Kemp elimination: A probe reaction to study ionic liquids properties

Article abstract of DOI:10.1021/jo702662z

(Chemical Equation Presented) The amino induced elimination of benzisoxazole into the relevant o-cyanophenolate ion (Kemp elimination) has been studied in [bmim] [BF₄] solution at 298 K. To have information about the interactions between reactants and ionic liquid, the reaction has been carried out at different temperatures (293-313 K). Several primary, secondary, and tertiary amines have been used to study the effect of amine structure on the reaction rate. The collected data show that the amine structure seems to have a crucial role in determining the reaction rate. Furthermore, as different cation or anion structures of an ionic liquid can significantly affect its properties, the title reaction has been performed in four different ionic liquids ([bmim][PF₆], [bmim][NTf₂], [bm₂im][NTf ₂], and [bmpyrr][NTf₂]), using pyrrolidine and piperidine as model amines. An H-donor negative solvent (MeOH and [bmim][NTf₂]) effect on reaction rate was detected. Finally, a narrow range of activation parameters was calculated both for the reaction induced by different amines and for pyrrolidine and piperidine, in the presence of different ILs. This fact suggests the occurrence of an early transition state.

Full text of DOI:10.1021/jo702662z

Kemp Elimination: A Probe Reaction To Study Ionic Liquids
Properties
Francesca D’Anna,* Sandra La Marca, and Renato Noto*
Dipartimento di Chimica Organica “E. Paternò”, UniVersità degli Studi di Palermo, Viale delle Scienze,
Parco d’Orleans II, 90128 Palermo, Italy
fdanna@unipa.it; rnoto@unipa.it
ReceiVed December 14, 2007
The amino induced elimination of benzisoxazole into the relevant o-cyanophenolate ion (Kemp elimination)
has been studied in [bmim][BF ] solution at 298 K. To have information about the interactions between
4
reactants and ionic liquid, the reaction has been carried out at different temperatures (293–313 K). Several
primary, secondary, and tertiary amines have been used to study the effect of amine structure on the
reaction rate. The collected data show that the amine structure seems to have a crucial role in determining
the reaction rate. Furthermore, as different cation or anion structures of an ionic liquid can significantly
affect its properties, the title reaction has been performed in four different ionic liquids ([bmim][PF
bmim][NTf ], [bm im][NTf ], and [bmpyrr][NTf ]), using pyrrolidine and piperidine as model amines.
An H-donor negative solvent (MeOH and [bmim][NTf ]) effect on reaction rate was detected. Finally, a
6
],
[
2
2
2
2
2
narrow range of activation parameters was calculated both for the reaction induced by different amines
and for pyrrolidine and piperidine, in the presence of different ILs. This fact suggests the occurrence of
an “early” transition state.
Introduction
been recently described as polymeric hydrogen bonded su-
pramolecules whose order degree can be differently affected
by guest reactants molecules. Furthermore, as has been pointed
Organic chemists are very interested in the development of
environmental friendly catalysts and solvents. Among solvents
4
out recently, ILs are able to give π-π and π-cation interactions
much interest is focused on room temperature ionic liquids
that in some cases may significantly affect the reactivity in these
1
(
RTILs). This is due to their low vapor pressure and negligible
5
solvent media.
flammability, which make ionic liquids (ILs) attractive candi-
dates as replacements for volatile organic solvents.
2
(1) (a) Welton, T. Chem. ReV. 1999, 99, 2071–2083. (b) Wasserscheid, P.;
Keim, M. Angew. Chem., Int. Ed. Engl. 2000, 39, 3772–3789. (c) Rogers, R. D.;
Seddon, K. R. Ionic Liquids: Industrial Application to Green Chemistry; ACS
Symp. Ser. No. 818; American Chemical Society: Washington, DC, 2002. (d)
Rogers, R. D.; Seddon, K. R. Ionic Liquids as Green SolVents. Progress and
Prospects; ACS Symp. Ser. No. 856; American Chemical Society: Washington,
DC, 2003. (e) Rogers, R. D.; Seddon, K. R.; Volker, S. Green Industrial
Application of Ionic Liquids, NATO Science Series II. Mathematics, Physics and
Chemistry; Kluwer: Dordrecht, The Netherlands, 2003; p 92. (f) Ionic Liquids
in Synthesis; Wasserscheid, P., Welton, T., Eds.; Wiley-VCH: Weinheim,
Germany, 2008. (g) Chiappe, C.; Pieraccini, D. J. Phys. Org. Chem. 2005, 18,
275–297. (h) Rogers, R. D.; Seddon, K. R. Ionic Liquids III A: Fundamentals,
Progress, Challenges and Opportunities; ACS Symp. Ser. No. 901; American
Chemical Society: Washington, DC, 2005. (i) Harper, J. B.; Kobrak, M. N. Mini-
ReV. Org. Chem. 2006, 3, 253–269. (j) Chowdhury, S.; Mohan, R. S.; Scott,
J. L. Tetrahedron 2007, 63, 2363–2389.
The nature of solvent is frequently one of the most important
factors in determining the outcome of a given reaction and the
study of solvent properties becomes, under this light, an area
of considerable importance. Studying ILs, this subject can
assume a particular perspective. Indeed, being formed only by
ions, ILs can produce different electrostatic environments, which
3
can significantly affect a given reaction. On the other hand,
ILs formed by aromatic cations, such as imidazolium ions, have
*
Corresponding author. Phone: +39091596919. Fax: +39091596825 (F.D.
and R.N.).
1
0.1021/jo702662z CCC: $40.75  2008 American Chemical Society
J. Org. Chem. 2008, 73, 3397–3403 3397
Published on Web 03/21/2008

D’Anna et al.
CHART 1. Schematic Representation of Substrates and
Among the methods used to achieve a better understanding
of the properties of a given solvent, the kinetic study of a probe
reaction, having a well-known mechanism, may represent a
useful tool. Therefore, we believed it would be interesting to
investigate kinetically the effect that ILs have on a classical
organic reaction such as an elimination reaction. We have
previously studied the effect of ILs on the amine-induced
elimination of 1,1,1-tribromo-2,2-bis(phenyl-substituted)ethane
Reaction Studied
5
b
(
1). The reaction studied in this work is the base-induced
elimination of benzisoxazole (2) into the relevant o-cyanophe-
nolate (3), the so-called Kemp elimination, for which a concerted
E2 elimination mechanism has been accepted. This reaction has
been previously studied in water, using hydroxide or trimethy-
6
7
lamine as bases, as well as in the presence of cationic vesicles,
The reaction was carried out in 1-butyl-3-methylimidazolium
8
9
biocatalysts (enzymes and antibodies), and cyclodextrins. The
Kemp elimination proceeds through a charge separated and
ordered transition state, in which both C-H and N-O bonds
are extensively cleaved. For all the above reasons, according
to our previous findings,
tetrafluoroborate solution ([bmim][BF ]) at various amines
4
concentrations (0.00870–0.022 M) over the temperature range
293–313 K.
6
It is well-known that differences in both the cation and the
anion may have significant effects on RTILs structure and
properties. Under this light, we chose pyrrolidine and piperidine
as model amines and we studied the title reaction in different
ILs, such as [bmim][PF6], [bmim][NTf2], [bm2im][NTf2], and
5
b–d
it should be a suitable probe to
study ionic liquids properties (Chart 1).
The reaction rates were measured spectrophotometrically at
98 K, following the appearance of 3 at 346 nm. We chose as
bases some primary, secondary, and tertiary amines with
different structures (cyclic or acyclic), basicities, and steric
requirements that are widely soluble in the used ionic liquids
2
[
bmpyrr][NTf2] [where bm2im ) 1-butyl-2,3-dimethylimida-
zolium, bmpyrr ) N-butyl-N-methylpyrrolidinium, NTf2 )
bis(trifluoromethylsulfonylimide)] (Chart 3).
+
+
Among these, bm2im and bmpyrr have a lesser hydrogen
(
Chart 2).
+
+
bond donor ability than bmim ; furthermore, the bmpyrr is
not able to give π-π interactions that in some cases determine
a strong effect on reactivity. At last, the substitution of the anion
part of [bmim][X] going from [BF4 ] and [PF6 ] to [NTf2 ]
should cause a different packing and possibly a different
catalytic effect.
(
2) (a) Seddon, K. R.; Stark, A.; Torres, M.-J. Pure Appl. Chem. 2000, 72,
-
-
-
2
275–2287. (b) Rodgers, R. D.; Seddon, K. R. Science 2003, 302, 792–793. (c)
Forsyth, S. A.; Pringle, J. M.; MacFarlane, D. R. Aust. J. Chem. 2004, 57, 113–
1
19.
(
3) (a) Sheldon, R. Chem. Commun. 2001, 2399–2407. (b) Lancaster, N. L.;
Welton, T.; Young, G. B. J. Chem. Soc., Perkin Trans. 2 2001, 2267–2270. (c)
Chiappe, C.; Conte, V.; Pieraccini, D. Eur. J. Org. Chem. 2002, 2831–2837. (d)
Lancaster, N. L.; Salter, P. A.; Welton, T.; Young, G. B. J. Org. Chem. 2002,
To have a comparison with reactivity in conventional organic
solvents, the reaction was also studied in MeOH and DMF
solutions.
67, 8855–8861. (e) Chiappe, C.; Pieraccini, D.; Saullo, P. J. Org. Chem. 2003,
68, 6710–6715. (f) Chiappe, C.; Pieraccini, D. J. Org. Chem. 2004, 69, 6059–
6064. (g) Akiyama, T.; Suzuki, A.; Fuchibe, K. Synlett 2005, 1024, 1026. (h)
Results and Discussion
Ranu, B. C.; Jana, R. J. Org. Chem. 2005, 70, 8621–8624. (i) Conte, V.; Floris,
B.; Galloni, P.; Mirruzzo, V.; Scarso, A.; Sordi, D.; Strukul, G. Green Chem.
2
005, 7, 262–266. (j) Laali, K. K.; Sarca, V. D.; Okazaki, T.; Brock, A.; Der, P.
First, we checked for the possible occurrence of the spontane-
ous reaction in IL solution. We verified that the substrate remains
unchanged, in IL solution, for at least 48 h. This suggests that
Org. Biomol. Chem. 2005, 3, 1034–1042. (k) Man, B. Y. W.; Hook, J. M.; Harper,
J. B. Tetrahedron Lett. 2005, 46, 7641–7645. (l) Lindén, A. A.; Johansson, M.;
Hermanns, N.; Bäckvall, J.-E. J. Org. Chem. 2006, 71, 3849–3853. (m) Zhao,
X.; Alper, H.; Yu, Z. J. Org. Chem. 2006, 71, 3988–3990. (n) Chiappe, C.;
Piccioli, P.; Pieraccini, D. Green Chem. 2006, 8, 277–281. (o) Yoshino, T.; Imori,
S.; Togo, H. Tetrahedron 2006, 62, 1309–1317.
uncatalyzed reaction did not proceed at an appreciable rate (k
-5 -1
,
1.2 × 10 s ). As substrate and amines were added to IL
10
as concentrated solution in 1,4-dioxane (see the Experimental
Section), we checked for the reaction in this solvent. However,
also in this case, in the presence of the highest pyrrolidine (the
most efficient base catalyst) concentration used in this work,
the reaction did not proceed.
(
4) (a) Dupont, J. J. Braz. Chem. 2004, 15, 341–350. (b) Consorti, C. S.;
Suarez, P. A.; de Souza, R. F.; Burrow, R. A.; Farrar, D. H.; Lough, A. J.; Loh,
W.; da Silva, L. H. M.; Dupont, J. J. Phys. Chem. B 2005, 109, 4341–4349. (c)
Dupont, J.; Suarez, P. A. Z. Phys. Chem. Chem. Phys. 2006, 8, 2441–2452.
(
5) (a) Iwata, K.; Kakita, M.; Hamaguchi, H. J. Phys. Chem. B 2007, 111,
4
6
914–4919. (b) D’Anna, F.; Frenna, V.; Pace, V.; Noto, R. Tetrahedron 2006,
2, 1690–1698. (c) D’Anna, F.; Frenna, V.; Noto, R.; Pace, V.; Spinelli, D. J.
In IL solution, we did not have any evidence about the
formation of intermediates and we confidently supposed that
the mechanism of the base-catalyzed elimination of the ben-
zisoxazole proceeded, as well as in water solution, through a
Org. Chem. 2006, 71, 9637–9642. (d) D’Anna, F.; Frenna, V.; La Marca, S.;
Noto, R.; Pace, V.; Spinelli, D. Tetrahedron 2008, 64, 672–680.
(
6) (a) Casey, M. L.; Kemp, D. S.; Paul, K. G.; Cox, D. D. J. Org. Chem.
1
973, 38, 2294–2301. (b) Kemp, D. S.; Cox, D. D.; Paul, K. G. J. Am. Chem.
Soc. 1975, 97, 7312–7318.
7) (a) Peréz-Juste, J.; Hollfelder, F. Org. Lett. 2000, 2, 127–130. (b) Klijn,
6
(
concerted E2 elimination.
J. E.; Engberts, J. B. F. N. J. Am. Chem. Soc. 2003, 125, 1825–1833. (c) Klijn,
J. E.; Engberts, J. B. F. N. Org. Biomol. Chem. 2004, 2, 1789–1799.
It is noteworthy that a significant blue-shift in the UV–vis
spectrum of o-cyanophenolate (3) was detected in [bmim][BF4]
solution with respect to MeOH (λMAX ) 327 and 346 nm in
MeOH and [bmim][BF4], respectively).
Kinetic Data in [bmim][BF4]. Among amines used, the most
basic pyrrolidine was the most effective, whereas diisoprop-
(
8) (a) Thorn, K. K. S. N.; Hilvert, D. J. Am. Chem. Soc. 1996, 118, 8184–
8
1
185. (b) Hollfelder, F.; Kirby, A. J.; Tawfik, D. S. J. Am. Chem. Soc. 1997,
19, 9578–9579. (c) Hollfelder, F.; Kirby, A. J.; Tawfik, D. S.; Kikuchi, K.;
Hilvert, D. J. Am. Chem. Soc. 2000, 122, 1022–1029. (d) Hotta, K.; Kikuchi,
K.; Hilvert, D. HelV. Chim. Acta 2000, 83, 2183–2191. (e) Hollfelder, F.; Kirby,
A. J.; Tawfik, D. S. J. Org. Chem. 2000, 66, 5866–5874. (f) Lemma-Gray, P.
Biotechnol. Bioeng. 2001, 74, 524–527. (g) Hu, Y.; Houk, K. N.; Kikuchi, K.;
Hotta, K.; Hilvert, D. J. Am. Chem. Soc. 2004, 126, 8197–8205. (h) Seebeck,
F. P.; Hilvert, D. J. Am. Chem. Soc. 2005, 127, 1307–1312.
(10) The kinetic solutions contain 13% (v:v) of 1,4-dioxane. According to
5
d
(9) McCraken, P. G.; Ferguson, C. G.; Vizitiu, D.; Walkinshaw, C. S.; Wang,
our previous report, however, the obtained results can be well referred to neat
ionic liquid, i.e., to a partially organized solvent medium.
Y.; Thatcher, G. R. J. J. Chem. Soc., Perkin Trans. 2 1999, 911, 912.
3
398 J. Org. Chem. Vol. 73, No. 9, 2008

Kemp Elimination
CHART 2. List of Amines Used
CHART 3. Structures of Ionic Liquids
For a quick overall evaluation purpose, the reactivity ranges
(
∆k) as a function of amines nature are reported in Figure 1.
Data reported in Figure 1 show that the reactivity range increases
along the series as follows: primary amines < tertiary amines
<
secondary amines. We recently found the same order,
studying the ion pair stability of 4-nitrophenol/amine in IL
1
2
solution. Furthermore, as far as secondary amines are con-
cerned, ∆k values increase on going from acyclic to cyclic
amines. A comparison among reactivity and basicity ranges
indicates that this latter factor is not the only one involved in
affecting the reaction rate.
TABLE 1. Calculated Second-Order Rate Constants at 298 K for
the Elimination Reaction of 2, in [bmim][BF4] Solution, in the
Presence of Amines
2
-1
pKBH+
a
amine
10 kII (M s-1
)
R
+
Indeed, ∆pKBH is larger for tertiary amines than for
butylamine
0.569 ( 0.049
0.408 ( 0.027
0.157 ( 0.031
2.30 ( 0.08
2.62 ( 0.14
10.9 ( 0.4
3.76 ( 0.10
5.99 ( 0.11
3.99 ( 0.24
n.r.
0.982
0.989
0.999
0.997
0.992
0.996
0.993
0.999
0.991
10.68
10.66
+
secondary ones (∆pKBH ) 0.49 and 0.67 for secondary and
cyclohexylamine
diisopropylamine
dibutylamine
N-butyl-N-methylamine
pyrrolidine
tertiary amines, respectively), but ∆k results are larger for the
-
1
-1
11.20
latter ones (∆k ) 0.107 and 0.045 M
tertiary amines, respectively).
s
for secondary and
11.27
11.12
10.89
10.78
8.33
A similar result can be obtained considering amines of the
same class. As a matter of fact, in the presence of primary
amines, cyclohexylamine and butylamine have comparable
piperidine
hexamethyleneimine
heptamethyleneimine
morpholine
+
pKBH values, but the reaction in the presence of butylamine is
2
,2,6,6-tetramethylpiperidine
2.40 ( 0.18
3.34 ( 0.10
6.36 ( 0.34
1.86 ( 0.03
n.r.
0.992
0.998
0.993
0.999
11.07
10.75
10.46
10.08
7.41
faster (kbua/kcha ) 1.4). Again, in the presence of secondary
amines, significant differences in reactivity are not due to large
differences in basicity. For example, dibutylamine and pyrro-
triethylamine
N-methylpyrrolidine
N-methylpiperidine
N-methylmorpholine
+
lidine have similar pKBH values (11.20 and 11.27, respectively),
but kpyrr/kdbua ) 4.8. Furthermore, piperidine is a stronger base
a
See ref 11.
+
than hexamethyleneimine (pKBH ) 11.12 and 10.89, respec-
tively), but kpip/kh6mi ) 0.63. Finally, among tertiary amines,
ylamine was least effective. Moreover, morpholine and N-
methylmorpholine were not able to induce the elimination
reaction. This behavior is different from that previously observed
for the E2 elimination on 1,1,1-tribromo-2,2-bis(phenyl-substi-
5
b
tuted)ethane (1). Indeed, for this compound only secondary
cyclic amines were capable of inducing the reaction.
For all reactive amines, the observed rate constant kobs shows
a linear correlation as a function of base concentration, according
to eq 1
kobs ) i + k [Am]
(1)
II
Second-order rate constant values (kII) as a function of amine,
+
11
together with pKBH values collected in aqueous solution, are
reported in Table 1 (Data collected at different amine concentra-
tions are reported in Table 5 of the Supporting Information).
(
11) (a) Frenna, V.; Vivona, N.; Consiglio, G.; Spinelli, D. J. Chem. Soc.,
Perkin Trans. 2 1985, 1865, 1868. (b) Hall, H. R., Jr. J. Am. Chem. Soc. 1957,
9, 5441–5444. (c) Lide, D. R. CRC Handbook of Chemistry and Physics, 80th
ed.; CRC: London, UK, 1999-2000.
FIGURE 1. Plot of reactivity ranges as a function of the nature of the
4
amine relative to the base-catalyzed elimination of 2 in [bmim][BF ]
7
solution.
J. Org. Chem. Vol. 73, No. 9, 2008 3399

D’Anna et al.
triethylamine is a stronger base than N-methylpiperidine (pKBH+
10.75 and 10.46 for triethylamine and N-methylpiperidine,
)
respectively), but the reaction proceeds faster in the presence
of the last amine (kmpip/ktea ) 1.9). However, irrespective of
structure, tertiary amines in [bmim][BF4] are more efficient than
-3
-1 -1
6
in water solution [kII,303K ) (8.0 ( 0.7) × 10
M
s ].
As we previously said, reactivity of secondary amines
increases on going from acyclic to cyclic ones. In particular,
for the three acyclic amines used, kII values increase on going
from diisopropylamine to dibutylamine and N-butyl-N-methyl-
amine. Amines having linear alkyl groups (dbua and bma) react
faster than amines having branched alkyl groups such as
diisopropylamine. In the presence of cyclic amines, reactivity
decreases on going from pyrrolidine to piperidine or to other
larger amines. It is noteworthy that a similar result was
5
b
previously obtained for amino-induced elimination of 1.
However, in this case the effect due to the amine ring size is
more important. In fact, kpyrr/kh7mi ) 22.7 and 2.73 for 1 and 2,
respectively.
FIGURE 2. Plot of absorbance as a function of time relative to the
2
pyrrolidine-catalyzed elimination of 2 in [bmpyrr][NTf ] at 298 K
(
[pyrr] ) 0.0217 M).
The different amine effect on elimination reaction of 2 and
1
could be due to different substrate geometries. The former
substrate has suitable geometry for elimination, whereas 1 should
be constrained in order to assume a planar conformation, which
possibly allows the electronic effects exerted by the aromatic
rings on the reaction route to be maximized. This results in a
larger sensitivity to structural requirements of the amine.
Moreover, the significant role played by cyclic amine structure
emerges comparing the reactivity of N-butyl-N-methylamine,
pyrrolidine, and N-methylpyrrolidine. Indeed, going from bma
to pyrr an increase in reactivity (∼4) is observed. A similar
result is obtained going from the secondary amine (bma),
generally more reactive, to the tertiary amine (mpyrr) (∼2.4).
In the presence of cyclic tertiary amines, reactivity decreases
on going from pyrrolidine to piperidine derivatives. It is
noteworthy that the comparison among pyrrolidine, piperidine,
and corresponding N-methylderivatives shows that the methy-
lation of amine nitrogen has a comparable effect on the
reactivity. Indeed, quite similar reactivity ratios can be calculated
between secondary and corresponding tertiary amines (kpyrr/kmpyrr
FIGURE 3. UV–vis spectra of [bmpyrr][NTf
2
] in the presence of [pyrr]
)
0.0217 M collected in 1 h.
)
1.7 and kpip/kmpip ) 2.0), in spite of more significant
+
differences in basicity (∆pKBH ) 0.81 and 1.04 respectively).
On the whole, collected data show that the amino-catalyzed
elimination of 2 is very sensitive to amine structure (i.e.,
flexibility, steric hindrance, and geometry). In contrast, the base
strength plays a minor role. Similar results were previously
typical trend for a simple kinetic process, with the only exception
being the pyrrolidine-induced elimination in [bmpyrr][NTf2].
In the latter case at least two kinetically prominent steps were
responsible for the observed trend (Figure 2).
In fact, first or higher order kinetic equations do not fit
the experimental trace. From a careful analysis of the system
we found that the absorbance of the IL-pyrrolidine mixture
changed as function of time. Furthermore, a variation in the
UV–vis spectrum of [bmpyrr][NTf2] induced by pyrrolidine was
observed. For example, in Figure 3 the spectra of IL in the
presence of pyrrolidine collected over a time range are reported.
Spectral analysis shows that the absorbance at 300 nm and at
13
reported by Yadav et al. studying the nucleophilic substitution
reaction in the presence of secondary amines. Considering
secondary acyclic amines, the reactivity seems to be a function
of alkyl group structure and of steric hindrance on the amino
nitrogen. The considerable influence of the amine structure on
the reaction rate could be further evidence for the occurrence
of a concerted E2 mechanism, where a crowded transition state
suffers from the effect of base structure.
∼
360 nm first increases as a function of the time and then
decreases to a constant value.
Kinetic Data in Different ILs. We chose pyrrolidine and
piperidine to study the target reaction in different ILs, because
the former amine was the best catalyst, whereas piperidine,
1
The IL-pyrrolidine mixtures were analyzed also by H NMR
1
5
b
spectroscopy. H NMR spectra, recorded at different times, are
according to previous reports, did not induce significant
variations in organized RTIL structure. In all the cases
considered, the absorbance as a function of time showed a
reported in Figure 4 (spectra recorded at t ) 1.5 and 24 h are
available in Figure 5 of the Supporting Information). In all cases
the analyzed solutions were perfectly clear.
As can be seen, after adding 75 µL of 1,4-dioxane all signals
related to the [bmpyrr] cation were enlarged and lose their
multiplicity. The same trend was observed in the presence of a
(
(
12) D’Anna, F.; Noto, R. Tetrahedron 2007, 63, 11681–11685.
13) Yadav, J. S.; Reddy, B. V. S.; Basak, A. K.; Navsaiah, A. V. Tetrahedron
Lett. 2003, 42, 4364–4366.
3
400 J. Org. Chem. Vol. 73, No. 9, 2008

Kemp Elimination
FIGURE 4. 1H NMR spectra of (a) neat [bmpyrr][NTf
2
2 2
], (b) [bmpyrr][NTf ] in the presence of 75 mL of 1,4-dioxane, and (c) [bmpyrr][NTf ] in
the presence of a dioxane solution of [pyrr] ) 0.0217 M at t ) 0 min.
pyrrolidine solution in 1,4-dioxane. However, according to
variations detected in the UV–vis spectra, after 1.5 h, all signals
showed the typical multiplicity that became more defined in
of similar viscosity. So unusual features of ILs can be attributed
to a microheterogeneous environment of the moving particles.
14
To avoid invalidating the kinetic data by variations in the
UV–vis spectrum of the [bmpyrr][NTf2]-pyrrolidine mixture,
the kinetic runs were registered by using as a blank a sample
of IL containing the same amine concentration as the kinetic
run. In this manner, excellent pseudo-first-order curves were
obtained (see the inset of Figure 1).
In Table 2 second-order rate constant values (kII) as a function
of ILs and amines are reported. Furthermore, for a useful
comparison rate constant values related to pyrrolidine- and
piperidine-catalyzed elimination of 2 in MeOH and DMF are
2
4 h.
1
The variation in UV–vis and H NMR spectra cannot be
accounted for by the occurrence of an IL degradation reaction.
Indeed, the IL-pyrrolidine mixture, was extracted with diethyl
ether after some days and the analysis of extract did not show
any presence of either methylamine or other reaction products.
5b
Therefore, according to data previously reported, we ascribed
the variations in UV–vis spectra to a reorganization process of
IL molecules induced by amine. It has been recently claimed
that several experimental methods indicate broadly distributed
dynamics of the ions, which is not typical for molecular solvents
(14) Weingärtner, H. Angew. Chem., Int. Ed. 2008, 47, 654–670.
J. Org. Chem. Vol. 73, No. 9, 2008 3401

D’Anna et al.
TABLE 2. Calculated Second-Order Rate Constants at 298 K for
the Pyrrolidine- and Piperidine-Catalyzed Elimination Reaction of 2,
in Different Solvent Media
TABLE 3. Activation Parameter Values for the Elimination
Reaction of 2, in [bmim][BF4] Solution, in the Presence of Amines
amine
∆H* (kJ/mol)
∆S* (J/(K mol))
1
kII, pip (M-1 s-1
solvent
MeOH
DMF
kII, pyrri (M- s-1
)
R
)
R
butylamine
58 ( 7
51 ( 4
54 ( 1
53 ( 1
51 ( 2
46 ( 1
54 ( 1
49 ( 2
49 ( 3
51 ( 3
52 ( 2
52 ( 3
57 ( 1
-134 ( 21
-156 ( 13
-139 ( 5
-139 ( 3
-146 ( 7
-149 ( 4
-131 ( 4
-146 ( 6
-148 ( 10
-147 ( 9
-140 ( 7
-135 ( 10
-130 ( 5
0.0195 ( 0.001
0.097 ( 0.007
0.109 ( 0.004
0.101 ( 0.007
0.0686 ( 0.0033 0.994 0.0234 ( 0.0007 0.997
0.0756 ( 0.004
0.101 ( 0.004
0.990 0.0124 ( 0.0007 0.993
cyclohexylamine
diisopropylamine
dibutylamine
N-butyl-N-methylamine
pyrrolidine
piperidine
hexamethyleneimine
heptamethyleneimine
0.986 0.0210 ( 0.002
0.996 0.0376 ( 0.001
0.986 0.0267 ( 0.002
0.980
0.993
0.990
[
[
[
[
[
bmim][BF4]
bmim][PF6]
bmim][NTf2]
bm2im][NTf2]
bmpyrr][NTf2]
0.992 0.0303 ( 0.0005 0.999
0.996 0.0389 ( 0.001 0.998
2
,2,6,6-tetramethylpiperidine
also reported (data collected at different amines concentrations
are reported in Table 6 of the Supporting Information).
The collected data show that the elimination process occurs
easier in ILs than in MeOH. Furthermore, the reaction proceeds
faster in DMF than in MeOH, so an H-donor negative solvent
triethylamine
N-methylpyrrolidine
N-methylpiperidine
-
effect could be operating, as well as data for [NTf2 ] ILs (see
interaction between the IL anion and the hydrogen atom on the
ammonium nitrogen. On the other hand, the IL cation could
after) seem to indicate.
Taking into account bmim ILs, the reaction rate decreases
+
+
coordinate by means of its acidic hydrogen atom (H2 on bmim ,
+
18
along the series as follows: [bmim][BF4] > [bmim][PF6] >
CH3 on bm2im ) the oxygen atom on the isoxazole unit,
favoring the O-N bond breaking. Furthermore, as the transition
state of the title reaction has a lower aromatic character than
[
bmim][NTf2]. However, the detected order is completely
different from that determined on the basis of solvent parameter
1
9
20
ꢀ
values (ꢀ ) 0.376, 0.207, and 0.243 for [bmim][BF4],
the substrate, also π-π and π-cation interactions should
act by stabilizing the ground state to a higher degree than the
transition state. Data collected here show that the anion
coordination ability and the cation ability to give π-π interac-
tions are the most important effects. Indeed, reactivity decreases
1
5
[
bmim][PF6], and [bmim][NTf2], respectively). On the other
-
hand, considering [NTf2 ] ILs, the reaction rate decreases along
the series as follows: [bmpyrr][NTf2] > [bm2im][NTf2] >
[
bmim][NTf2], i.e., the two ILs with less acidic protons favor
-
-
-
the elimination. Also in this case the observed reactivity trend
cannot be explained only on the basis of R values (R ) 0.427,
on going from [BF4 ] and [PF6 ] ILs up to [NTf2 ] IL,
according to the lower coordination ability of the last anion.
+
0
.381, and 0.617 for [bmpyrr][NTf2], [bm2im][NTf2], and
On the other hand, reactivity increases on going from bmim
1
5
+
+
[
bmim][NTf2], respectively). Likewise, the reactivity trend is
and bm2im ILs up to bmpyrr IL, according to the different
ability to give π-π interactions and/or to act as H-donor species.
Activation Parameters. It has been reported that, for kinetics
carried out in RTILs, a sharp curvature in Arrhenius or Eyring
plots could be observed as a consequence of some significant
different from that obtained considering classical polarity
parameters such as ETN and π* (ET ) 0.544, 0.576, 0.644 for
N
[
bmpyrr][NTf2], [bm2im][NTf2], and [bmim][NTf2], respec-
tively; π* ) 0.954, 1.083, and 0.984 for [bmpyrr][NTf2],
1
5
21
[
bm2im][NTf2], and [bmim][NTf2], respectively).
structural changes in the RTIL. So for a careful analysis of
-
However, for the same anion ([NTf2 ]), the reaction seems
the temperature effect, the reaction was carried out at five
temperatures going from 293 to 313 K.
to be favored in the presence of an aliphatic cation such as
+
bmpyrr . Fixing the cation, the reaction seems to be favored
In all the cases considered, an excellent linear correlation of
log(kA,R/T) versus 1/T was obtained, indicating that the above
upsetting effect is not operating in the analyzed range and that
the calculated activation parameters depend only on the nature
of the elimination process. In Table 3 activation parameter
by anions having a higher symmetry and coordination ability,
-
-
16
such as [BF4 ] and [PF6 ].
As far as comparison among reactivity data in conventional
solvents and in different ILs is concerned, the higher reactivity
of 2 in ionic liquid media cannot be ascribed to higher polarity
effects. An increase in this parameter should favor the reaction;
some of the ILs used in this work have ENR values comparable
to those for MeOH and lower than those for DMF (ENR ) 217.2,
(18) Crowhurst, L.; Lancaster, N. L.; Pérez Arlandis, J. M.; Welton, T. J. Am.
Chem. Soc. 2004, 126, 11549–11555.
(19) (a) Atwood, J. L. Liquid Chlatrates, Inclusion Compounds; Atwood,
J. L., Davies, J. E. D., MacNicol, D. D., Eds.; Academic Press: London, UK,
1984; Vol. 1. (b) Su, B.-M.; Zhang, S.; Zhang, Z. C. J. Phys. Chem. B 2004,
108, 19510–19517. (c) Iwata, K.; Kakita, M.; Hamaguchi, H. J. Phys. Chem. B
2
18.5, 218.0, 217.7, and 221.0 for [bmim][BF4], [bmim][PF6],
17
[bmim][NTf2], MeOH, and DMF, respectively). Probably the
2
007, 111, 4914–4919.
higher reactivity could be a consequence of electrostatic
interactions. Bearing in mind the transition state structure, ions
constituting ILs could induce different effects. According to data
(20) (a) Hanke, C. G.; Johansson, A.; Harper, J. B.; Lynden-Bell, R. M. Chem.
Phys. Lett. 2003, 374, 85–90. (b) Holbrey, J. D.; Reichert, W. M.; Nieuwen-
huyzen, M.; Sheppard, O.; Hardacre, C.; Rogers, R. D. Chem. Commun. 2003,
4
76, 477. (c) Harper, J. B.; Lyndell-Bell, R. M. Mol. Phys. 2004, 102, 85–94.
1
8
5c
previously reported by Welton et al., and confirmed by us,
an ammonium-charged transition state could be stabilized by
(d) Deetlefs, M.; Hardacre, C.; Nieuwenhuyzen, M.; Sheppard, O.; Soper, A. K.
J. Phys. Chem. B 2005, 109, 1573–1598. (e) Lachwa, J.; Bento, I.; Duarte, M. T.;
Lopes, J. N. C.; Rebelo, L. P. N. Chem. Commun. 2006, 2445–2447. (f) Lyndell-
Bell, R. M.; Del Pópolo, M. G.; Youngs, T. G. A.; Kohanoff, J.; Hanke, C. G.;
Harper, J. B.; Pinilla, C. C. Acc. Chem. Res. 2007, 40, 1138–1145. (g) Hardacre,
C.; Holbrey, J. D.; Nieuwenhuyzen, M.; Youngs, T. G. A. Acc. Chem. Res. 2007,
40, 1146–1155.
(15) (a) Muldoon, M. J.; Gordon, C. M.; Dunkin, I. R. J. Chem. Soc., Perkin
Trans. 2 2001, 433, 435. (b) Crowhurst, L.; Mawdsley, P. R.; Pérez-Arlandis,
J. M.; Salter, P. A.; Welton, T. Phys. Chem. Chem. Phys. 2003, 5, 2790–2794.
(
16) (a) Golding, J. J.; MacFarlane, D. R.; Spiccia, L.; Forsyth, M.; Skelton,
(21) (a) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361–363. (b) Handy, S. T.; Okello, M. J. Org. Chem. 2005, 70, 1915–
1918.
B. W.; White, A. H. Chem. Commun. 1998, 1593–1594. (b) McLean, A. J.;
Muldoon, M. J.; Gordon, C. M.; Dunkin, I. R. Chem. Commun. 2002, 1880–
1
881. (c) Holbrey, J. D.; Reichert, W. M.; Nieuwenhuyzen, M.; Sheppard, O.;
Hardacre, C.; Rogers, R. D. Chem. Commun. 2003, 476–477.
17) Carmichael, A. J.; Seddon, K. R. J. Phys. Org. Chem. 2000, 13, 591–
95.
(22) Gordon, C. M.; McLean, A. J.; Muldoon, M. J.; Dunkin, I. R. In Ionic
Liquids as Green SolVents. Progress and Prospects; Rogers, R. D., Seddon, K. R.,
Eds.; ACS Symp. Ser. No. 856; American Chemical Society: Washington, DC,
2003; pp 357–369.
(
5
3
402 J. Org. Chem. Vol. 73, No. 9, 2008

Kemp Elimination
TABLE 4. Activation Parameter Values for the Pyrrolidine- and
Piperidine-Elimination Reaction of 2, in Different Solvent Media
observed in conventional solvents. Activation parameters seem
to indicate a low degree of C-H bond breaking in the transition
state. Consequently, a scarce relevance of variation in amines
and ILs structure on activation parameters was detected.
Finally, it is interesting to note that the IL effect is a function
of the probe reaction used. In particular, relative to the utilized
ILs in the present work, we have verified that an aliphatic cation
and an anion having higher coordination ability are more
efficient in catalyzing the reaction.
∆
H*pyrr
(kJ/mol)
∆S*pyrr
∆H*pip
∆S*pip
solvent
MeOH
DMF
(J/(K mol))
(kJ/mol)
(J/(K mol))
48 ( 2
45 ( 3
46 ( 1
49 ( 3
49 ( 2
45 ( 3
46 ( 2
-154 ( 5
-152 ( 11
-149 ( 4
-144 ( 10
-144 ( 8
-154 ( 8
-152 ( 7
41 ( 2
55 ( 2
54 ( 1
48 ( 2
51 ( 1
49 ( 2
45 ( 2
-183 ( 8
-124 ( 7
-131 ( 4
-156 ( 8
-145 ( 5
-149 ( 7
-163 ( 5
[
[
[
[
[
bmim][BF4]
bmim][PF6]
bmim][NTf2]
bm2im][NTf2]
bmpyrr][NTf2]
Experimental Section
values, collected in [bmim][BF4] solution, as a function of
amine, are reported (data collected at different temperatures are
available in Table 7 of the Supporting Information).
In Table 4 activation parameter values collected for the
pyrrolidine and piperidine elimination of 2 in different solvent
media are reported (data collected at different temperature are
available in Table 8 of the Supporting Information).
Materials. Commercial 1,4-dioxane, [bmim][BF
PF ] were used without any other purification. [bmim][NTf
bm im][NTf ], and [bmpyrr][NTf ] were prepared according to a
procedure previously reported. All ionic liquids, before use, were
dried on a vacuum line at 60 °C for at least 2 h, then stored in a
dryer under argon and over calcium chloride. Amines were freshly
distilled before use. Commercial 1,4-benzisoxazole was freshly
distilled before use.
4
], and [bmim]-
[
[
6
2
],
2
2
2
24
The activation parameters calculated span a narrow range.
q
Kinetics Measurements and Calculations. UV–vis spectra and
kinetic measurements were carried out by using a spectrophotometer
equipped with a Peltier temperature controller, able to keep the
temperature constant within 0.1 K. Kinetic runs were carried out
over the temperature range 293–313 K. The sample for a typical
kinetic run was prepared by mixing into a quarz cuvette (optical
path 0.2 cm) 500 µL of IL, 50 µL of a solution of substrate in
1,4-dioxane, and then 25 µL of a concentrated solution of amine
in 1,4-dioxane, previously thermostated. The obtained solution was
thermostated and the opportune volume of amine solution was
added. The concentration of substrate was constant and equal to
In particular mean values of ∆H ) (52 ( 2) kJ/mol and of
q
∆
S (-142 ( 8) J/(K mol) were calculated for elimination
induced by different amines in [bmim][BF4]. Similar results
were also obtained for pyrrolidine-induced elimination in
q
conventional solvent and ILs [mean values: ∆H ) (47 ( 2)
q
kJ/mol and of ∆S ) (-150 ( 8) J/(K mol)]. A little wider
spread of values was calculated for piperidine [mean values:
q
q
∆
H ) (49 ( 2) kJ/mol and of ∆S ) (-150 ( 6) J/(K mol)].
The above data about the reaction in ILs are not surprising.
Indeed, they confirm that weak interactions are operative. The
scarce relevance of amine structure on activation parameters
could be a consequence of an “early” transition state, having a
low C-H bond-breaking degree. This would not allow signifi-
cant differences to be detected by varying the amine structure.
For the same reason only small differences were observed on
changing the reaction media.
-
4
-3
2
2
.0 × 10 M; the amine concentration ranged from 8.7 × 10 to
-3
2 × 10 M. The reactions were studied over at least three half-
lives. In every case, the correlation coefficients were higher than
.9998. The apparent first-order rate constants obtained were
0
reproducible within (3%. All kinetic data were analyzed by means
of the KALEIDAGRAPH 3.0.1 software.
1
H NMR Measurements. NMR spectra were collected on a 300
MHz spectrometer. In NMR measurements the opportune volumes
of RTIL and cosolvent were mixed in a 5 mm NMR tube. A steam
coaxial capillary tube loaded with DMSO-d was used for the
6
Conclusions
On the whole, collected data show that the target reaction is
sensitive to amine structure rather than to amine basicity. It is
interesting to note that once more, for an E2 reaction, cyclic
secondary amines are more effective than acyclic ones. This
should be a consequence of a partially ordered structure of
the solvent medium. Moreover, for the studied reaction classical
solvent parameters do not unequivocally allow the effect of ILs
to be rationalized. This is consistent with data reported by
Kobrak about the differences in solvation phenomena in ionic
external lock of the NMR magnetic field/frequency and its signal
was used as the H NMR external reference at 2.56 ppm.
1
Acknowledgment. We thank MIUR (PRIN 20066034372)
and the University of Palermo for financial support.
Supporting Information Available: Rate constants collected
at different amines concentrations and at different temperature
1
values and H NMR spectra. This material is available free of
2
3
charge via the Internet at http://pubs.acs.org.
liquids and molecular solvents. Actually, a relationship
between polarity and the microscopic (molecular) electrostatic
behavior of the IL may exist, but it is quite different than that
JO702662Z
(
24) (a) Schröder, U.; Wadhawan, J. D.; Compton, R. G.; Marken, F.; Suarez,
(
23) (a) Kobrak, M. N. J. Chem. Phys. 2006, 125, 064502. (b) Kobrak, M. N.
J. Chem. Phys. 2007, 127, 184507. (c) Kobrak, M. N. J. Phys. Chem. B 2007,
11, 4755–4762. (d) Kobrak, M. N. Green Chem. 2008, 10, 80–86.
P. A.; Consorti, C. S.; de Souza, R. F.; Dupont, J. New J. Chem. 2000, 24,
1009–1015. (b) Cammarata, L.; Kazarian, S. G.; Salter, P. A.; Welton, T. Phys.
Chem. Chem. Phys. 2001, 3, 5192–5200.
1
J. Org. Chem. Vol. 73, No. 9, 2008 3403