Formation and reactions of alkylzinc reagents in room-temperature ionic liquids

Article abstract of DOI:10.1021/jo051772m

The presence of a suitable amount of bromide or chloride ions was found to be critical in forming the alkylzinc reagents from alkyl iodides and zinc metal in the room-temperature ionic liquid, N-butylpyridinium tetrafluoroborate. β-Hydride transfer in the reactions of butylzinc reagents with aldehydes can also be reduced by a bromide ion.

Full text of DOI:10.1021/jo051772m

Formation and Reactions of Alkylzinc Reagents in
Room-Temperature Ionic Liquids
Man Chun Law,†,‡ Kwok-Yin Wong, and Tak Hang Chan*
‡
,†,‡
Department of Chemistry, McGill University, Montreal, Quebec, Canada H3A 2K6, and
Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University,
Hung Hom, Hong Kong SAR, China
tak-hang.chan@mcgill.ca; bcchanth@polyu.edu.hk
Received August 22, 2005
The presence of a suitable amount of bromide or chloride ions was found to be critical in forming
the alkylzinc reagents from alkyl iodides and zinc metal in the room-temperature ionic liquid,
N-butylpyridinium tetrafluoroborate. â-Hydride transfer in the reactions of butylzinc reagents with
aldehydes can also be reduced by a bromide ion.
Introduction
fore cannot be used for the preparation or reactions of
such reactive organometallic reagents.7 The study of
organometallic reactions in room-temperature ionic liq-
uids (RTILs) has thus far been limited to the less-reactive
Organozinc reagents have been studied extensively
since they were first prepared by Frankland in 1848.
Organozinc reagents of the type RZnX are usually
prepared in conventional organic solvents such as tetra-
hydrofuran, hexane, toluene, and ether. The choice of
solvent plays an important role in the reactivity and
stability of these compounds. Because of concern for the
environment, there has been considerable recent research
into replacing the use of volatile organic solvents as the
1
8
allyl metal reagents or their variations. This is because
the commonly used imidazolium-based RTILs react with
reactive organometallic reagents to give imidazol-2-
9
ylidenes (N-heterocyclic carbenes, NHCs). Recently, we
2
found that the RTIL N-butylpyridinium tetrafluorobo-
rate, [bpy][BF
4
], was stable toward diethylzinc and
preformed diethylzinc could be used to react with alde-
3
4
5
reaction media. Water, supercritical carbon dioxide,
9,10
4
hydes in [bpy][BF ] to give the adducts in high yields.
6
and room-temperature ionic liquids are the most com-
monly explored media to replace organic solvents. Reac-
tive organometallic reagents such as alkylzinc, Grignard,
or alkyllithium are well-known to react vigorously with
water or carbon dioxide. Those two “clean” media there-
Furthermore, the ionic liquid could be easily recovered
9
and reused. Because the preformed organozinc reagents
have to be prepared in volatile organic solvents, we are
therefore interested in the possibility of generating
organozinc reagents directly from zinc metal and alkyl
halides in RTILs. In a recent report,10 preformed phen-
ylmagnesium bromide in THF was also found to be stable
in phosphonium ionic liquids and could undergo the
normal reactions expected of Grignard reagents; however,
†
McGill University.
‡
The Hong Kong Polytechnic University.
(
1) Frankland, E. Liebigs Ann. Chem. 1849, 71, 171.
(
2) (a) For the preparation of organozinc reagents, see: Synthetic
Methods of Organometallic and Inorganic Chemistry; Hermann, W.
A., Ed.; Thieme: Stuttgart, Germany, 1999; Vol. 5. (b) Huo, S. Org.
Lett. 2003, 5, 423.
(7) For the rather extensive study of less-reactive organometallic
reactions in aqueous media, see, for examples: (a) Li, C. J.; Chan, T.
H. Tetrahedron 1999, 55, 11149. (b) Petrier, C.; Luche, J. L. J. Org.
Chem. 1985, 50, 910. (c) Keh, C. C. K.; Wei, C.; Li, C. J. J. Am. Chem.
Soc. 2003, 125, 4062 and references therein.
(8) (a) Law, M. C.; Wong, K. Y.; Chan, T. H. Green Chem. 2002, 4,
161. (b) Gordon, C. M.; Ritchie, C. Green Chem. 2002, 4, 124. (c)
Kitazume, T.; Kasai, K. Green Chem. 2001, 3, 30. (d) Gordon, C. M.;
Mccluskey, A. Chem. Commun. 1999, 1431.
(9) Law, M. C.; Wong, K. Y.; Chan, T. H. Green Chem. 2004, 6, 241.
(10) Gorodetsky, B.; Ramnial, T.; Branda, N. R.; Clyburne, J. A. C.
Chem. Commun. 2004, 1972.
(3) (a) Anastas, P. T.; Warner, J. C. Green Chemistry: Theory and
Practice; Oxford Science Publications: Oxford, NY, 1998. (b) Abraham,
M.; Moens, L. Clean Solvents: Alternative Media for Chemical Reac-
tions and Processing; ACS Symposium Series No. 819; American
Chemical Society: Washington, DC, 2001.
(
4) Li, C. J.; Chan, T. H. Organic Reactions in Aqueous Media; John
Wiley & Sons: New York, 1997.
5) Jessop, P. Chemical Synthesis Using Supercritical Fluids; Wiley-
VCH: Weinheim, Germany, 1999.
6) Rogers, R. D.; Seddon, K. R. Ionic Liquids: Industrial Applica-
tions to Green Chemistry; Oxford University Press: Washington, 2002.
(
(
10.1021/jo051772m CCC: $30.25 © 2005 American Chemical Society
10434
J. Org. Chem. 2005, 70, 10434-10439
Published on Web 11/11/2005

Alkylzinc Reagents in Room-Temperature Ionic Liquids
SCHEME 1
TABLE 1. Chemical Shifts of Methylene Protons of
Various Ethylzinc Compounds in Different Solvents
compound
toluenea Et2Ob CD3CN c
ethylzinc species prepared in RTIL
0.28
EtZnCl
EtZnBr
EtZnI
0.65
0.69
0.34
0.13
0.22
0.28
0.35
0.21
Et2Zn
SCHEME 2
a
0% solution in toluene. b 10% solution in Et2O. c 2 M solution
1
of EtZn species in [bpy][BF4] with 3 mmol of pyridine dissolved in
CD3CN.
TABLE 2. Ethylation of Benzaldehyde in RTILa
the Grignard reagent could not be formed from magne-
sium and the halide in the ionic liquid.10 The issues that
are of interest to us are: (1) Can alkylzinc compounds
be formed directly from alkyl halides and zinc metal in
RTIL? (2) What is the structure of the alkylzinc reagent
formed? (3) Do they show a reactivity pattern in RTIL
different from that in traditional organic solvents?
ratio of Zn/EtI [bpy][Br] added temp yield of 3a
b
entry
(mmol)
(% Br)
(°C)
(%)
1
2
3
4
5
6
7
1:1
1:2
(5.1)
(5.1)
(5.1)
(5.1)
(5.1)
(5.1)
(5.1)
(0)
0.05 g (4.3)
0.06 g (5.1)
0.075 g (6.4)
0.1 g (8.5)
50/rt
50/rt
50/rt
50/rt
40/40
rt/rt
50/rt
50/rt
50/rt
50/rt
50/rt
50/rt
50/rt
50/rt
0
0
2:1
65
84
5
1.5:1.5
2:2
2:2
8
2:2
(92)
0
c
8
2:2
Results and Discussion
c
d
9
2:2
0
c
d
1
1
1
1
0
1
2
3
2:2
89
A. Formation of the Alkylzinc Intermediate in
c
c
c
c
d
2:2
83
[
bpy][BF
4
]. Initially, we examined the reaction of zinc
d
2:2
(94)
4
metal with ethyl iodide (1a) in [bpy][BF ] to give the
d
d
2:2
0.15 g (12.8)
0.2 g (17.1)
89
10
presumed ethylzinc intermediate which was then
quenched with benzaldehyde (2a) to give the adduct
alcohol 3a (Scheme 1). The reaction was found to be
highly irreproducible, giving at times little or no forma-
tion of the ethylzinc intermediate (see below) and poor
yield of the alcohol 3a. After consideration of various
parameters such as the mesh size, the origin of the zinc
metal used, the ambient temperature, or the humidity,
we came to realize that the reaction depended critically
14
2:2
a
Reaction conditions: Zn and EtI in the indicated mmol ratios
were stirred in RTIL (1 mL) for 1 h at 50 °C. Benzaldehyde (1
mmol) was then added, and the stirred mixture was brought to
-
the indicated temperature for 12 h. Impure [bpy][BF4] (5.1% Br )
b
was used except for with entries 8-14. Product yield was either
1
determined by H NMR or by isolation (isolated yields in paren-
c
d
theses). Entries 8-14: pure [bpy][BF4] was used. Entries 9-14:
indicated amount of [bpy][Br] was added to pure [bpy][BF4].
on the batch and purity of [bpy][BF
used was usually prepared from [bpy][Br] followed by
a metathesis reaction with NaBF . Because the meta-
thetic reaction is an equilibrium process (Scheme 2), the
bpy][BF ] generated inevitably contained a certain
amount of the bromide ion. When [bpy][BF ] had been
meticulously purified using the chromatography method,
we were able to obtain pure [bpy][BF ] with less than
.2% of bromide ion, determined by an electrochemical
method (see Supporting Information). The impure [bpy]-
BF ] obtained before chromatographic purification, on
4 4
] used. The [bpy][BF ]
11
To deduce the nature of the ethylzinc intermediate, we
1
examined the H NMR of the solution in CD
3
CN or THF-
4
8
d but could not observe any meaningful signals. How-
ever, by adding a small amount of pyridine to the reaction
mixture to stabilize the ethylzinc intermediate followed
[
4
4
12
3
by dissolution in CD CN, we were able to detect the
ethylzinc species clearly. By comparing the chemical shift
of the methylene protons of the ethylzinc species with
those in organic solvents14 (Table 1), we concluded that
ethylzinc halide species were likely formed in the ionic
liquid.
4
0
[
4
the other hand, contained about 5.1% of bromide ion (in
The ethylzinc reagent thus generated reacted with
benzaldehyde to give the adduct 3a after the normal
aqueous workup and extraction. The reaction was rela-
tively inefficient at room temperature; however, at 50 °C,
the yield was quite satisfactory (Table 2). The optimal
ratio of Zn and EtI to benzaldehyde was found to be 2:2:1
to give a high yield (90%) of the adduct. This high
reactivity of the ethylzinc intermediate in RTIL is rather
surprising, as alkylzinc halides are generally unreactive
the form of [bpy][Br]). The reaction of zinc metal with
ethyl iodide could only occur in the impure [bpy][BF
When zinc metal powder was stirred in pure [bpy]-
BF ], nothing happened even for an extended period. On
the other hand, a deep blue color developed when zinc
metal powder (1 mmol) was stirred in impure [bpy][BF
1 mL, 5.1% bromide) and the zinc metal slowly dissolved.
4
]!
[
4
4
]
(
In the presence of ethyl iodide, the blue color slowly
changed to a green color which indicated the formation
of the ethylzinc intermediate. It is important to note that
the blue color appeared not to be due to the reduction of
the pyridinium ionic liquid by metallic zinc,13 as the NMR
spectra of the recovered ionic liquid did not show any
reduction product.
15
toward aldehydes in organic solvents. Some kinds of
activation are usually required.16 Previously, it had been
(13) It is well-known that substituted pyridinium ions can undergo
reductive coupling to give highly colored species. See: Gale, R. J.;
Osteryoung, R. A. J. Electrochem. Soc. 1980, 127, 2167.
(
14) (a) Robert, C.; Tang, J. C. J. Am. Chem. Soc. 1954, 76, 2262.
(
11) Law, M. C.; Wong, K. Y.; Chan, T. H. Green Chem. 2002, 4,
28.
12) Park, S.; Kazlauskas, R. J. J. Org. Chem. 2001, 66, 8395.
(b) Boersma, J.; Noltes, J. G. Tetrahedron Lett. 1966, 1521. (c) Hota,
N. K.; Willis, J. J. Organomet. Chem. 1967, 9, 169. (d) Boersma, J.;
Noltes, J. G. J. Organomet. Chem. 1967, 9, 551.
3
(
J. Org. Chem, Vol. 70, No. 25, 2005 10435

Law et al.
reported that addition of Bu
4
NI accelerated the nickel/
TABLE 3. Zinc-Mediated Cross-Coupling of Ethyl
Iodide with Aldehydes with Added [bpy][Br]a
palladium-catalyzed cross-coupling between benzylic zinc
bromides and alkyl iodide/alkenyl trifates.17 However, in
4
the present case, addition of Bu NI to the RTIL either
prior to or after the formation of the ethylzinc intermedi-
ate did not have any appreciable effect in our reaction.
The possibility that pyridine, which may be present as
an impurity from the ionic liquid, may play a role in the
activation can also be ruled out. Addition of pyridine to
the organozinc intermediate prior to the addition of
benzaldehyde in fact reduced the yield of 3a.
B. Effect of Adding Bromide or Chloride Ions. To
demonstrate the importance of the bromide ion in affect-
4
ing the reaction, we used pure [bpy][BF ] mixed with
various amounts of [bpy][Br] as the reaction media for
reaction isolated yield
mole ratio
Zn/EtI
time
(h)
of 3
entry
R′ of R′CHO
(%)
1
2
3
4
5
6
7
8
9
p-ClPh
2:2
3:3
2:2
3:3
2:2
2:2
2:2
1:1
1:1
12
48
12
48
12
12
12
12
12
3c, 85
3d, 99
3e, 63
3f, 99
3g, 92
3h, 63
3i, 95
p-CNPh
p-MeOPh
p-BrPh
2,6-dichloroPh
trans-PhCHdCH-
2-pyridyl
b
(CH3)3C-
n-C7H15-
0
b
0
the same ethylation of benzaldehyde (entries 9-14). It
-
a
could be seen that there was a critical window of Br
Reaction conditions: [bpy][BF4] (1 mL) was used without
purification as the reaction media. Zinc and ethyl iodide (in mmol
specified) were added and stirred at 50 °C for 1 h. The carbonyl
compound (1 mmol) was added to the reaction mixture and stirred
concentration for the optimal results. With the amount
-
of Br below 4.3% (entries 8 and 9), there was no
-
formation of adduct 3a. The optimal amount of Br was
for the specified time at 50 °C. The mixture was worked up to
found to be 8.5% (entry 10) which gave 3a in 94% isolated
isolate 3. b The starting carbonyl compound was completely con-
-
yield. However, as the amount of Br was increased to
sumed. The product was isolated as a mixture of complex products.
1
7.1%, the yield of 3a was only 10% with a significant
amount of benzaldehyde unreacted. One consequence of
having a larger amount of [bpy][Br] was that the reaction
media became more viscous, eventually making the
reaction difficult.
TABLE 4. Zinc-Mediated Cross-Coupling of Ethyl
Iodide with Aldehydes with Added [bpy][Cl]a
4
Because purification of [bpy][BF ] by chromatography
was relatively tedious and led to substantial loss of
mole ratio [bpy][Cl] added yield of 3
-
-
materials, we used the impure [bpy][BF
4
] (5.1% Br ) to
entry
R′ of R′CHO
Zn/EtI
(% Cl )
(%)
examine the ethylation of various aldehydes (Table 3).
It was clear that all aryl aldehydes could be converted
to the corresponding ethylated alcohols 3 in relatively
good yields. Cinnamaldehyde reacted in a 1,2-fashion to
give the ethylated alcohols (entry 6). The replacement of
aryl with a heteroaryl 2-pyridyl moiety presented no
particular difficulty (entry 7). On the other hand, ali-
phatic aldehydes reacted to give a complex mixture with
complete consumption of the aldehydes (entries 8 and 9).
It was possible that, in addition to ethylation, aldol
condensation, elimination, addition, and reduction reac-
tions all occurred.18
1
2
3
4
5
6
Ph
2:2
2:2
2:2
2:2
2:2
2:2
0.047 g (5.1)
0.08 g (8.5)
0.08 g (8.5)
0.08 g (8.5)
0.08 g (8.5)
0.08 g (8.5)
3a, 0
Ph
3a, 87
3d, 86
3e, 53
3f, 83
3h, 68
p-CNPh
p-MeOPh
p-BrPh
trans-PhCHdCH
a
Reaction conditions: pure [bpy][BF4] (1 mL) was used as the
reaction media. Zinc and ethyl iodide (in mmol specified) were
added and stirred at 50 °C for 1 h. The carbonyl compound (1
mmol) was added to the reaction mixture and stirred for 12 h at
50 °C. The mixture was worked up to isolate 3.
entries 3-6) were also examined, and they showed
results similar to those with [bpy][Br]. Addition of [bpy]-
[I], however, failed to show the blue color when stirred
with zinc metal powder in [bpy][BF ] and did not give
4
adduct 3a under similar conditions.
C. Other Organozinc Reagents. We have also
examined the formation of the n-butylzinc reagent in the
4
reaction of n-butyl iodide with zinc in [bpy][BF ] (Table
5). As in the case of ethyl iodide, no formation of the blue
color or the green alkylzinc reagent was observed in pure
Addition of [bpy][Cl] was also able to initiate the
formation of an organozinc halide intermediate. A deep
blue color developed when zinc metal powder (1 mmol)
was stirred in [bpy][BF
4
] with added [bpy][Cl] to give
8
4
.5% chloride, and the zinc metal slowly dissolved (Table
, entry 2). With the amount of Cl below 8.5% (entry 1),
-
there was no formation of adduct 3a. Ethylations of
various aldehydes with activation of [bpy][Cl] (Table 4,
[
bpy][BF
4
] and unreacted benzaldehyde was recovered on
(
15) See, for examples: (a) Knochel, P.; Singer, R. D. Chem. Rev.
1
993, 93, 2117. (b) Knochel, P. Synlett 1995, 393. (c) Tamaru, Y.;
quenching the reaction mixture (entry 1). When impure
[
but no green color intermediate formed upon addition of
n-BuI. Under these conditions, the added benzaldehyde
was converted to give the pinacol coupling product
Nakamura, T.; Sakaguchi, M.; Ochiai, H.; Yoshida, Z. Chem. Commun.
-
4
bpy][BF ] (5.1% Br ) was used, the blue color appeared,
1
988, 610.
16) See: (a) Inoue, S.; Yokoo, Y. J. Organomet. Chem. 1972, 39,
1. (b) Soms a´ k, L.; N e´ meth, I. J. Carbohydr. Chem. 1993, 12, 679. (c)
(
1
Soms a´ k, L.; Madaj, J.; Wiœniewski, A. J. Carbohydr. Chem. 1997, 16,
1
075. (d) Ishino, Y.; Mizuno, T.; Ishikawa, A.; Kobayashi, J. I. Synlett
002, 2116. (e) Zhou, J.; Fu, G. C. J. Am. Chem. Soc. 2003, 125, 12527.
2
(PhCHOH)
2
(4) in low yield among other unidentified
(
f) Soms a´ k, L.; Czifr a´ k, K.; Veres, E. Tetrahedron Lett. 2004, 45, 9095.
products (entry 2). With a still higher amount of [bpy]-
[
blue color appeared followed by the formation of the green
n-butylzinc intermediate. Addition of benzaldehyde to the
solution gave a mixture of the butylated alcohol 3b
together with benzyl alcohol 5 in a ratio of 49:51 (entry
4
(17) Addition of Bu NI had been found to accelerate the nickel/
-
4
Br] added to the pure [bpy][BF ] (to give 8.5% Br ), the
palladium-catalyzed cross-coupling between benzylic zinc bromides and
alkyl iodide/alkenyl trifates, respectively. See: Piber, M.; Jensen, A.
E.; Rottlaender, M.; Knochel, P. Org. Lett. 1999, 1, 1323.
(
18) The possibility of aldol condensation under the ionic liquid/zinc
halide conditions was evident in the attempted reaction of (MeO)
COMe. Only the aldol adduct was isolated in good yield.
2
CH-
10436 J. Org. Chem., Vol. 70, No. 25, 2005

Alkylzinc Reagents in Room-Temperature Ionic Liquids
TABLE 5. Zinc-Mediated Cross-Coupling of n-Butyl
TABLE 6. Recycling of [bpy][BF4] in the Ethylation of
Benzaldehydea
Iodide with Benzaldehyde in RTILa
yield of 3a in each cycle (%)c
mole ratio [bpy][X] added
[
bpy][X] added
mole ratio
entry
Zn/EtI
(% X)
1
2
3
4
5
entry
(% X)
X
Zn/n-BuI
ratio 3b/5
b
b
1
2:2
3:3
2:2
Cl (8.5)
Cl (8.5)
Br (8.5)
87
94
94
80
95
67
90
97
89
85
93
75
81
94
80
1
2
3
4
5
6
7
8
9
(0)
Br
Br
Br
Br
Br
Br
Br
Br
Cl
Cl
2:2
2:2
2:2
2:2
2:2
2:2
2:2
2:2
2:2
2:2
2
b
c
b
impure (5.1)
0.1 g (8.5)
0.13 g (11.1)
0.15 g (12.8)
0.17 g (14.5)
0.2 g (17.1)
0.17 g (14.5)
0.08 g (8.5)
0.17 g (18.4)
3
49:51
a
Reaction conditions: Zn and EtI in the indicated mmol ratios
were stirred in RTIL (1 mL) for 1 h at 50 °C. The carbonyl
compound (1 mmol) was added to the reaction mixture and stirred
for 12 h at 50 °C. Sodium oxalate was used to quench the reaction
mixture, followed by precipitation of inorganic salt with acetone.
47:53
54:46
75 (70):25
4d 7:53
b
Entries 1-3: pure [bpy][BF4] was used with the indicated
12 (5 only)
0
c
amount of [bpy][X] added at each cycle. Isolated yields.
1
0
a
Reaction conditions: pure [bpy][BF4] (1 mL) with the indicated
was extracted with organic solvents (usually ether).
There was usually little loss of the ionic liquid in the
extraction process because the extract in each workup
was free of ionic liquid according to its NMR. However,
the recovered ionic liquid may contain inorganic products
insoluble in the organic solvent extract. In the present
reaction, it was the presumed formation of ZnXOH as
the inorganic product which may remain in the recovered
ionic liquid and interfere with subsequent alkylation
amount of [bpy][Br] added was used, except in entry 2, as the
reaction media. Zinc and n-butyl iodide (in mmol specified) were
added and stirred at 50 °C for 1 h. Benzaldehyde (1 mmol) was
added to the reaction mixture and stirred for 12 h at room
temperature. The mixture was worked up, and the ratio of 3b/5
1
b
was determined by H NMR. Benzaldehyde was recovered.
c
d
Pinacol product 4 was obtained with undetermined yield. Tri-
cyclohexylphosphine (3 mmol) was added after the formation of
n-BuZnX.
4
reactions. Indeed, when the ionic liquid [bpy][BF ],
3
). The formation of 5 was attributed to the well-known
recovered after extraction of the organic product and
diluted with a small amount of methylene chloride to
precipitate the inorganic zinc salt, was reused for the
alkylzinc formation, the reaction failed. Apparently, the
inorganic zinc salt was not completely removed by this
recovery process. On the other hand, when the reaction
mixture was quenched with anhydrous sodium oxalate
instead of water and the resultant inorganic salts (pre-
sumably zinc oxalate and sodium halide) were precipi-
tated from the ionic liquid by the addition of a small
propensity of â-hydride reduction of aldehydes by alky-
lzinc. However, when we increased the amount of the
bromide ion even further by adding more [bpy][Br] to the
pure [bpy][BF
4
] as the reaction media, the alkylation
product 3b increased substantially with 5 suppressed
-
(
entries 4-7). The optimal amount of Br concentration
appeared to be 14.5%, at which point the ratio of 3b to 5
was found to be 75:25 and the alcohol 3b could be isolated
in 70% yield after chromatography. Recently, Fu et al.
have found that some hindered phosphines are quite
4
amount of acetone, the IL [bpy][BF ] thus recovered could
useful in suppressing â-hydride elimination for alkyl-
metallic reagents that possess â-hydrogen.19 We thus
then be reused for the reaction to give the ethylated
product 3a in similar yields. The results are summarized
in Table 6. The ionic liquid, added with either [bpy][Br]
attempted the addition of tricyclohexylphosphine to the
-
ionic liquid reaction using the optimal Br concentration.
(
entry 3) or [bpy][Cl] (entries 1 and 2) could be recycled
The result was disappointing, and all the starting ben-
zaldehyde was recovered (entry 8). Addition of [bpy][Cl]
instead gave little of 3b with a poor yield of 5 as the
product (entry 9). Unlike [bpy][Br], further increasing
and reused readily to give the alkylation product with
similar yields for at least five cycles (Table 6). In the case
of entry 2, where a greater excess of Zn and EtI were
used, there was no noticeable change in yields throughout
the five cycles and only a slight reduction in the yield of
[
bpy][Cl] did not improve the yield of 3b.
We have also examined the reaction of phenyl iodide
3
a, by a few percent in each cycle, was observed in the
4
with zinc in [bpy][BF ] under various reaction conditions.
other cases (entries 1 and 3).
An organometallic species appeared to have been formed.
However, on quenching the reaction mixture with ben-
zaldehyde, a complex mixture was invariably obtained
as the reaction product. Further studies will be required
to clarify the reactions involved in this case.
D. Recovery and Reuse of IL. One problem antici-
pated in the use of ionic liquids as solvents for organo-
metallic reactions is the issue of recycling and reusing
the ionic liquids after removal of the organic product.
After the formation of the alkylzinc intermediate and its
subsequent reaction with aldehyde, the reaction mixture
was usually quenched with water and the organic product
E. Mechanistic Discussions. We have shown that
alkylzinc reagents can be formed from alkyl iodide and
zinc metal in the ionic liquid [bpy][BF ]. The bromide or
4
chloride ion was found to be essential for the reaction to
occur. The function of the bromide ion (or chloride ion)
was not due to the possible conversion of the alkyl iodide
to the alkyl bromide under the reaction conditions. The
use of ethyl bromide in place of ethyl iodide in pure [bpy]-
[
4
BF ] failed to lead to the formation of the ethylzinc
intermediate. The bromide (or chloride) ion appeared to
have a role in assisting the zinc metal to dissolve in the
ionic liquid, forming a blue colored solution. This facili-
tates the formation of the alkylzinc intermediate subse-
quently. Scheme 3 presents a plausible mechanism to
(
19) Netherton, M. R.; Dai, C.; Neuschutz, K.; Fu, G. C. J. Am. Chem.
Soc. 2001, 123, 10099.
J. Org. Chem, Vol. 70, No. 25, 2005 10437

Law et al.
SCHEME 3
3 mL of purified and predried [bpy][BF
4
] in a glass electro-
chemical cell and performed by scanning from -0.2 to +1.4
4
V. The background of pure [bpy][BF ] did not show any signal
corresponding to the bromide anion. A stock solution of the
bromide ion was prepared by dissolving a weighed amount of
pure N-butylpyridinium bromide into pure [bpy][BF
Calibration was done by addition of the prepared stock solution
0.1 mL) to a solution of purified [bpy][BF ] (3 mL) and by
repetition a few times between consecutive CV runs. Figure 1
in Supporting Information) shows the cyclic voltammograms
for the oxidation of bromide (0.01-0.03 M) in [bpy][BF ] at a
4
] (3 mL).
(
4
(
4
Pt electrode. A wave was observed at E ) 700 mV vs platinum.
An anodic wave observed on the forward potential sweep is
attributable to the oxidation of bromide to bromine, and the
2
0
cathodic wave is due to the reverse process. A calibration
curve was obtained by the linear plot of current against
bromide ion concentration (Figure 2 in Supporting Informa-
tion). The unknown bromide concentration of the crude [bpy]-
account for the observations. The bromide (or chloride)
ion reacts with the zinc metal to generate ZnX (X ) Br
+
or Cl) and the solvated electron in ionic liquid, giving the
blue color typical of solvated electrons. The solvated
electron then reacts with alkyl iodide to generate a
4
[BF ] was then determined using the calibration curve.
General Procedure for the Preparation of Alkylzinc
+
Reagents in Ionic Liquid. To a 50 mL reaction vessel
radical anion intermediate, which can react with ZnX
4
containing the colorless ionic liquid [bpy][BF ] (1 mL) (Figure
to give RZnX (X ) Br or Cl). The RZnX then reacts with
aldehyde to give both the alkylation product 3 and/or the
hydride reduction product 5. In the n-BuZnX case, where
â-hydride transfer to aldehydes normally predominates,
the presence of more bromide ion may lead to the
3
in Supporting Information) was added metallic zinc (2 mmol).
Commercially available zinc dust (purity 99.9, -325 mesh),
zinc powder (purity 99.999, -40 mesh or purity 99.999, -100
+200 mesh), and zinc granules (purity 99.8+, -10 +50 mesh)
could all be used without any pretreatment. Zinc metal
powder, which had been stored for a long time, could be
-
formation of complex RZnXBr which may facilitate the
2
1
washed with dilute acid before use. For ionic liquid, which
contained between 5% and 12.8% bromide ion, the solution
alkyl transfer more than the â-hydride transfer, thus
accounting for the greater amount of 3.
slowly turned blue (Figure 4 in Supporting Information) after
1
stirring at room temperature for /
2
h. Alkyl iodide (2 mmol)
Conclusion
was then added, and the solution was stirred at 50 °C
overnight and then turned green (Figure 5 in Supporting
Information), indicating the formation of the alkylzinc inter-
We have shown that alkylzinc intermediates can be
formed directly from alkyl iodide and zinc metal in the
1
mediate. H NMR was then examined by adding pyridine (3
ionic liquid [bpy][BF
4
]. However, the presence of [bpy]-
mmol) to the reaction mixture followed by dissolving a drop
3
of the ionic liquid in d -acetonitrile in an NMR tube. One set
[
X] either as an impurity or as an additive was required
δ
of ethyl group absorption (one triplet at CH
3
) 1.18 ppm and
for the reaction to occur. The role of the X ion is believed
to be to facilitate the dissolution of zinc metal in the IL
to give the solvated electron to assist in the subsequent
reaction with the alkyl iodide. The structure of the
alkylzinc intermediate is likely to be RZnX according to
δ
one quartet at CH
8.
2
) 0.28 ppm) was observed with JAB (Hz)
)
General Procedure for the Alkylation of Aldehydes
with Organozinc Reagents in Ionic Liquid. A 50 mL
reaction vessel was charged with the ionic liquid (1 mL,
1
H NMR. The alkylzinc intermediate thus formed reacts
4
bromide containing [bpy][BF ]), powdered metallic zinc (2
efficiently with aryl aldehydes to give the corresponding
ethylated products. The IL can be recovered and reused
for at least five cycles.
mmol), and alkyl iodide (2 mmol) under an inert nitrogen
atmosphere. The reaction mixture was stirred at 50 °C for 1 h
followed by the addition of the carbonyl compound (1 mmol).
The mixture was stirred at the indicated temperature for the
time period (see Tables 2 and 3) under an inert nitrogen
atmosphere (Figure 6 in Supporting Information). The result-
Experimental Section
ing mixture was quenched with a few drops of water followed
Preparation of Ionic Liquids. The crude N-butylpyri-
dinium tetrafluoroborate [bpy][BF ] was prepared by a me-
4
tathesis reaction in acetonitrile (300 mL) from sodium tet-
rafluoroborate (121 g, 1.1 mol) (purchased commercially with
22
by extraction with Et
2
O (3 × 10 mL). After removal of the
ether solvent in vacuo, the residue was purified by flash
chromatography on silica gel to yield the pure adduct alcohol
3
. Specific conditions and yields of the alkylation of aldehydes
9
8% purity in fine powder form) and its bromide precursor
are given in Tables 2-4. Compounds 3a, 3b, and all the
[
bpy][Br] (216 g, 1.0 mol) which was in turn prepared from
isolated products in Table 3 are known compounds. Their
11
the microwave-assisted method. The mixture was stirred for
days at room temperature and then filtered. The filtrate was
evaporated, and the residue crude [bpy][BF ] was vacuum-
dried overnight at 70 °C (0.1 mmHg) and stored under
nitrogen. Pure [bpy][BF ] could be obtained by column chro-
matography of the crude [bpy][BF ] on silica gel with dichlo-
1
structures and purities were confirmed by their H NMR
3
spectra (Figures 7-13 in Supporting Information) and com-
4
parison with known samples previously synthesized.
9
Recycling and Reusing Ionic Liquids. After complete
4
reaction, an equivalent amount of sodium oxalate to zinc metal
4
1
2
romethane as eluent. The recovery of [bpy][BF
matography was about 60%.
4
] after chro-
(
20) Allen, G. D.; Buzzeo, M. C.; Villagran, C.; Hardacre, C.;
Compton, R. G. J. Electroanal. Chem. 2005, 575 (2), 311.
(21) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory
Chemicals; Oxford: Boston; Butterworth: Heinemann, 1996.
Electrochemical Determination of Bromide Ion Con-
centration in [bpy][BF ]. A conventional three-electrode
4
(22) It is assumed generally that, if necessary, extraction of organic
setup was utilized with all Pt electrodes acting as working,
counter-, and pseudoreference electrodes. The platinum work-
ing electrode was polished with a slurry of alumina suspension
in water on a clean cloth before use. Cyclic voltammetry was
carried out on a BAS 100B/W electrochemical workstation with
substrates from ionic liquids can be carried out with supercritical
carbon dioxide in place of nonpolar organic solvents. See: (a) Blan-
chard, L. A.; Hancu, D.; Beckman, E. J.; Brennecke, J. F. Nature 1999,
3
99, 28. (b) Brown, R. A.; Pollet, P.; McKoon, E.; Eckert, C. A.; Liotta,
C. L.; Jessop, P. G. J. Am. Chem. Soc. 2001, 123, 1254.
10438 J. Org. Chem., Vol. 70, No. 25, 2005

Alkylzinc Reagents in Room-Temperature Ionic Liquids
powder was added to the reaction mixture. The zinc residue
could be removed by dissolution of the ionic liquid in acetone
lished under the University Grants Committee of the
Hong Kong Special Administrative Region, China (Project
No. AoE/P-10/01). The award of a scholarship by McGill
University to M.C.L. is gratefully acknowledged. We
thank Michael C. Y. Tang for the electrochemical
determination of bromide ion concentration.
(
5 mL) or methylene chloride, and the solid precipitates of Zn-
) and NaX were filtered. The ionic liquid was recovered
(
2 4
C O
on removal of acetone/methylene chloride followed by extrac-
tion with Et
2
O (2 × 5 mL) and dried by further heating
overnight at 70 °C (0.1 mmHg). The experimental procedure
for ethylation in recycled ionic liquid was the same as that
described above.
Supporting Information Available: Cyclic voltammo-
1
grams and H NMR spectra of 3a, 3b, and other isolated
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. This work was supported by a
grant from NSERC to T.H.C., a HKRGC grant (PolyU
5
002/03P), and the Areas of Excellence Scheme estab-
JO051772M
J. Org. Chem, Vol. 70, No. 25, 2005 10439