Direct preparation of allylic indium(III) reagents from allylic alcohols via a reductive transmetalation of π-allylnickel(II) with indium(I) iodide

Article abstract of DOI:10.1021/jo049394t

InI-mediated direct allylation of carbonyl compounds with allylic alcohols proceeded smoothly with catalytic amounts of Ni(acac)₂ and PPh ₃ to give the corresponding homoallylic alcohols in high yields. Allylindium compounds were shown to be the real allylating agents in the present system. Substituted allylic alcohols gave branched homoallylic alcohols with syn-selectivity irrespective of the geometry of the starting allylic alcohols, whereas high anti-selectivity was observed when a bulky substituent is present in the allylic alcohols. The outcome of the diastereoselectivity is discussed on the basis of the reaction mechanism, comparing with the corresponding Pd-catalyzed version. Another distinct behavior between the Ni- and Pd-catalyzed allylation was demonstrated in the reaction of hex-1,5-diene-3,4-diol derivatives: the Pd catalyst did not give any coupling product, whereas the Ni-catalyzed InI-mediated reaction with benzaldehyde afforded the 1:1 and 1:2 adduct diols selectively depending on the reaction conditions.

Full text of DOI:10.1021/jo049394t

Dir ect P r ep a r a tion of Allylic In d iu m (III) Rea gen ts fr om Allylic
Alcoh ols via a Red u ctive Tr a n sm eta la tion of π-Allyln ick el(II) w ith
In d iu m (I) Iod id e
Tsunehisa Hirashita, Shinya Kambe, Hiromitsu Tsuji, Hideki Omori, and Shuki Araki*
Omohi College, Graduate School of Engineering, Nagoya Institute of Technology,
Gokiso-cho, Showa-ku, Nagoya 466-8555, J apan
araki.shuki@nitech.ac.jp
Received April 12, 2004
InI-mediated direct allylation of carbonyl compounds with allylic alcohols proceeded smoothly with
catalytic amounts of Ni(acac)2 and PPh3 to give the corresponding homoallylic alcohols in high
yields. Allylindium compounds were shown to be the real allylating agents in the present system.
Substituted allylic alcohols gave branched homoallylic alcohols with syn-selectivity irrespective of
the geometry of the starting allylic alcohols, whereas high anti-selectivity was observed when a
bulky substituent is present in the allylic alcohols. The outcome of the diastereoselectivity is
discussed on the basis of the reaction mechanism, comparing with the corresponding Pd-catalyzed
version. Another distinct behavior between the Ni- and Pd-catalyzed allylation was demonstrated
in the reaction of hex-1,5-diene-3,4-diol derivatives: the Pd catalyst did not give any coupling
product, whereas the Ni-catalyzed InI-mediated reaction with benzaldehyde afforded the 1:1 and
1:2 adduct diols selectively depending on the reaction conditions.
In tr od u ction
from allylic alcohols. The direct use of allylic alcohol is
advantageous in view of synthetic efficiency and green
chemistry; it facilitates the access to a broad range of
allylindium reagents without transformation to allylic
alcohol derivatives bearing a wasteful leaving group. We
report here that allylindium reagents can be prepared
directly from allylic alcohols via reductive transmetala-
tion of a π-allylnickel intermediate with InI by the use
of Ni(acac)2 and PPh3. The coupling in situ with carbonyl
compounds gave the corresponding homoallylic alcohols
in a much shorter time than that involving the Pd
catalyst.
Allylation of carbonyl compounds with a variety of
allylmetal reagents has been widely studied for carbon-
carbon bond formation.1 Among them, allylindium re-
agents have received much attention in the past decade,
because they are water-tolerant reagents.2 Allylindium
reagents are easily prepared from allylic bromides or
allylic iodides with metallic indium in polar solvents
including water. Recently, we developed Pd-catalyzed InI-
mediated allylation of aldehydes,3 where various allylic
compounds including allylic alcohols are used as the
allylic source.3a Although Pd-catalyzed allylation of car-
bonyl compounds with allylic alcohols has hitherto been
achieved by using SnCl2 and Et3B as reductants for the
intermediate π-allypalladium, it usually requires a longer
time compared with the reaction employing allylic ha-
lides.4 The Pd-catalyzed In-mediated allylation also
proceeds slowly and there remains a need to develop
efficient methods for the preparation of allylating agents
Resu lts a n d Discu ssion
By using allyl acetate as a representative allylic com-
ponent, Barbier-type nickel-catalyzed indium-mediated
allylations of benzaldehyde were examined under various
conditions. Ni(cod)2 and Ni(acac)2 gave the corresponding
homoallylic alcohol 1a in moderate yields in 1,3-dimethyl-
2-imidazolidinone (DMI) (Table 1, entries 1 and 2). With
heating, the yield was increased to 82% yield (entry 3).
The use of the triphenylphosphine ligand accelerated the
reaction and quantitative yields of 1a were attained
(entries 4-6).
Next, allylic compounds with different leaving groups
were subjected to this allylation (Table 2). Allyl methyl
sulfide afforded 1a in moderate yield in 24 h with
catalytic amounts of Ni(acac)2 (5 mol %) and PPh3 (20
mol %) (entry 1). Allyl chloride, allyl phenyl ether, and
even allyl alcohol (2a ) gave 1a in high yields in only 1 h
(entries 2-4). The reaction of 2a in THF or CH2Cl2 also
proceeded smoothly to give 1a in high yields (entries 5
(1) (a) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207. (b) Rouch,
W. R. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I.,
Eds.; Pergamon Press: Oxford, UK, 1991; Vol. 2, Chapter 1.1, p 1.
(2) (a) Cintas, P. Synlett 1995, 1087. (b) Marshall, J . A. Chemtracts:
Org. Chem. 1997, 10, 481. (c) Li, C.-J .; Chan, T.-H. Tetrahedron 1999,
55, 11149. (d) Araki, S.; Hirashita, T. In Main Group Metals in Organic
Synthesis; Yamamoto, H., Oshima, K., Eds.; Wiley-VCH: Weinheim,
Germany, 2004; Vol. 1, pp 323-386.
(3) (a) Araki, S.; Kamei, T.; Hirashita, T.; Yamamura, H.; Kawai,
M. Org. Lett. 2000, 2, 847. (b) Araki, S.; Kameda, K.; Tanaka, J .;
Hirashita, T.; Yamamura, H.; Kawai, M. J . Org. Chem. 2001, 66, 7919.
(c) Araki, S.; Kambe, S.; Kameda, K.; Hirashita, T. Synthesis 2003,
751. (d) Hirashita, T.; Kamei, T.; Satake, M.; Horie, T.; Shimizu, S.
Org. Biomol. Chem. 2003, 1, 3779.
(4) (a) Takahara J . P.; Masuyama Y.; Kurusu, Y. J . Am. Chem. Soc.
1992, 114, 2577. (b) Kimura, M.; Tomizawa, T.; Horino, Y.; Tanaka,
S.; Tamaru, Y. Tetrahedron Lett. 2000, 41, 3627.
10.1021/jo049394t CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/30/2004
5054
J . Org. Chem. 2004, 69, 5054-5059

Preparation of Allylic In(III) Reagents from Allylic Alcohols
TABLE 1. Ni-Ca ta lyzed In -Med ia ted Rea ction of Allyl
Aceta te w ith Ben za ld eh yd ea
TABLE 3. Allyla tion of Ben za ld eh yd e w ith Va r iou s
Allylic Alcoh olsa
entry
catalyst (mol %)
Ni(cod)2 (10)
Ni(acac)2 (10)
Ni(acac)2 (10)
Ni(acac)2 (5), PPh3 (20)
NiBr2(PPh3)2 (10)
Ni(PPh3)4 (10)
conditions
yield, %
1
2
3
4
5
6
rt, 20 h
rt, 20 h
70 °C, 3 h
rt, 1 h
rt, 2 h
rt, 1 h
34
50
82
97
98
100
a
All reactions were carried out with allyl acetate (1.0 mmol)
and benzaldehyde (0.50 mmol) in DMI (3 mL).
TABLE 2. Allyla tion w ith Va r iou s Allylic Com p ou n d s in
Differ en t Solven tsa
entry
X
solvent
DMI
DMI
DMI
DMI
THF
CH2Cl2
DMI-H2O
t, h
yield, %
1
2
3
4
5
6
7
SMe
Cl
24
1
1
77
96
97
96
OPh
OH
OH
OH
OH
1
5 (13)b
97 (76)b
95
5
3
trace
a
All reactions were carried out with allyl compound/benz-
aldehyde/InI ) 2/1/2, Ni(acac)2 (5 mol %), and PPh3 (20 mol %) at
b
room temperature. Pd(PPh3)4 (5 mol %) was used in place of
Ni(acac)2.
and 6). It is noted that the present Ni-catalyzed allylation
with allyl alcohol (2a ) proceeded much faster than that
with the Pd catalyst. The allylation did not proceed in
aqueous DMI due to a decomposition of the nickel
catalyst (entry 7). These results show that DMI is the
best solvent regarding the reaction time.
Various allylic alcohols were then examined for the
coupling with benzaldehyde in DMI and the results are
summarized in Table 3. A cinnamylindium reagent,
prepared from cinnamyl alcohol, gave the corresponding
branched homoallylic alcohol 1b exclusively in moderate
yield (entry 1). The yield was easily improved by increas-
ing the amount of Ni(acac)2 (entry 2). Again, high
efficiency of the Ni-catalyzed cinnamylation was observed
compared with the Pd-catalyzed one (entry 3). Crotyl
alcohol also gave the branched alcohol 1c with syn-
selectivity (entries 3 and 4). The secondary allylic alcohol
2c′ is usable for the present allylation (entry 5). Almost
coincident diastereoselectivities were observed in the
reactions of the isomeric allylic alcohols (entries 3-5).
Both (E)- and (Z)-2d gave the corresponding homoallylic
alcohol 1d with coincident syn-diastereoselectivity (en-
tries 6 and 7). Interestingly, the reaction of (Z)-2d
proceeded more quickly than that of (E)-2d . Again, the
corresponding secondary allyl alcohol 2d ′ gave 1d with
syn-selectivity (entry 8). In contrast, the allylic alcohols
bearing R-branched groups gave the anti-adducts selec-
tively (entries 9-11). The reaction with isobutyl-substi-
tuted allylic alcohol 2g afforded 1g with similar syn-
selectivity to 2c-d (entry 12). The cyclic allylic alcohol
2h gave the coupling product 1h in poor yield (entry 13).
a
All reactions were carried out with allylic alcohol (1.0 mmol),
PhCHO (0.50 mmol), and InI (1.0 mmol) in DMI at room temper-
ature. Numbers in parentheses show syn:anti ratios. c With
b
Pd(PPh3)4 in place of Ni(acac)2/PPh3.
Prenylindium cannot be prepared from 2i even with an
increasing amount of the nickel catalyst (entry 14).
J . Org. Chem, Vol. 69, No. 15, 2004 5055

Hirashita et al.
TABLE 4. Dia ster eoselectivity in th e Cr otyla tion of Va r iou s Ald eh yd esa
crotyl compod
Ni(acac)2/PPh3,
yield, %
entry
X
E:Z
R
mol %
conditions
(syn:anti)b
1
2
3
4
5
6
7
8
9
OAc
Cl
OAc
Cl
OH
OAc
OH
OH
OH
OH
OH
86:14
85:15
86:14
85:15
0:100
0:100
84:16
84:16
84:16
84:16
84:16
Ph
Ph
Ph
Ph
Ph
Ph
15/60
15/60
15/60
10/40
20/80
20/80
40/160
40/160
40/160
40/160
40/160
DMI, 4 h
DMI, 1 h
THF, 6 h
THF, 20 h
DMI, 2 h
DMI, 2 h
DMI, 6 h
DMI, 6 h
DMI, 5.5 h
DMI, 6 h
DMI, 3 h
1c: 90 (75:25)
1c: 100 (80:20)
1c: 63 (59:41)
1c: 94 (70:30)
1c: 81 (79:21)
1c: 84 (81:19)
3c: 87 (54:46)
4c: 82 (50:50)
5c: 55 (81:19)
6c: 56 (72:28)
7c: 88 (8:70:4:18)c
PhCH2CH2
n-C6H13
c-C6H11
Me2CH
PhMeCH
10
11
a
b
All reactions were carried out in a ratio of crotyl compound/RCHO/InI ) 2/1/2. Determined by 1H NMR. c Diastereomeric ratio.
TABLE 5. Allyla tion of 3-P h en ylp r op a n a l a n d Acetop h en on e w ith Va r iou s Allylic Alcoh olsa
entry
2
R1
R2
t, h
yield, %
syn:antib
1
2
3
4
5
6
7
8
9
2a
PhCH2CH2
PhCH2CH2
PhCH2CH2
PhCH2CH2
PhCH2CH2
PhCH2CH2
PhCH2CH2
Ph
H
H
H
H
H
H
H
Me
Me
Me
Me
1
20
6
22
24
22
16
8
3a : 95
3c: 63
3c: 90
3d : 27
3d : 48
3i: 29
3k : 92
8a : 84
8c: 55
8c: 56
8k : 30
-
2cc
2c′
(Z)-2d
2d ′
2i′
52:48
52:48
d
59:41e
-
2k
-
-
87:13
86:14
-
2a
2cc
2c′
2k
Ph
Ph
Ph
17
8
24
10
11
a
All reactions were carried out with 2 (0.50 mmol), 3-phenylpropanal (0.25 mmol) or acetophenone, InI (0.50 mmol), Ni(acac)2 (20 mol
%), and PPh3 (80 mol %) in DMI (3 mL). Determined by 1H NMR. c E:Z ) 84:16. Not determined. e Determined by GC analysis of the
corresponding acetate.
b
d
However, 2i′, an allylic isomer of 2i, gave the correspond-
ing branched homoallylic alcohol 1i in high yield (entry
15). Even the sterically demanding tertiary allylic alcohol,
linalool (2j), underwent allylation efficiently to give 1j
in good yield (entry 16). Methallyl alcohol (2k ) afforded
1k in high yield (entry 17).
(entries 7-11). The reaction of 3-phenylpropanal with
crotyl alcohol gave 3c in high yield. A larger amount (40
mol %) of Ni(acac)2 was required to complete the re-
action in comparison with the case of benzaldehyde
(entry 7). Heptanal gave 4c in high yield (entry 8). The
syn-selectivity was not observed for these linear ali-
phatic aldehydes. On the contrary, cyclohexanecarbox-
aldehyde afforded 5c syn-selectively, albeit in moderate
yield (entry 9). 2-Methylpropanal also afforded 6c syn-
selectively (entry 10). 2-Phenylpropanal selectively gave
one diastereomer 7c of the possible four isomers (entry
11).
Next, the allylation of 3-phenylpropanal and acetophe-
none with various allylic alcohols was investigated and
the results were summarized in Table 5. The reaction of
2a gave 3a in high yield (entry 1). Although crotyl alcohol
(2c) gave 3c in only 63% yield, its isomer 2c′ afforded 3c
in 90% yield in a shorter time (entries 2 and 3). The same
tendency was observed in the reaction of (Z)-2d and 2d ′
(entries 4 and 5). Again, the diastereoselectivity was
lower than that with benzaldehyde (entries 3-8 in Table
3). Although the prenylation of benzaldehyde with the
The unique syn-selective crotylation of benzaldehyde
observed in Table 3 prompted us to investigate the
reaction of a variety of crotyl compounds. The reaction
of E-rich crotyl acetate and crotyl chloride in place of
crotyl alcohol gave 1c also syn-selectively (Table 4,
entries 1 and 2). The corresponding Pd-catalyzed reaction
with crotyl chloride afforded 1b with a slight anti-
selectivity (46:54). In THF the syn-selectivity was de-
creased (entries 3 and 4). (Z)-Crotyl alcohol and (Z)-crotyl
acetate also gave 1c syn-selectively (entries 5 and 6),
indicating that the stereochemical outcome of the present
Ni-catalyzed crotylation is independent from the geom-
etry of the starting allylic compounds, and suggesting
that the crotylation proceeds via the same intermediate.
To examine the generality of the syn-selectivity, croty-
lation of a variety of aliphatic aldehydes was studied
5056 J . Org. Chem., Vol. 69, No. 15, 2004

Preparation of Allylic In(III) Reagents from Allylic Alcohols
TABLE 6. Nick el-Ca ta lyzed Allyla tion of Ben za ld eh yd e
in th e P r esen ce or Absen ce of In d iu m Sa ltsa
Syn products were formed predominantly from nickel-
catalyzed reactions of benzaldehyde, except for the cases
involving allylic alcohols with a bulky substitutent (Table
3, entries 1, 2, 9, 10, 11, and 15). In contrast, the
diastereoselectivity of the Pd-catalyzed InI-mediated
allylation can be controlled by a geometry of allylic
alcohol derivatives: (Z)- and (E)-crotyl acetates react with
benzaldehyde to give the corresponding syn- and anti-
homoallylic alcohols, respectively.3d As the reaction of
allylindium with aldehyde proceeds via a six-membered
transition state, the syn preference in the nickel-
catalyzed allylation suggests the existence of (Z)-allylic
indium intermediates regardless of the geometry of
starting allylic alcohols, unless a sterically demanding
substitution is present. A π-crotylnickel complex is known
to exist as an equilibrium of the syn- and anti-π-allyl
isomers.6 Therefore, it is reasonable to postulate the
intermediacy of anti-π-allylic nickel complexes, which
give upon transmetalation (Z)-allylic indium reagents
more readily than syn isomers give (E)-allylic indium
reagents.7 The poor diastereoselectivity with 3-phenyl-
propanal and heptanal suggests that the energy differ-
ence is small between the two transition states, in which
the alkyl group of the aldehydes occupies equatorial and
axial positions.
entry
Ni(PPh3)4, mol %
In salt, mol %
t, h
yield, %
1
2
3
4
200
200
200
10
none
6
6
6
1
0
0
0
InCl3 (200)
In(OH)3 (200)
InI (200)
100
a
All reactions were carried out with PhCHO, allyl alcohol (200
mol %), and InI (200 mol %) in THF-DMI-hexane (2:1:1) at room
temperature.
tertiary alcohol 2i′ proceeded smoothly, the reaction of
3-phenylpropanal gave 3i in lower yield (entry 6). â-Sub-
stituted allylic alcohol 2k afforded the corresponding
product 3k in high yield (entry 7). The allylation of
acetophenone also proceeded smoothly in the presence
of Ni(acac)2 (20 mol %) giving 8a in high yield (entry 8).
The allyl alcohols substituted by a methyl group at the
R, â, or γ position gave the corresponding homoallylic
alcohols 8c and 8k syn-selectively in moderate yields
(entries 9-11).
Mech a n istic Con sid er a tion . π-Allylnickel halides
are known as nucleophilic reagents; for example, dimeric
π-allylnickel bromide reacts with carbonyl compounds to
give the corresponding homoallylic alcohols.5 This fact
raises a question about the real allylating agent in the
present reaction. There is a possibility that π-allylnickel
directly reacts with the aldehyde, where InI simply plays
a role as a reductant for Ni(II) to Ni(0), and the resulting
In(III) salt acts as a Lewis acid to promote the reaction.
To examine this possibility, we carried out the reaction
of allyl alcohol with benzaldehyde in the presence of an
equimolar amount of a Ni(0) catalyst without InI: thus,
Ni(PPh3)4 was prepared by treatment of a mixture of
Ni(acac)2 and PPh3 with DIBAL and the reaction was
performed in the presence or absence of InX3 (X ) Cl or
OH, 200 mol %). The results are summarized in Table 6.
The reactions without InI did not give homoallylic alcohol
1a (Table 6, entries 1-3). In contrast, the reaction
proceeded smoothly in the presence of InI (200 mol %)
and Ni(PPh3)4 (10 mol %) to give 1a quantitatively (entry
4). These results clearly indicate that the key nucleophilic
intermediate in the present allylation is not an allylnickel
species but allylindium compounds. The most plausible
reaction mechanism involving allylindium is depicted in
Scheme 1. A Ni(0) catalyst is preformed from Ni(acac)2
and InI. A π-allylnickel intermediate, generated from
allylic alcohol and the Ni(0) catalyst, undergoes a reduc-
tive transmetalation with InI to afford an allylindium
reagent, which allylates benzaldehyde.
Ta n d em Allyla t ion w it h Hexa -1,5-d ien e-3,4-d iol
Der iva tives. As a further example, hexa-1,5-diene-3,4-
diol derivatives 9a -c were subjected to the Ni-catalyzed
InI-mediated allylation (Scheme 2). The reaction of diol
9a with 2 equiv of benzaldehyde gave the 1:2 adduct 11
together with a small amount of 12 (Table 7, entry 1).
When monoacetate 9b was reacted with 3 equiv of the
aldehyde, 11 was obtained in high yield with a small
amount of 12 (entry 2). Addition of water increased the
ratio of 12 (entry 3). In contrast, the reaction performed
in carefully dried THF gave 11 exclusively (entry 4). The
effect of water on the preference for 12 can be accounted
for by the equilibrium of the intermediates A and B. As
allylic indium reagents are coupled with carbonyl com-
pounds at the γ-carbon, diols 11 and 12 are considered
to come from A and B, respectively. The formation of A
is favored by an intramolecular chelation, whereas water
in the solvent may prevent the chelation of the alkoxide
to indium and hence the acyclic intermediate B becomes
more important. The use of a 2-fold amount of 9b to the
aldehyde led to an exclusive formation of the 1:1 adduct
10 (entry 5). Diacetate 9c gave the 1:2 adduct 11 in 50%
yield as a major product (entry 6). When the product diol
10 was subjected to the Ni-catalyzed InI-mediated reac-
tion with benzaldehyde, 11 and 12 were formed in 47%
and 28% yields, respectively. Again, addition of water
reversed the ratio (11: 29%, 12: 47%). In contrast, 9a -c
did not give any coupling product with Pd(PPh3)4. These
results exhibit additional differences in the nature of the
Pd- and Ni-catalyzed allylation. It is reported that the
Ni-catalyzed allylation of diethylamine with allylic alco-
hols is much more efficient than the corresponding Pd-
involving one.8 This fact is in accord with the high
It is interesting to note that the distinction between
the nickel and palladium catalysts in the InI-mediated
allylation is clearly marked by the diastereoselectivity.
(5) (a) Billington, D. C. Chem. Soc. Rev. 1985, 14, 93. (b) Hegedus,
L. S.; Wagner, S. D.; Siirala-Hansen, K. J . Org. Chem. 1975, 40, 593.
(c) Hegedus, L. S.; Evans, B. R.; Korte, D. E.; Waterman, E. L.; Sjo¨berg,
K. J . Am. Chem. Soc. 1976, 98, 3901. (d) Semmelhack, M. F.; Wu, E.
S. C. J . Am. Chem. Soc. 1976, 98, 3384. (e) Semmelhack, M. F.;
Yamashita, A.; Tomesch, J . C.; Hirotsu, K. J . Am. Chem. Soc. 1978,
100, 5565. (f) Semmelhack, M. F.; Brickner, S. J . J . Am. Chem. Soc.
1981, 103, 3945. (g) Hegedus, L. S.; Varaprath, S. Organometallics
1982, 1, 259.
(6) (a) Lehmkuhl, H.; Rufinska, A.; Mehler, K.; Benn, R.; Schroth,
G. Liebigs Ann. Chem. 1980, 744.
(7) A Z-selective alkylation of crotylnickel with dimethylmalonate
is also reported: Bricout, H.; Carpentier, J .-F.; Mortreux, A. Tetrahe-
dron Lett. 1996, 37, 6105.
J . Org. Chem, Vol. 69, No. 15, 2004 5057

Hirashita et al.
SCHEME 1
hydrochloric acid (1 N) was added and the product was
extracted with ether. The extracts were washed with water
and dried over anhydrous sodium sulfate. The solvent was
removed, and the residue was separated by chromatography
on silica gel (EtOAc:hexane ) 1:4) to give 1-phenylbut-3-en-
1-ol (1a ) (71 mg, 96%).
SCHEME 2
Products 1a ,9 1b,10 1c,9 1d ,4a 1e,11 1f,12 1h ,9 1i,9 1j,9 1k ,9
3a ,9 3c,9 3i,9 3k ,9 4c,13 5c,9 6c,14 7c,15 8a ,4a 8c,16 and 8k 17 are
all known compounds.
2-Isobu tyl-1-p h en ylbu t-3-en -1-ol (1g): 1H NMR (300
MHz, CDCl3) (syn isomer) δ 0.84 (dd, J ) 12, 6.6 Hz, 6H),
1.09-1.37 (m, 2H), 1.42-1.67 (m, 1 H), 1.92 (br s, 1H), 2.26-
2.61 (m, 1H), 4.60 (d, J ) 5.6 Hz, 1H), 5.13-5.30 (m, 2H),
5.38-5.76 (m, 1H), 7.18-7.38 (m, 5H). 1H NMR (300 MHz,
CDCl3) (anti isomer) δ 0.76 (dd, J ) 18, 6.6 Hz, 6 H), 4.34 (d,
J ) 7.8 Hz, 1H), 4.96-5.13 (m, 2H). 13C NMR (50 MHz, CDCl3)
δ 21.1, 21.4, 24.0, 24.1 25.4, 38.9, 39.7, 49.3, 50.7, 117.1, 118.4,
126.5, 126.8, 127.1, 127.4, 127.7, 128.0, 138.8, 139.2, 142.1.
IR (neat, cm-1) 3400, 1635. Anal. Calcd for C14H20O: C, 82.30;
H, 9.87. Found: C, 82.15; H, 10.10.
1-P h en yl-4-p r op ylh ex-5-en -3-ol (3d ): 1H NMR (200 MHz,
CDCl3) δ 0.88 (t, J ) 6.7 Hz, 3H), 1.12-1.49 (m, 4H),
1.63-1.95 (m, 2H), 1.99-2.20 (m, 1H), 2.55-2.94 (m,
2H), 3.42-3.54 (m, 1H), 5.02-5.20 (m, 2H), 5.47-5.71 (m,
1H), 7.15-7.32 (m, 5H). IR (neat, cm-1) 3400, 1495. Anal.
Calcd for C15H22O: C, 82.52; H, 10.16. Found: C, 82.28; H,
10.10.
1,2-Dieth en yl-3-p h en ylp r op a n -1,3-d iol (10): 1H NMR
(200 MHz, CDCl3) (major isomer) δ 2.28 (br s, 2H), 2.26-2.38
(m, 1H), 4.20-4.29 (m, 1H), 4.90-5.35 (m, 5H), 5.69-6.07 (m,
2H), 7.21-7.35 (m, 5H). 1H NMR (200 MHz, CDCl3) (minor
isomer) δ 2.47-2.60 (m, 1H). Elemental analysis was per-
formed on the diacetate. Diacetate of 10: 1H NMR (200 MHz,
CDCl3) δ 2.02-2.08 (m, 6H), 2.60-2.80 (m, 1H), 4.79-5.28 (m,
4H), 5.63-6.03 (m, 2H). Anal. Calcd for C17H20O4: C,70.81;
H, 6.99. Found: C, 71.07; H, 6.94.
efficiency of the present Ni-catalyzed allylation with
allylic alcohols.
Con clu sion
1
The umpolung of π-allylnickel with InI provides a new
route for allylindium reagents by the direct use of allylic
alcohols. The high availability of a wide range of allylic
alcohols greatly enhances the usefulness of the present
Ni-catalyzed reaction. High syn-selectivity is generally
observed in the coupling with carbonyl compounds,
regardless of the geometry of allylic alcohols. Further
applications of this new process are currently underway.
2-Eth en yl-1,6-d ip h en ylh ex-3-en -1,6-d iol (11): H NMR
(200 MHz, CDCl3) (major isomer) δ 2.20-2.50 (m, 2H), 2.72
(br s, 2H), 2.94-3.08 (m, 1H), 4.37-4.66 (m, 2H), 4.78-5.86
(m, 3H), 7.10-7.38 (m, 10H). 1H NMR (200 MHz, CDCl3)
(minor isomer) δ 3.37-3.58 (m, 1H). Elemental analysis was
performed on the diacetate. Diacetate of 11: 1H NMR (200
MHz, CDCl3) (major isomer) δ 2.02-2.06 (m, 6H), 2.42-2.72
(m, 2H), 3.09-3.26 (m, 1H), 4.77-5.20 (m, 2H), 5.22-5.44 (m,
(9) Kobayashi, S.; Nishino, K.; J . Org. Chem. 1994, 59, 6620.
(10) Coxon, J . M.; Van Elk, S. J .; Steel P. J . Tetrahedron 1989, 45,
1029.
Exp er im en ta l Section
(11) Lombardo, M.; Marganti, S.; Tozzi, M.; Trombini, C. Eur. J .
Org. Chem. 2002, 2823.
(12) Lombardo, M.; Marganti, S.; Trombini, C. J . Org. Chem. 2000,
68, 8786.
(13) Fu¨rstner, A.; Voigtla¨nder, D. Synthesis 2000, 7, 959.
(14) Steel, P. G.; Thomas, E. J . J . Chem. Soc., Perkin Trans. 1 1997,
Typ ica l Exp er im en ta l P r oced u r e (Ta ble 2, en tr y 4).
To a mixture of Ni(acac)2, prepared by heating Ni(acac)2‚2H2O
(7.0 mg, 0.025 mmol) under reduced pressure, PPh3 (26 mg,
0.1 mmol), and indium(I) iodide (0.24 g, 1.0 mmol) in DMI (1.5
mL) were added allyl alcohol (68 µL, 1.0 mmol) and benzal-
dehyde (52 µL, 0.50 mmol). The reaction mixture was stirred
at room temperature under argon. The reaction was monitored
by TLC. After 1.0 h, the benzaldehyde disappeared. Diluted
371.
(15) Brinkmann, H.; Hoffmann, R. W. Chem. Ber. 1991, 123,
2395.
(16) Schloss, J . D.; Paquette, L. A. Synth. Commun. 1998, 28,
2887.
(8) Bricout, H.; Carpentier, J .-F.; Mortreux, A. J . Mol. Catal. A:
Chem. 1998, 136, 243.
(17) Bates, R, B.; Beavers, W. A.; Gordon, B., III J . Org. Chem. 1979,
44, 3800.
5058 J . Org. Chem., Vol. 69, No. 15, 2004

Preparation of Allylic In(III) Reagents from Allylic Alcohols
TABLE 7. Nick el-Ca ta lyzed Rea ction of Hexa -1,5-d ien e-3,4-d iol Der iva tivesa
yield, %, and diastereomeric ratiob
entry
9 (9:PhCHO)
solvent
t, h
10
11
12
1
9a (1:2)
9b (1:3)
9b (1:3)
9b (1:3)
9b (2:1)
9c (1:3)
DMI
DMI
DMI
THF
DMI
DMI
3
2.5
20
15
22
1.5
0
65 (39:17:12:32)
61 (29:30:21:20)
34 (25:27:26:22)
61 (26:17:15:42)
0
6c
2
trace
trace
trace
84 (1:32:66:1)
0
17 (3:5:55:37)
27 (4:8:75:13)
0
0
3d
4
5
6
50 (26:17:13:44)
11 (20:40:25:15)
a
b
All reactions were performed on a 0.25-mmol scale of 9 in the presence of Ni(acac)2 (20 mol %) and PPh3 (80 mol %). Yields and
diastereomeric ratios were determined by GC analysis. c Diastereomeric ratio was not determined. H2O (3 mmol) was added.
d
2H), 5.50-5.81 (m, 3H). 1H NMR (200 MHz, CDCl3) (minor
isomer) δ 3.44-3.56 (m, 1H). Anal. Calcd for C24H26O4: C,
76.17; H, 6.92. Found: C, 75.86; H, 7.07.
performed on the diacetate. Diacetate of 12: Anal. Calcd for
C
24H26O4: C, 76.17; H, 6.92. Found: C, 76.04; H, 6.86.
2,3-Dieth en yl-1,4-d ip h en ylbu ta n -1,4-d iol (12): 1H NMR
(300 MHz, CDCl3) (major isomer) δ 1.90 (br s, 2H), 2.19 (t, J
) 9.0 Hz, 2H), 4.42 (d, J ) 9.0 Hz, 2H), 5.15 (dd, J ) 17, 1.8
Hz, 2H), 5.41 (dd, J ) 10, 1.8 Hz, 2H), 5.93 (dt, J ) 17, 10
Hz), 7.03-7.13, 7.24-7.37 (m, 10H). 1H NMR (300 MHz,
CDCl3) (minor isomer) δ 2.21 (t, J ) 8.0 Hz, 2H), 4.50 (d, J )
8.4 Hz, 2H), 5.52-5.78 (m, 2H). Elemental analysis was
Ack n ow led gm en t. This work was partially sup-
ported by a Grant-in-Aid for Scientific Research (No.
14340195) from the Ministry of Education, Science,
Sport and Culture, J apan.
J O049394T
J . Org. Chem, Vol. 69, No. 15, 2004 5059