allyl-
OH
3
4
The stereochemistry at the carbinol center was initially
assigned based on the relative chemical shifts of the
carbinol centers in the 13C NMR spectra. The carbinol
signal further downfield should belong to the axial-addi-
tion product.2,4,11 Attempts to corroborate this assign-
ment by measurement of NOE enhancements led to
inconclusive results, so each epimer was hydrogenated
catalytically. The 13C NMR data and melting point of
the saturated tertiary alcohol derived from the minor
epimer matched the data for 4 in the literature.10,12
Thus, the saturated tertiary alcohol derived from the
major epimer was 3.
*
Corresponding author. Tel.: +1 (902) 494 1664; fax: +1 (902) 494
1310; e-mail: Jean.Burnell@dal.ca
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.03.034

3292
L. Zhao, D. J. Burnell / Tetrahedron Letters 47 (2006) 3291–3294
An allylindium reagent was prepared in DMF,13 and
this was reacted with 4-tert-butylcyclohex-2-enone at
room temperature for 6 h. A 1:1 mixture of 1 and 2
was obtained in a combined yield of 85%. No change
in the ratio of the epimers was seen when allylindium
reagents were prepared in different solvents, although the
reaction rate was affected: DMF/H2O 1:1 (1 day), H2O
(2 days), and dry THF (incomplete after 7 days).
Paquette and Lobben9 had seen the same trend in the
rates. That the ratio of epimers did not change was
unexpected, as it is known that the nature of the indium
reagent is not the same in aqueous7 and non-aqueous
solutions,6,14 and some changes in stereoselectivity had
been observed in different solvents with 4-tert-
butylcyclohexanone.7
O
OH
or
OH
7
8
HO
OH
9
H
NOE
H
H
O
Allylbismuth15,16 is a nucleophile that has received only
scant attention. The stereoselectivity of its additions to
enones was unknown. Allylbismuth reacted with 4-
tert-butylcyclohex-2-enone in THF under reflux to pro-
vide a mixture of 1 and 2 in a ratio of 2.2:1, respectively.
H
H
H
NOE
10
Cholest-4-en-3-one presents facial alternatives that are
very similar to those of 4-tert-butylcyclohex-2-enone.
Allylmagnesium bromide reacted with this enone to give
a 6.3:1 mixture of the epimeric alcohols 5 and 6.17 The
relative positions of the carbinol signals in their 13C
NMR spectra indicated that the major isomer was 5,
the product of axial addition. The allylindium reagent
in DMF provided a 4.5:1 ratio of 5 and 6, respectively.
The allylbismuth reagent in THF under reflux gave 5
and 6 in a ratio of 4.2:1, respectively.
product in CH2Cl2 was stirred with a catalytic amount
of Grubbs’ ‘second generation’ ruthenium catalyst.19
The idea was that the relative stereochemistry of 8, but
not 7, might make the molecule amenable to ring-closing
metathesis. However, the only product (93%) was the
result of intermolecular, cross-metathesis (9).20 A survey
of derivatives of the addition product led to the (b-naph-
thyl)methyl ether (10),20 which had a fortuitously dis-
1
persed H NMR spectrum that allowed unambiguous
assignment of the relative stereochemistry by measure-
ment of the NOE enhancements shown above. There-
fore, the structure of the addition product was 7,20 the
result of axial addition to the carbonyl. This result was
consistent with a significant attenuation of the rate of
equatorial attack by the sterically hindering methyl
group. The allylindium reagent in DMF would not react
with the even more hindered enone 11, to which allyl-
magnesium added with reluctance.21
C8H17
O
C8H17
HO
5
MeO
C8H17
+
O
OH
H
6
TBSO
H
OH
Carvone is a cyclohex-2-enone with a significant confor-
mational bias but with additional steric bulk in the plane
of the carbonyl by virtue of the methyl group. This
feature was expected to hinder equatorial addition more
than axial addition. Addition of allylmagnesium
bromide to (R)-carvone in ether produced only one
product. In DMF, the allylindium reagent reacted with
(R)-carvone to give the same product18 exclusively. On
the other hand, the allylbismuth reagent in THF showed
no sign of reaction with (R)-carvone, even after
many days under reflux. The stereochemistry of the
single product could not be assigned by comparison of
11
At the moment, any interpretation of these results can
only be simplistic as the structures of the allylindium
reagent (in DMF) and of allylbismuth are not well
understood, and the mechanisms by which all three
organometallics react may not be similar. Cyclohexa-
none has an inherent tendency to be attacked by nucleo-
philes axially, but other considerations, including steric
effects, may disfavor axial attack. In the reaction of
cyclohexanone with allylmagnesium, any axial-addition
tendency must be almost exactly balanced by other
forces because the axial:equatorial ratio was reported
to be almost 1:1.6,7 The inherent tendency for axial
1
13C NMR shifts, and overlap in the H NMR spectrum
led to ambiguous NOE data. In an initial attempt to
determine the structure chemically, a solution of the

L. Zhao, D. J. Burnell / Tetrahedron Letters 47 (2006) 3291–3294
3293
For 4: mp 71–73 °C, lit.12 73–74 °C; IR (cast): 3380 cmÀ1
;
addition with cyclohexanone must be overcome by the
other considerations with the allylindium reagent,
because the equatorial-addition product was reported
to predominate.7 The inherent tendency for axial attack
is greater for cyclohexenone than for cyclohexanone.1 In
our hands, in the reactions with the cyclohexenones,
allylmagnesium gave highest proportion of the axial-
addition products, whereas in additions of the allyl-
indium reagent and allylbismuth, the proportions of
axial-addition products were less. Indeed, with the
simplest case, 4-tert-butylcyclohex-2-enone, the allyl-
indium reagent returned a 1:1 mixture of the axial-
and equatorial-addition products.
1H NMR: d 1.66 (2H, m), 1.57 (2H, m), 1.39 (4H, m),
1.26–1.35 (5H, m), 0.91 (3H, m), 0.86 (9H, s); 13C NMR: d
70.9 (0), 48.2 (1), 46.9 (2), 37.6 (2), 32.6 (0), 27.8 (3), 22.7
(2), 16.6 (2), 15.0 (3); lit.12 (CDCl3, 80 MHz instrument): d
70.7, 48.1, 46.8, 37.5, 32.4, 27.6, 22.6, 16.4, 14.8.
11. (a) Buckwalter, B. L.; Burfitt, I. R.; Felkin, H.; Joly-
Goudket, M.; Maemura, K.; Salomon, M. F.; Wenkert,
E.; Wovkulich, P. M. J. Am. Chem. Soc. 1978, 100, 6445–
6450; (b) Trost, B. M.; Gunzner, J. L.; Dirat, O.; Rhee,
Y. H. J. Am. Chem. Soc. 2002, 124, 10396–10415.
12. Welch, S. C.; Prakasa Rao, A. S. C.; Lyon, J. T.; Assercq,
J.-M. J. Am. Chem. Soc. 1983, 105, 252–257.
13. Allylindium procedure: Indium metal (0.50 mmol) and allyl
iodide (0.75 mmol) in DMF (2.0 mL) were stirred under
argon for 1 h during which time most of the indium
dissolved giving a grey solution. The enone (0.50 mmol) in
DMF (1.0 mL) was introduced dropwise, and the solution
was stirred at room temperature. Reaction progress was
monitored by TLC. Aqueous workup provided the pro-
duct(s). The ratio of products was determined by NMR.
14. Gynane, M. J. S.; Worrall, I. J. J. J. Organomet. Chem.
1974, 81, 329–334.
Acknowledgements
Support from the Natural Sciences and Engineering
Research Council of Canada and the Killam Trust is
gratefully acknowledged.
15. (a) Makoto, W.; Hidenori, O.; Kinya, A. Tetrahedron
Lett. 1986, 27, 4771–4774; (b) Makoto, W.; Munekazu,
H.; Yoshihiro, K.; Norikazu, M. Bull. Chem. Soc. Jpn.
1997, 70, 2265–2267; (c) Xu, X.; Zha, Z.; Wang, Z. Synlett
2004, 1171–1174.
References and notes
1. Wu, Y.-D.; Houk, K. N.; Florez, J.; Trost, B. M. J. Org.
Chem. 1991, 56, 3656–3664.
16. Allylbismuth procedure:15 The enone (0.6 mmol), allyl
bromide (1.3 mmol), BiCl3 (0.85 mmol), and zinc powder
(1.3 mmol) in THF (10 mL) were heated to reflux under
argon. Reaction progress was monitored by TLC. A solid
was removed by filtration, and the THF was evaporated
under vacuum. Ether was added, and this solution was
washed with 0.5 M HCl, H2O, and saturated NaHCO3.
The solution was dried (MgSO4), and the ether was
evaporated under vacuum. The ratio of products was
determined by NMR.
2. (a) Trost, B. M.; Florez, J.; Jebaratnam, D. J. J. Am.
Chem. Soc. 1987, 109, 613–615; (b) Trost, B. M.; Florez,
J.; Haller, K. J. J. Org. Chem. 1988, 53, 2396–2398.
3. Klemeyer, H. J.; Paquette, L. A. J Org. Chem. 1994, 59,
7924–7927.
4. Lindsay, H. A.; Salisbury, C. L.; Cordes, W.; McIntosh,
M. C. Org. Lett. 2001, 3, 4007–4010.
5. Gajewski, J. J.; Bocian, W.; Harris, N. J.; Olson, L. P.;
Gajewski, J. P. J. Am. Chem. Soc. 1999, 121, 326–334.
6. Araki, S.; Ito, H.; Butsugan, Y. J. Org. Chem. 1988, 53,
1833–1835.
17. For 5: IR (cast): 3348 cmÀ1 1H NMR: d 5.89 (1H, m),
;
7. Reetz, M. T.; Haning, H. J. Organomet. Chem. 1997, 541,
117–120.
8. (a) Chan, T. H.; Yang, Y. J. Am. Chem. Soc. 1999, 21,
3228–3229; (b) Podlech, J.; Maier, T. C. Synthesis 2003,
633–655; (c) Lucas, P.; Gajewski, J. J.; Chan, T. H. Can. J.
Anal. Sci. Spect. 2003, 48, 1–6.
5.10–5.19 (3H, m), 2.33 (1H, dd, J = 13.7, 6.9 Hz), 2.25
(1H, dd, J = 13.7, 8.0 Hz), 2.20 (1H, dt, J = 5.5, 14.0 Hz),
1.99 (2H, m), 1.70–1.89 (3H, m), 1.04 (3H, s), 0.90 (3H, d,
J = 7.1 Hz), 0.865 (3H, d, J = 6.5 Hz), 0.860 (3H, d,
J = 6.5 Hz), 0.68 (3H, s), and other signals unresolved
0.70–1.68; 13C NMR: d 147.1 (0), 134.2 (1), 125.6 (1), 118.7
(2), 71.1 (0), 56.5 (1), 56.4 (1), 54.4 (1), 45.7 (2), 42.7 (0),
40.1 (2), 39.7 (2), 37.7 (0), 36.4 (2), 36.2 (1), 36.0 (1), 34.9
(2), 33.5 (2), 32.6 (2), 32.5 (2), 28.4 (2), 28.2 (1), 24.4 (2),
24.1 (2), 23.0 (3), 22.8 (3), 21.4 (2), 19.2 (3), 18.9 (3), 12.2 (3).
For 6: IR (cast): 3422 cmÀ1; 1H NMR: d 5.85 (1H, m), 5.24
(1H, s), 5.09–5.14 (2H, m), 2.21–2.30 (2H, m), 2.20 (1H, m),
1.97–2.03 (2H, m), 1.80 (1H, m), 1.72 (1H, m), 0.94 (3H, s),
0.90 (3H, d, J = 6.8 Hz), 0.87 (3H, d, J = 6.4 Hz), 0.86 (3H,
d, J = 6.1 Hz), 0.66 (3H, s), and other signals unresolved
0.77–1.68; 13C NMR: d 149.6 (0), 134.3 (1), 124.1 (1),
118.4 (2), 69.3 (0), 56.4 (1), 56.3 (1), 54.5 (1), 47.6 (2), 42.7
(0), 40.1 (2), 39.7 (2), 37.8 (0), 36.4 (2), 36.1 (1), 36.0 (1), 33.5
(2), 32.9 (2), 32.7 (2), 32.2 (2), 28.4 (2), 28.2 (1), 24.5 (2), 24.1
(2), 23.0 (3), 22.8 (3), 21.7 (2), 18.9 (3), 18.1 (3), 12.2 (3).
18. 1,2-Addition of an allylindium reagent to carvone was
reported previously: Kim, H. Y.; Choi, K. I.; Pae, A. N.;
Koh, H. Y.; Choi, J. H.; Cho, Y. S. Synth. Commun. 2003,
33, 1899–1904, but the stereochemistry of the product was
not determined.
9. Paquette, L. A.; Lobben, P. C. J. Org. Chem. 1998, 63,
5604–5616.
10. All NMR spectra were run in CDCl3 on a 500 MHz
instrument.
For 1: IR (neat): 3375 cmÀ1 1H NMR: d 5.90 (1H, m),
;
5.74 (1H, dt, J = 10.2, 1.8 Hz), 5.61 (1H, dt, J = 10.2,
2.2 Hz), 5.10–5.17 (2H, m), 2.34 (1H, dd, J = 7.2,
13.6 Hz), 2.26 (1H, dd, J = 7.5, 13.6 Hz), 1.95 (1H,
complex d, J = 13 Hz), 1.90 (1H, m), 1.76 (1H, m), 1.56
(1H, m), 1.38 (1H, m), 0.88 (9H, s); 13C NMR: d 134.4 (1),
133.9 (1), 130.4 (1), 118.9 (2), 70.8 (0), 46.0 (2), 45.6 (1),
36.2 (2), 33.0 (0), 27.4 (3), 22.4 (2).
1
For 2: IR (neat): 3416 cmÀ1; H NMR: d 5.83–5.89 (2H,
m), 5.68 (1H, dt, J = 10.5, 2.1 Hz), 5.09–5.14 (2H, m), 2.28
(2H, symmetrical m), 1.76–1.80 (2H, m), 1.69 (1H, m),
1.45–1.56 (2H, m), 0.90 (9H, s); 13C NMR: d 134.0 (1),
133.2 (1), 132.5 (1), 118.5 (2), 68.7 (0), 47.3 (2), 46.7 (1),
35.7 (2), 32.7 (0), 27.5 (3), 20.2 (2).
For 3: mp 77–78 °C; IR (cast): 3293 cmÀ1 1H NMR: d
;
1.80 (2H, br d, J = 12.5 Hz), 1.67 (2H, m), 1.47 (2H, m),
1.32–1.39 (4H, m), 1.01–1.11 (3H, m), 0.95 (3H, t,
J = 7.2 Hz), 0.86 (9H, s); 13C NMR: d 72.5 (0), 47.8 (1),
39.1 (2), 39.0 (2), 32.5 (0), 27.9 (3), 24.7 (2), 16.1 (2), 15.0
(3).
19. Scholl, S.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett.
1999, 1, 953–956.
1
20. For 7: IR (neat): 3403, 1644 cmÀ1; H NMR: d 5.87 (1H,
m), 5.47 (1H, narrow m), 5.16 (1H, br d, J = 11.0 Hz),
5.13 (1H, br d, J = 18.5 Hz).4.73 (2H, br s), 2.47 (1H, dd,

3294
L. Zhao, D. J. Burnell / Tetrahedron Letters 47 (2006) 3291–3294
J = 13.8, 7.2 Hz), 2.32–2.37 (2H, m), 2.05–2.14 (2H, m),
For 10: IR (neat): 1640, 1602 cmÀ1; 1H NMR: d 7.81 (4H,
m), 7.43 (3H, m), 6.03 (1H, m), 5.74 (1H, narrow m), 5.11
(1H, d, J = 9.5 Hz), 5.10 (1H, d, J = 17.5 Hz), 4.72 (1H,
s), 4.71 (1H, s), 4.61 (1H, d, J = 12.5 Hz), 4.37 (1H, d,
J = 12.5 Hz), 2.67 (1H, ddd, J = 14.3, 6.1, 1.2 Hz), 2.44
(1H, dd, J = 14.3, 8.2 Hz), 2.38 (1H, m), 2.10 (1H, br d,
J = 10 Hz), 1.95–2.08 (2H, m), 1.82 (1H, t, J = 13.0 Hz),
1.72 (6H, s); 13C NMR: d 149.3, 137.7, 136.6, 135.2, 133.7,
132.9, 128.1, 128.0, 127.9, 127.8, 126.1, 125.7, 125.5, 125.4,
117.4 (2), 109.0 (2), 80.5 (0), 64.1 (2), 43.9 (2), 39.7 (1), 34.7
(2), 31.2 (2), 21.1 (3), 17.6 (3).
1.96 (1H, m), 1.74 (3H, br s), 1.73 (3H, s), 1.50 (1H, t,
J = 12.2 Hz); 13C NMR: d 149.2 (0), 138.3 (0), 133.9 (1),
124.1 (1), 118.8 (2), 109.3 (2), 74.0 (0), 43.1 (2), 40.7 (2),
39.4 (1), 31.0 (2), 21.0 (3), 17.2 (3).
For 9: IR (neat): 3415 cmÀ1; 1H NMR: d 5.63 (2H, narrow
m), 5.30 (2H, narrow m), 4.80 (2H, s), 4.77 (2H, s), 2.46
(2H, dd, J = 14.3, 4.0 Hz), 2.21–2.32 (4H, m), 2.07 (2H, br
d, J = 12.5 Hz), 1.97 (2H, m), 1.87 (2H, m), 1.77 (6H, s),
1.64 (6H, s), 1.40 (2H, t, J = 12.5 Hz); 13C NMR: d 149.0
(0), 138.2 (0), 129.3 (1), 123.8 (1), 109.2 (2), 73.7 (0), 41.7
(2), 40.4 (2), 39.3 (1), 30.9 (2), 20.7 (3), 17.1 (3).
21. Zhao, L.; Burnell, D. J. Org. Lett. 2006, 8, 155–157.