Iron-catalyzed allylation of ketones

Article abstract of DOI:10.1016/j.tetlet.2006.06.134

Allylation of ketones has been efficiently performed using iron-complex catalysis and led to homoallylic alcohols in high yields.

Full text of DOI:10.1016/j.tetlet.2006.06.134

Tetrahedron Letters 47 (2006) 6255–6258
Iron-catalyzed allylation of ketones
*
´
Muriel Durandetti and Jacques Perichon
`
Laboratoire d’Electrochimie, Catalyse et Synthese Organique, UMR 7582 CNRS-Universite Paris 12 Val de Marne,
´
2 rue Henri-Dunant, 94320 Thiais, France
Received 12 May 2006; revised 22 June 2006; accepted 26 June 2006
Available online 14 July 2006
Abstract—Allylation of ketones has been efficiently performed using iron-complex catalysis and led to homoallylic alcohols in high
yields.
Ó 2006 Elsevier Ltd. All rights reserved.
The allylation of carbonyl compounds by means of
allylic metal reagents is a very important class of C–C
bond forming reactions, since it gives rise to syntheti-
cally useful homoallylic alcohols.1 These compounds
can be easily converted to various building blocks useful
for the synthesis of natural products.2 Different methods
of allylation have been developed mainly based on the
nucleophilic character of the allylmetal obtained from
allyl bromides and metallic species (metal = Li,3 Mg,4
Al,5 Zn,6 Ni,7 etc.). More recently allylic organometallic
compounds, such as allyl-chromium,8 -indium,9 -manga-
nese,10 -silane,11 -boronate,12 -stannane,13 etc., have
been examined. Allylation of carbonyl compounds from
allylic acetate requires the use of palladium as catalyst,
in addition to another metal or a reducing salt.14 In
most cases, only aldehydes are reactive. The allylation
of ketones seems more difficult, except in the presence
of SmI2.15 Allylstannane or allylsilane also permits the
allylation of ketones, and enantioselective approaches
are even known in this field.16 In our laboratory, we
have shown that a cobalt-catalyzed coupling reaction
of allyl acetates with carbonyl compounds using zinc
dust as reducing metal leads to homoallylic alcohols.17
Another access to allylic anions is their electrochemical
generation, for example using catalytic amount of tin
and allyl bromide.18 We already described some electro-
chemical processes for the synthesis of homoallylic
alcohols.19
With the intention of having a less toxic process, iron
species have been added to the relatively large list of cat-
alysts known to carry out allylation reactions.20 Ally-
lation of aldehydes with allyl trimethylsilane occurs
also with FeCl3.20b In the same time, we initiated a work
investigating iron catalysis21 because of the abundance
of the metal, its low toxicity, and its cheap price. We
have recently reported iron-catalyzed electrochemical
allylation reactions of carbonyl compounds.22 In this
reaction homoallyl alcohols are prepared from alde-
hydes or ketones and allylic acetates, using an electro-
chemical process catalyzed by iron complexes. The
reaction proved to be highly regioselective, affording
almost exclusively branched homoallyl alcohols. How-
ever, electrochemical reactions are often considered as
being less scalable than conventional classical methods.
Thus, electrochemical processes are infrequently applied
in industrial scale. We have then reported a chemical
method for an iron Reformatsky type reaction, using
FeBr2 as catalyst in the presence of an appropriate
reducing metal.23 We intended to show that this method
can be suitable for the activation of allyl acetates. We
present in this paper the investigation of iron-catalyzed
coupling reaction from allyl acetates with ketones
(Scheme 1), leading to homoallylic alcohols.
We first conducted a series of experiments with allyl
acetate and ketones under the reaction conditions
previously used for the Reformatsky reaction23
*
Corresponding author at present address: Laboratoire des Fonctions
´ ´
R
R'
R'
R
FeBr2, L
´
Azotees & Oxygenees Complexes de L’IRCOF, UMR CNRS 6014,
+
AcO
Mn, solvent
´
Universite et INSA de Rouen, 76821 Mont Saint-Aignan Cedex,
O
OH
France. Tel.: +33 (0)2 35 52 24 37; fax: +33 (0)2 35 52 29 71; e-mail:
muriel.durandetti@univ-rouen.fr
Scheme 1. The allylation reaction.
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.06.134

6256
M. Durandetti, J. Pe´richon / Tetrahedron Letters 47 (2006) 6255–6258
(i.e., FeBr2bipy (0.25 equiv) as catalyst in DMF or
FeBr2 (0.15 equiv) without ligand in acetonitrile). Man-
ganese metal was used as reducing metal of iron salts.
We observed at first that the reaction is best performed
at 80 °C, and that chemical yields are better using 25%
of FeBr2bipy in DMF, than FeBr2 in acetonitrile. How-
ever, even at this temperature, chemical yields of homo-
allylic alcohols are low: about 30–40%, whatever the
method used, the amount of iron salts or the ketones.
The reaction appears to stop, since carbonyl compounds
are recovered at the end (see the kinetics in Fig. 1). We
tried then to optimize the reaction conditions using
FeBr2bipy in DMF (Scheme 2).
iron salts by acetate ions generated during the process.
However, we found that the yields could not be in-
creased by the use of larger amounts of the iron catalyst
(even with 1 equiv), or with addition of 1 or 2 equiv of
Me3SiCl, even if the reaction occurs quickly in this case
(Table 1, entries 1 and 2). Moreover, when we ran the
reaction without carbonyl compounds, we obtained
the quantitative formation of hexadiene, thus confirm-
ing the formation of the p-allyl-iron complex. As allyl
acetate is totally consumed, nothing prevents the reac-
tion to go at the end. We then tried to add another salt
like LiCl or ZnBr2. Chemical yields increased only in the
presence of ZnBr2. As shown in Figure 1, best results are
obtained with 50 mol % of ZnBr2. We could suppose
that the p-allyl-iron, generated during the process,
reacted with ZnBr2 to lead a p-allyl-zinc species.
In all cases, the reactions seem to be blocked at one
stage (Table 1). Ketones and allyl acetates are no more
consumed. We supposed then a complexation step of
After optimization the best conditions were found to be
0.25 equiv of FeBr2bipy in the presence of 0.5 equiv of
ZnBr2, providing 50 to 80% of homoallylic alcohols in
5 h approximatively as shown in Table 2.25 Chemical
yields are moderate to good. Therefore, the use of iron
in the presence of zinc salts allows the coupling between
allyl acetate and ketones. This method is efficient with
aliphatic, cyclic, as well as aromatic ketones (Table 2).
No conjugated addition was observed with cyclohexe-
none, thus indicating that the reaction is regiospecific
(Table 2, entry 5).
n (mmol)
7
Without ZnBr2
25% of ZnBr2
6
35% of ZnBr2
5
50% of ZnBr2
75% of ZnBr2
4
3
2
1
We then tried to realize the reaction between cyclohexa-
none and allyl acetate under the same conditions but
without FeBr2bipy as catalyst. In this case, no coupling
product was obtained even after five days. The iron
catalysis is necessary to obtain the coupling product.
0
t (h)
0
2
4
6
8
10
Figure 1. Evolution of the cross-coupling product between aceto-
phenone and allylOAc as a function of ZnBr2.
Table 2. Allylation of ketones using FeBr2bipy as catalyst, in the
presence of ZnBr2
Entry
1
Ketones
Yields (%) of
R
R'
R'
R
coupling producta
FeBr2bipy
+
AcO
Mn, DMF, 80 °C
O
OH
79
O
Scheme 2. Allylation catalyzed by FeBr2bipy in DMF as solvent.
2
3
4
61
O
Table 1. Allylation of ketones using FeBr2bipy as catalyst
51
O
Entry
Ketones
Yields (%) of
coupling producta
O
41b
1
2
40b
43c
O
O
5
6
71
68
O
3
4
39
31
O
O
7
53
O
S
O
a Isolated yields, based on initial ketone.
b Reaction time: 30 h.
c Reaction conducted with 1 equiv of Me3SiCl; reaction time: 3 h.
a Isolated yields, based on initial ketone.
b 50% of ketone is recovered.

M. Durandetti, J. Pe´richon / Tetrahedron Letters 47 (2006) 6255–6258
6257
The chemistry of allylic compounds also includes the
additional regiochemical aspect. So we ran a series of
experiments with the same experimental procedure,
using crotyl acetate (or its allylic isomer), instead of allyl
acetate. Reactions involving allylic derivatives generally
give two isomeric products due to allylic transposition
(Scheme 3).
In the case of dissymetric carbonyl compounds (Table 3,
entries 5 and 7), we obtained the coupling product, as
two couples of diastereoisomers, with a very moderate
diastereoselectivity. The ratio anti/syn is 55/45.
Even if the reaction is regioselective in favor of the
branched product, the coupling product isomerizes into
the linear coupling product if the alkoxide is left to stir
after the end of the reaction (Table 3, entry 3). We have
access to both isomers, depending on the reaction time.
If the crude product is hydrolyzed immediately after the
end of the reaction (ca. 5 h), the branched product is ob-
tained as the main product. Instead, if the desired prod-
uct is the linear isomer, we just have to wait 24 h before
quenching the reaction. An equilibrium should exist
between the kinetic and the thermodynamic isomers,
probably initiated by a reversible reaction of allylation
(Scheme 4).
The ratio of branched to linear alcohol (B/L) is known
to depend mainly on the nature of the metal in the orga-
nometallic reagent.1a,24 In our hands, the major product
formed in these reactions is the branched one B, with an
approximate ratio B/L of 9/1 (Table 3). However, the
regioselectivity decreased with reaction time (Table 3,
entries 5 and 6) and isomerization was complete after
a day (Table 3, entry 3).
The yields are moderate to good and do not seem to
greatly depend on the nature of the butenyl acetate
(crotyl or its isomer) (Table 3, entries 1 and 2). So, crotyl
acetate and 1-methyl prop-2-enyl acetate are as efficient
in this coupling reaction: same yield, and same B/L
ratio.
We supposed that the reversibility is due to the presence
of zinc in the solution. If the reaction is run with crotyl
acetate without ZnBr2, homoallylic alcohols are
obtained in poor yields (ca. 30%) with a B/L ratio of
R1
R2
MII
+
R2
R1
R2
R1
O
B
L
OAc
FeBr2bipy 0.25 eq,
ZnBr2 0.5 eq,
HO
O
+
+
R
1 R2
R2
R2
DMF, Mn 2.5 eq, 80 °C
R1
HO
time
R1
AcO
MIIO
MIIO
B 5 hours
L 24 hours
Scheme 3. Regioselectivity of the iron-catalyzed addition of crotyl
acetate to ketones.
Scheme 4. Equilibrium between homoallylic alkoxides.
Table 3. Regioselectivity in the iron allylation reaction
Entry
Ketones
Allylic compounds
Yieldsa (%)
80
Ratio B/L
O
O
1
95/5
O
O
O
2
3
4
77
86
41
94/6
O
O
O
O
82/18b
90/10
O
O
O
O
O
5
51c
80/20d
O
O
O
6
7
69
75/25d
87/13
O
O
O
69c
O
a Isolated yields, based on initial ketone.
b Ratio B/L = 20/80 if the alkoxide is hydrolyzed after only 24 h.
c Product B: anti/syn: 55/45.
d Reaction time is longer than 5 h, so ratio B/L = 95/5 after 3 h and 80/20 after 6 h.

6258
M. Durandetti, J. Pe´richon / Tetrahedron Letters 47 (2006) 6255–6258
12. (a) Arnauld, T.; Barrett, A. G. M.; Seifried, R. Tetra-
hedron Lett. 2001, 42, 7899–7901; (b) Felpin, F.-X.;
Bertrand, M.-J.; Lebreton, J. Tetrahedron 2002, 58,
7381–7389.
13. (a) Kamble, R. M.; Singh, V. K. Tetrahedron Lett. 2001,
42, 7525–7526; (b) Nokami, J.; Otera, J.; Sudo, T.;
Okawara, R. Organometallics 1983, 2, 191–193.
14. (a) Masuyama, Y.; Kinugawa, N.; Kurusu, Y. J. Org.
Chem. 1987, 52, 3704–3706; (b) Araki, S.; Kamei, T.;
Hirashita, T.; Yamamura, H.; Kawai, M. Org. Lett. 2000,
2, 847–849.
95/5 even after 12 h. The reversibility of allylation using
allylic organozinc reagent has already been reported by
Miginiac et al.,6a,b who mentioned that homoallylic
alcohol are obtained initially with a ratio B/L of
50/50, and of 0/100 after 12 h.
In conclusion, we have reported in this paper an original
method of efficient addition of allylic acetates to
ketones, enabling the preparation of valuable target
molecules: homoallylic alcohols. The method is also very
easy, cheap, and low toxic, compared to the other meth-
ods described in the literature, where palladium com-
plexes, in the presence of another reducing salt, must
be used to activate allyl acetate. The reduction of FeII
by manganese metal probably leads to Fe0, which is sta-
bilized by 2,20-bipyridine. Reactions are regioselective:
the branched product B is the major product, but the
process permits the obtention of the linear isomer.
15. Tabuchi, T.; Inanaga, J.; Yamaguchi, M. Tetrahedron
Lett. 1986, 27, 1195.
16. (a) Casolari, S.; D’Addario, D.; Tagliavini, E. Org. Lett.
1999, 1, 1061; (b) Tietze, L. F.; Volkel, L.; Wulff, C.;
Weigand, B.; Bittner, C.; McGrath, P.; Johnson, K.;
Schafer, M. Chem. Eur. J. 2001, 7, 1304–1308; (c)
Pignataro, L.; Benaglia, M.; Annunziata, R.; Cinquiti,
M.; Cozzi, F. J. Org. Chem. 2006, 71, 1458; (d) Denmark,
S. E.; Fu, J.; Coe, D. M.; Su, X.; Pratt, N. E.; Griegel, B.
D. J. Org. Chem. 2006, 71, 1513; (e) Denmark, S. E.; Fu,
J.; Lawler, M. J. J. Org. Chem. 2006, 71, 1523.
´
17. Gomes, P.; Gosmini, C.; Perichon, J. Synthesis 2003,
References and notes
1909–1915.
18. Torii, S.; Uneyama, K.; Matsuda, H. Tetrahedron Lett.
1984, 25, 6017–6020.
1. Reviews: (a) Yamamoto, Y.; Asao, N. Chem. Rev. 1993,
93, 2207–2293; (b) Denmark, S. E.; Fu, J. Chem. Rev.
2003, 103, 2763–2793.
2. (a) Ramachandran, P. V.; Reddy, M. V. R.; Brown, H. B.
Tetrahedron Lett. 2000, 41, 583–586; (b) Reddy, M. V. R.;
Rearick, J. P.; Hoch, N.; Ramachandran, P. V. Org. Lett.
2001, 3, 19–20.
´
19. (a) Durandetti, S.; Sibille, S.; Perichon, J. J. Org. Chem.
`
1989, 54, 2198–2204; (b) Rollin, Y.; Derien, S.; Dunach,
´
E.; Gebehenne, C.; Perichon, J. Tetrahedron 1993, 49,
´
´
7723–7732; (c) Durandetti, M.; Nedelec, J.-Y.; Perichon, J.
Org. Lett. 2001, 3, 2073–2076.
20. (a) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Chem. Rev.
2004, 104, 6217; (b) Watahiki, T.; Oriyama, T. Tetra-
hedron Lett. 2002, 43, 8959–8962.
3. Alonso, E.; Guijarro, D.; Ramon, D. J.; Yus, M.
Tetrahedron 1999, 55, 11027–11038.
4. (a) Blomberg, C.; Hartog, F. A. Synthesis 1977, 18; (b)
Yamamoto, Y.; Maruyama, K. J. Org. Chem. 1983, 48,
1564–1565.
´
21. Durandetti, M.; Meignein, C.; Perichon, J. Org. Lett.
2003, 5, 317–320.
´
22. Durandetti, M.; Meignein, C.; Perichon, J. J. Org. Chem.
5. Courtois, G.; Miginiac, L. J. Organomet. Chem. 1974, 69,
1–44.
2003, 68, 3121–3124.
´
23. (a) Durandetti, M.; Perichon, J. Patent application, FR
6. (a) Miginiac, P.; Bouchoule, C. Bull. Chim. Soc. Fr. 1968,
4675–4676; (b) Barbot, F.; Miginiac, P. J. Organomet.
Chem. 1977, 132, 445–454; (c) Ranu, B. C.; Majee, A.;
Das, A. R. Tetrahedron Lett. 1995, 36, 4885–4888; (d)
Sklute, G.; Marek, I. J. Am. Chem. Soc. 2006, 128, 4642–
4649.
7. Hegedus, L. S.; Wagner, S. D.; Waterman, E. L.; Siirala-
Hansen, K. J. Org. Chem. 1975, 40, 593–598.
8. (a) Nowotny, S.; Tucker, C. E.; Jubert, C.; Knochel, P.
J. Org. Chem. 1995, 60, 2762–2772; (b) Baati, R.;
Gouverneur, V.; Mioskowski, C. J. Org. Chem. 2000, 65,
1235–1238.
9. (a) Khan, F. A.; Prabhudas, B. Tetrahedron 2000, 56,
7595–7599; (b) Lombardo, M.; Girotti, R.; Morganti, S.;
Trombini, C. Org. Lett. 2001, 3, 2981–2983; (c) Hirashita,
T.; Kambe, S.; Tsuji, H.; Omori, H.; Araki, S. J. Org.
Chem. 2004, 69, 5054–5059.
´
2004–7230; (b) Durandetti, M.; Perichon, J. Synthesis
2006, 1542–1548.
24. Ito, A.; Kishida, M.; Kurusu, Y.; Masuyama, Y. J. Org.
Chem. 2000, 65, 494–498.
25. General procedure for the cross-coupling of allylic
acetates with carbonyl compounds: 10 mmol of carbonyl
compounds and 13 mmol of allylic acetates were added
in a stirred flask under argon at 80 °C with 30 mL of
DMF. Then 1.35 g of Mn (25 mmol) is introduced,
followed by 1.1 g of ZnBr2 (5 mmol), 0.39 g of 2,20-
bipyridine (2.5 mmol), and 0.54 g of FeBr2 (2.5 mmol)
and finally 20 lL of CF3CO2H to activate manganese
metal. The reaction is conducted at 80 °C and was
monitored by GC-analysis and quenched after the
carbonyl compounds were consumed (ca. 5 h for the
branched product or 24 h for the linear product). The
mixture was then hydrolyzed with 1 N HCl and diluted
with diethyl ether. The aqueous layer was extracted with
diethyl ether, the combined organic layers were washed
with water and saturated NaCl solution, dried over
MgSO4, and the solvent was evaporated. The oil thus
obtained was purified by column chromatography to give
the desired compounds.
10. (a) Cahiez, G.; Chavant, P.-Y. Tetrahedron Lett. 1989, 30,
7373; (b) Furstner, A.; Brunner, H. Tetrahedron Lett.
¨
1996, 37, 7009–7012; (c) Kakiya, H.; Nishimae, S.;
Shinokubo, H.; Oshima, K. Tetrahedron 2001, 57, 8807.
11. (a) Davis, A. P.; Jaspars, M. J. Chem. Soc. Perkin Trans.
1992, 2111–2118; (b) Yadav, J. S.; Chand, P. K.; Anja-
neyulu, S. Tetrahedron Lett. 2002, 43, 3783–3784.