SCHEME 3. Alkylation Route to Imidazolium Salt 3
TABLE 1. Grignard Reactions in RTIL 5a
A more satisfactory, though longer, route employed standard
imidazole alkylation chemistry (Scheme 3). Thus, alkylation of
2-isopropylimidazole with n-butyl bromide in the presence of
sodium hydroxide under phase-transfer conditions afforded
9
imidazole 4 in 98% yield. A second alkylation with methyl
iodide in refluxing methylene chloride then afforded the desired
iodide salt in 92% yield. Using this strategy, 10-20 g scale
reactions could be easily carried out with good reproducibility.
With the desired iodide salt in hand, standard anion metathesis
afforded the triflimide salt 5 as a very pale yellow liquid in
essentially quantitative yield. Qualitatively, this RTIL is slightly
more viscous than BMIM NTf2 but still readily poured and
transferred via pipet.
At this point, the use of RTIL 5 in main group organometallic
chemistry could be studied. There have been very few reports
of main group organometallic chemistry in RTILs. Most have
dealt with the use and/or generation of organozinc reagents in
imidazolium or pyridinium RTILs.10 These species are generally
less basic and less reactive, which may account for the greater
number of successful reports of their use in RTILs. Beyond
these reports, there have been only two reports of the use of
main group organometallic reagents in RTILs.1 In both cases,
Grignard reagents were employed. Clyburne and co-workers
reported that Grignard reagents could be successfully used in
tetraalkylphosphonium RTILs; the products could be separated,
a
Reaction conditions: 0.5 mmol of the carbonyl compound in 2 mL of
RTIL 5 and 0.55 mmol of the Grignard reagent. b Isolated yield. Identity
confirmed by comparison to commercially available material.
c
the anticipated addition products in good yield (Table 1, entries
and 2). A similar reaction with phenylmagnesium bromide
also afforded the anticipated addition product in good yield
Table 1, entry 3).
1
(
1,12
For the product extraction stage, pure ether could also be
employed. However, RTIL 5 does exhibit slight solubility in
ether, meaning that, over the course of several recyclings, the
volume of the RTIL remaining will decrease. On the other hand,
the use of the 10% ether in hexane solution was sufficient to
completely extract the products from the RTIL because NMR
analysis of the recovered RTIL layer showed no traces of either
the addition products or the starting aldehyde. At the same time,
NMR analysis of the concentrated crude organic extract layer
showed no trace of RTIL 5, indicating that the use of 10% ether
in hexanes avoids leaching of the RTIL during product
extraction.
For the recycling of the RTIL layer, two options were
explored. In one case, the aqueous layer was simply removed
following the extraction of the organic products. The remaining
RTIL was then dried under vacuum for 36-48 h prior to use
in the next Grignard addition. Although this option may leave
some magnesium salts in the RTIL layer, it worked satisfactorily.
1
1
and the RTIL could be reused in a relatively simple fashion.
More recently, Wilhelm and co-workers have reported a new
class of RTILs (2-phenylimidazolinium salts) that are also
compatible with Grignard reagents.12 Again, the products of the
addition of these reagents to carbonyl compounds could be
readily separated from the RTIL and the RTIL could be recycled
using simple extraction methods.
Our own efforts also targeted the addition of Grignard
reagents to aldehydes. The first aldehyde that was employed
was p-anisaldehyde. The reaction of this aldehyde with both
methylmagnesium chloride and vinylmagnesium bromide was
conducted in RTIL 5. The Grignard reagents were both used as
solutions in THF as supplied by commercial sources. Because
of the nonvolatile nature of the RTIL, no external cooling was
applied, although subsequent efforts have shown that the
reactions can also be carried out using an ice bath with no
solidification of the RTIL. The reaction vessels did become
slightly warm to the touch after addition of the Grignard reagent.
After 3 h, the reactions were quenched by the addition of dilute
The other option was to employ a method akin to that reported
by both Clyburne and Wilhelm.
11,12
In this case, after extraction
of the organic products, the RTIL layer was diluted with
methylene chloride, the aqueous layer was separated, and the
RTIL layer was washed with water. This organic layer was then
dried with magnesium sulfate, concentrated, and dried under
vacuum overnight. Because there did not appear to be any
significant difference between the two recycling methods, both
were used interchangeably during the following studies. It should
also be noted that the same two ionic liquid samples that were
used in the first two reactions (Table 1, entries 1 and 2) were
recycled repeatedly and used as the solvent for all of the
(0.1 N) hydrochloric acid. The organic products were then
extracted using 10% ether in hexanes to afford, after purification,
(9) Khabnadideh, S.; Rezaei, Z.; Khalafi-Nezhad, A.; Bahrinajafi, R.;
Mohamadi, R.; Farrokhroz, A. A. Bioorg. Med. Chem. Lett. 2003, 13, 2863-
2
865.
(10) Law, M. C.; Wong, K.-Y.; Chan, T. H. J. Org. Chem. 2005, 70,
1
6
2
0434-10439. Law, M. C.; Wong, K.-Y.; Chan, T. H. Green Chem. 2004,
, 241-242. Sirieix, J.; Ossberger, M.; Betzemeier, B.; Knockel, P. Synlett
000, 1613-1615.
16
experiments reported in this paper.
(
11) Ramnial, T.; Ino, D. D.; Clyburne, J. A. C. Chem. Commun. 2005,
3
25-327.
(13) Chang, C.; Kumar, M. P.; Liu, R. J. Org. Chem. 2004, 69, 2793-
2796.
(
12) Jurcik, V.; Wilhelm, R. Green Chem. 2005, 7, 844-848.
4
660 J. Org. Chem., Vol. 71, No. 12, 2006