A novel gallium/samarium-mediated coupling of epoxides with allyl halides in aqueous media and synthesis of homoallylic alcohols

Article abstract of DOI:10.1246/cl.2005.90

A mild and efficient protocol for the coupling of various substituted epoxides 1 with allylgallium reagent generated in situ in a Bu ₄NBr-DMF-H₂O system have been developed. The coupling is also found to be equally effective when samarium is used in lieu of gallium and the corresponding homoallylic alcohols 2 were obtained in almost comparable yields. Copyright

Full text of DOI:10.1246/cl.2005.90

90
Chemistry Letters Vol.34, No.1 (2005)
A Novel Gallium/Samarium-mediated Coupling of Epoxides with Allyl Halides
in Aqueous Media and Synthesis of Homoallylic Alcohols
Mukut Gohain and Dipak Prajapatiꢀ
Medicinal Chemistry Division, Regional Research Laboratory, Jorhat-785006, Assam, India
(Received October 12, 2004; CL-041203)
A mild and efficient protocol for the coupling of various substi-
O
Ga or Sm/Bu4NBr
R
tuted epoxides 1 with allylgallium reagent generated in situ in a
Bu4NBr–DMF–H2O system have been developed. The coupling is
also found to be equally effective when samarium is used in lieu
of gallium and the corresponding homoallylic alcohols 2 were ob-
tained in almost comparable yields.
R
OH
X
2
1
DMF-H2O, rt
Scheme 1.
only a few examples of synthetic reactions in the literature12 using
gallium, which belongs to the same group as the extensively
studied boron, aluminium, and indium13 metals. Previously,
indium has been used for the allylation of epoxides in aprotic sol-
vents.14 After comparing the first ionization and reduction poten-
tials of gallium with those of indium (Ga: FIP, 5.99 eV, Eo,
Gaþ3/Ga = ꢁ0:56 V; In: FIP, 5.79 eV, Eo, Inþ3/In = ꢁ0:345
V), Wang et al.15 expected similar properties between the two met-
als. As in the case of indium, the reduction potential of gallium is
not too negative, and thus it is not sensitive to water and does not
form oxides readily in air.
Metal mediated reactions in aqueous media have recently
found considerable application in organic synthesis as they offer
significant advantages over conventional reactions using dry or-
ganic solvents.1 The use of water as solvent can reduce or eliminate
environmental damage by organic wastes.2 Therefore, organic re-
actions in aqueous media have attracted great attention. Also, nu-
cleophilic ring opening of epoxides with organometallic reagents
has found widespread application in the formation of carbon–car-
bon ꢀ-bonds.3 Among the most common reagents employed are
organomagnesium, organocopper, organozinc, organoaluminium
and organoboron compounds.4 In some cases the basicity of these
reagents can promote undesired side reactions. A useful alterna-
tive, which often avoids these problems, is the Lewis acid-promot-
ed addition of less basic carbon nucleophiles.4b Many reagents act-
ing as Lewis acids have been used for the cleavage of epoxides and
their conversion into 1,2-diamino, 1,2-halo-, 1,2-alkoxy-, and 1,2-
diazido- alcohols,5 but only a few methods are reported for the al-
lylation of epoxides with allylmetal reagents.6 In a report7 it is stat-
ed that allylative ring opening of ethylene oxide can be performed
smoothly with allylsilanes under the influence of TiCl4. Introduc-
tion of methyl substituent, however, results in the formation of
mixtures of products.8 In another report, the successful allylation
of alkenyloxiranes with allyltin reagents in the presence of
Treatment of styrene oxide with allylgallium reagent in the
presence of 2 mol % tetrabutylammonium bromide (TBAB) af-
forded the corresponding 1-phenyl-4-penten-2-ol in 85% yield.
Analysis of the crude mixture did not indicate the formation of
any other regioisomer. Similarly various other substituted epoxides
were reacted smoothly with allylgallium reagent in aqueous media
to give the corresponding homoallyl alcohols in 75–85% yields.
Even the sterically hindered cyclic epoxides showed good yields
of the target products. The coupling also proceeds effectively when
allylsamarium reagent was used and corresponding homoallylic al-
cohols were obtained in almost comparable yields (Table 1). The
reactions are generally clean and no trace of side product could
be detected in the NMR spectrum of the crude product. The struc-
ture of the homoallylic alcohols thus obtained were unambiguous-
ly identified on the basis of their spectral data (IR, NMR, MS) and
finally by comparing with authentic samples.10,16 The electron-do-
nating or withdrawing groups at the aromatic ring did not seem to
effect the reaction significantly either in terms of yields or the rate
of the reaction. Similarly allylgallium or allylsamarium reagent re-
acted smoothly with epichlorohydrin (Entry 11) to give the corre-
sponding 1-chloro-5-hexen-2-ol in 75 and 70%, yields, respective-
ly without the formation of 1,3-dichloro-2-propanol. The results,
as summarized in Table 1, reveal the generality of this methodol-
ogy in terms of structural variations of the epoxides and regioselec-
tivity; in each case homoallylic alcohols were isolated in high
yields. Moreover, a nitro function was not reduced under the reac-
tion condition. Thus 3-nitro styrene oxide was successfully allylat-
ed. Usually a nitro group is sensitive to reduction by metals and is
not tolerated under Barbier conditions.17 In this sense, the use of
TBAB as an activating agent is superior to the use of Al, Fe, or
NaBH4 reported previously.18 Also, allyl iodide was found to be
as reactive as allyl bromide, but the reactivity of allyl chloride
was found to be much less. The use of Bu4NBr was found to be
important, as the allylation did not proceed at all with gallium or
samarium alone. It is obvious that activation of the metal is needed
for this reaction to proceed. We tried other metal salts like KBr and
MgBr2 in aqueous media as additives in place of TBAB and found
BF3 OEt2 is described.9 The allylative ring opening takes place
.
at the site of the alkenyl substitution, which stabilize the incipient
positive charge, so that it prevents the isomerization of oxiranes.
Therefore, in view of the great utility of epoxides as synthetic in-
termediates in organic chemistry, it is worthwhile to explore the
versatility of gallium metal on the epoxide-opening and its subse-
quent allylation for the synthesis of homoallylic alcohols. In con-
tinuation of our ongoing programme on metal mediated organic
transformations,10 we report herein the first example of gallium
mediated ring-opening allylation of epoxides and synthesis of ho-
moallylic alcohols under aqueous conditions using tetrabutylam-
monium bromide (TBAB) as an additive. In a very recent report11
Lu and Li et al. have developed a InCl mediated coupling of epox-
ides with allylbromide catalyzed by mesoporous silica supported
palladium (expensive nano Pd-26, nano Pd-6, PdCl2(PPh3)2,
Pd(OAc)2, [Pd(allyl)Cl]2, Pd(PPh3)4) to give homoallyl alcohols
in 47–76% yields. In contrast we have investigated that various ep-
oxides undergo allylation when reacted with stoichiometric-quan-
tity of the allylgallium reagent generated in situ in a Bu4NBr–
DMF–H2O system at room temperature to produce homoallylic
alcohols 2 in high yields.
We paid particular attention to gallium, a soft low valent and
comparatively less studied element of group IIIA; there have been
Copyright Ó 2005 The Chemical Society of Japan

Chemistry Letters Vol.34, No.1 (2005)
91
4
5
a) J. M. Klunder and G. H. Posner, in ‘‘Comprehensive Organic Synthesis,’’ ed. by
B. M. Trost, I. Fleming, and G. Pattenden, Pergmon, Oxford (1991), Vol. 3, Chap.
1.5, pp 223–226. b) M. Yamaguchi, in ‘‘Comprehensive Organic Synthesis,’’ ed.
by B. M. Trost, I. Fleming, and S. L. Schreiber, Pergmon, Oxford (1991), Vol. 1,
Chap. 3.11, pp 342–345. c) J. Tang, H. Yorimitsu, H. Kakiya, R. Inoue, H.
Shinokubo, and K. Oshima, Tetrahedron Lett., 38, 9019 (1997).
a) C. Einhorn and J. L. Luche, Chem. Commun., 1986, 1368. b) J. Otera, Y.
Niibo, N. Tatsumi, and H. Nozaki, J. Org. Chem., 53, 275 (1988). c) J. Otera,
Y. Yoshinaga, K. Hirakawa, and T. Nakata, Tetrahedron Lett., 26, 3219 (1985).
d) Y. Niibo, T. Nakata, J. Otera, and H. Nozaki, Synlett, 1991, 97. e) C. Moberg,
L. Rakos, and L. Tottie, Tetrahedron Lett., 33, 2191 (1992). f) S. Saito, S.
Yamashita, T. Nishikawa, Y. Yokoyama, M. Inaba, and T. Moriwake,
Tetrahedron Lett., 30, 4153 (1989). g) R.-H. Fan, X.-L. Hou, and L.-X. Dai,
J. Org. Chem., 69, 689 (2004).
Table 1. Allylation of epoxides in DMF-H2O (3:1) by Ga/Sm-TBAB system
Reaction Yields Reaction
Yields
Entry
Epoxides 1
Allyl halides
Productsa 2
time/h,Ga /%,Ga time/h, Sm /%, Sm
Br
1
2
O
3.5
3.5
6.0
4.0
4.0
85
83
75
86
87
4.0
4.0
7.0
4.5
4.5
80
80
70
82
80
2a
OH
OH
OH
OH
OH
I
O
2a
3
Cl
2a
O
O
O
Me
Br
I
4
Me
2b
2c
Cl
Cl
5
Cl
Cl
Br
4.5
4.5
81
80
5.0
5.0
80
76
6
O
2c
O
OH
O
O
6
a) D. F. Taber, J. H. Green, and J. M. Geremia, J. Org. Chem., 62, 9342 (1997).
b) L. E. Overman and P. A. Renhowe, J. Org. Chem., 59, 4138 (1994). c) T. Imai
and S. Nishida, J. Org. Chem., 55, 4849 (1990).
Br
Br
O
O
7
OH
OH
2d
O
2e
Me
4.0
82
5.0
76
Me
8
7
8
9
I. Fleming and I. Patterson, Synthesis, 1979, 446.
G. A. Molander and S. W. Andrews, J. Org. Chem., 54, 3114 (1989).
Y. Naruta and K. Maruyama, Chem. Lett., 1987, 963.
Br
9
4.5
4.5
3.5
4.0
80
78
75
76
4.0
4.5
4.5
4.5
75
75
70
72
O
2f
OH
O2N
O2N
O
10
11
12
Br
Br
Br
10 a) B. Baruah, A. Boruah, D. Prajapati, and J. S. Sandhu, Tetrahedron Lett., 38,
1449 (1997). b) M. Gohain, D. Prajapati, and J. S. Sandhu, Synlett, 2004, 235.
c) D. Prajapati, D. D. Laskar, and J. S. Sandhu, Tetrahedron Lett., 41, 8639 (2000).
11 N. Jiang, Q. Hu, C. S. Reid, Y. Lu, and C.-J. Li, Chem. Commun., 2003, 2318.
12 K. Saigo, Y. Hashimoto, N. Kihara, H. Umehara, and M. Hasegawa, Chem. Lett.,
1990, 831; K. Saigo, Y. Hashimoto, N. Kihara, K. Hara, and M. Hasegawa, Chem.
Lett., 1990, 1097; M. Falorni, L. Lardicci, and G. Giacomelli, Tetrahedron Lett.,
26, 4949 (1985); S. Araki, H. Ito, and Y. Batsugan, Appl. Organomet. Chem., 26,
475 (1988); S. Murahashi, H. Mitsui, T. Shiota, T. Tsuda, and S. Watanabe,
J. Org. Chem., 55, 1736 (1990); M. Shibasaki, H. Sasai, and T. Arai, Angew.
Chem., Int. Ed., 36, 1236 (1997).
13 For very recent reviews on indium metal see: K. K. Chouhan and C. G. Frost,
J. Chem. Soc., Perkin Trans. 1, 2000, 3015; C.-J. Li and T.-H. Chan,
Tetrahrdron, 55, 11149 (1999).
14 B. K. Oh, J. H. Cha, Y. S. Cho, K. I. Choi, H. Y. Koh, M. H. Chang, and
A. N. Pae, Tetrahedron Lett., 44, 2911 (2003).
15 Z. Wang, S. Yuan, and C.-J. Li, Tetrahedron Lett., 43, 5097 (2002).
16 J. S. Yadav, B. V. S. Reddy, and G. Satheesh, Tetrahedron Lett., 44, 6501 (2003).
17 N. Kalyanam and G. V. Rao, Tetrahedron Lett., 34, 1647 (1993).
18 a) M. Lombardo, S. Spada, and C. Trombini, Eur. J. Org. Chem., 1998, 2361.
b) P. G. M. Wuts and Y. W. Jung, J. Org. Chem., 53, 1957 (1988).
19 For activation of an organolead compound by Bu4NBr see: A. Tanaka, T.
Hamatani, S. Yamashita, and S. Torri, Chem. Lett., 1986, 1461; H. Tanaka, S.
Yamashita, T. Hamatani, Y. Ikemoto, and S. Torri, Chem. Lett., 1986, 1611.
OH
2g
O
Cl
Cl
2h
2i
OH
OH
O
aAll the products were characterised by NMR, IR, and MS spectroscopy.
these to be ineffective or to give poor yields. Approximately
0.2 equiv. of Bu4NBr was found to be sufficient for these reactions
to proceed and the use of a large excess did not lead either to higher
yields or faster reaction rates. It is also interesting to note that the
nature of the solvent controlled the formation of homoallylated
products. The reaction is not effective and various by-products
are formed when acetonitrile or THF is used as the solvent. The re-
action failed to produce any desired compound when DMF alone
was used as the solvent. Also, no isolable product was formed
when the reaction was run in water. After screening the reaction
conditions, the optimum solvent for this coupling reaction was
seen to be a 3:1 mixture of DMF–H2O. Although the detailed
mechanism of the reaction is not clear, it is likely that Bu4NBr ef-
fects the generation of an active allylgallium or allylsamarium re-
agent,19 and the epoxides undergo rearrangement to form the cor-
responding aldehydes A.20 These in situ generated aldehydes re-
acts rapidly with organogallium or organosamarium reagents to af-
ford the corresponding homoallylic alcohols (Scheme 2).10,21
20 For
a GaCl3-catalyzed isomerisation of epoxides to aldehydes see: G. S.
Viswanathan and C.-J. Li, Synlett, 2002, 1553.
21 a) C.-C. Wang, S.-Y. Luo, C.-R. Shie, and S. C. Hung, Org. Lett., 2, 847 (2000). b)
W. Lee, K.-H. Kim, M. D. Surman, and M. J. Miller, J. Org. Chem., 68, 139
(2003).
22 General procedure for the regioselective cleavage of epoxides and synthesis of ho-
moallylic alcohols: To a suspension of gallium (0.5 g, 2 mmol), allyl bromide
(0.48 g, 4 mmol) and TBAB (0.064 g, 0.2 mmol) was taken in 15 mL DMF–H2O
(3:1) in a round bottom flask placed in an ice-bath. The mixture was then stirred
for 15 min until all the metal dissolved to form a clear solution. To this mixture
was then added a solution of styrene oxide 1 (Entry 1, 0.24 g, 2 mmol) in DMF
(3 mL). The resulting mixture was stirred for 3 h at room temperature. After com-
pletion (monitored by tlc) the reaction mixture was quenched with NH4Cl solu-
tion, extracted with ethyl acetate (3 ꢂ 20 mL) and washed with water and brine
solution (3 ꢂ 20 mL). The combined ether extracts were dried over anhydrous so-
dium sulphate and the residue obtained thereafter on evaporation of the solvent
was subjected to column chromatography using ethyl acetate-hexane (1:5) as elu-
ent to afford the pure product 1-phenyl-4-penten-2-ol 2 in 85% yield. Similarly,
other epoxides (Entries 2 to 12) were reacted with allylgallium reagent to get
the corresponding homoallyl alcohols in high yields. Even the sterically hindered
cyclic epoxides and styrene oxides or epichlorohydrin showed good yields of the
target products. The reactions are generally clean and no trace of side product
could be detected in the NMR spectra of the crude products. The reaction with sa-
marium metal was carried out similarly and the corresponding homoallylic alco-
hols were isolated in comparable yields. All products were characterized by infra-
red and 1H NMR spectroscopy, and finally by comparison with authentic samples.
2a: (liquid), 1H NMR (CDCl3) ꢁ: 1.48 (br, s, 1H, OH), 2.05–2.40 (m, 2H), 2.65–
2.75 (m, 2H), 3.70–3.76 (m, 1H), 5.10–5.19 (m, 2H), 5.90 (m, 1H), 7.15–7.40 (m,
5H). IR ꢂmax/KBr/cmꢁ1: 3350, 2925, 1645, 1500, 1450, 1080, 1035, 945, 740.
EIMS m=z 162 Mþ, 121, 103, 92. Anal. Calcd for C11H14O: C, 81.48; H,
10.60. Found C, 81.55; H, 1052. 2g liquid: 1H NMR (CDCl3) ꢁ: 1.42–1.60 (m,
4H), 1.88 (br, s, 1H, OH), 2.48 (m, 2H), 2.95 (d, 2H, j ¼ 18 Hz), 5.10 (dd, 1H,
j ¼ 1:7, 10 Hz), 5.18 (dd, 1H, J ¼ 1:7, 18 Hz), 5,85–5.98 (m, 1H), 7.09–7.28
(m, 4H). IR ꢂmax/KBr/cmꢁ1: 3410, 2900, 1645, 1490, 1450, 1075, 1030, 945,
O
Ga/Sm, Bu4NBr
R
CHO
R
R
X
OH
1
A
2
Scheme 2.
In conclusion, this simple and easily reproducible coupling of
epoxides with allyl halide22 using samarium and gallium under
aqueous conditions affords various homoallylic alcohols of poten-
tially high synthetic utility in high yields, without the formation of
any undesirable side products.
We thank Department of Science and Technology (DST),
New Delhi for financial assistance to this work. One of us (MG)
thanks CSIR, New Delhi for the award of a senior research fellow-
ship to him.
References and Notes
1
For reviews see: a) C.-J. Li, Tetrahedron, 52, 5643 (1996). b) C.-J. Li, Chem. Rev.,
93, 2023 (1993). c) A. Labineau, J. Avgi, and Y. Queneau, Synthesis, 1994, 741.
a) C.-J. Li and T. H. Chan, ‘‘Organic Reations in Aqueous Media,’’ John Wiley &
Sons, New York (1997). b) P. A. Grieco, ‘‘Organic Synthesis in Water,’’ Blackie
Academic & Professional, London (1998).
2
745. EIMS m=z 188 Mþ
, Anal. Calcd for C13H16O: C, 82.98; H, 8.51.
Found C, 83.05; H, 8.61. 2i: (liquid), 1H NMR (CDCl3) ꢁ: 1.45–2.50 (m, 5H,
–CH2CH2–, OH), 3.25–4.05 (m, 3H, –CH2ClCH–), 4.70–6.20 (m, 3H,
–CH2=CH–). EIMS m=z 135 Mþ, 100, 84, 43. Anal. Calcd for C6H11OCl: C,
53.55; H, 8.18; Found C, 53.61; H, 8.24.
3
M. Bartok and K. L. Lang, in ‘‘The Chemistry of Functional groups: The Chem-
istry of Ethers, Crown Ethers, Hydroxyl Groups and Their Sulphur Analogues,’’
ed. by S. Patai, John Wiley and Sons Ltd., New York (1980), Suppl. E, Part 2,
Chap. 14, pp 609–681.
Published on the web (Advance View) December 18, 2004; DOI 10.1246/cl.2005.90