Solvent reorganisation as the driving force for rate changes of Menschutkin reactions in an ionic liquid

Article abstract of DOI:10.1039/b909171h

The effect on the rate of reaction of each of a series of Menschutkin processes on changing from a molecular solvent to an ionic liquid was investigated. In each case, the rate acceleration observed at room temperature could be attributed to the change in

Full text of DOI:10.1039/b909171h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Solvent reorganisation as the driving force for rate changes of Menschutkin
reactions in an ionic liquid†
Hon Man Yau,a Andrew G. Howe,‡a James M. Hook,b Anna K. Croftc and Jason B. Harper*a
Received 11th May 2009, Accepted 19th June 2009
First published as an Advance Article on the web 9th July 2009
DOI: 10.1039/b909171h
The effect on the rate of reaction of each of a series of Menschutkin processes on changing from a
molecular solvent to an ionic liquid was investigated. In each case, the rate acceleration observed at
room temperature could be attributed to the change in the entropy of the system on reaching the
transition state, offsetting any enthalpic cost.
Introduction
Ionic liquids, which are typically made up of a bulky organic cation
and ◦a charge-diffuse anion,1–4 are salts which are liquid below
100 C.5 The ability to change the nature of the solvent through
modification of the components means that a variety of solvent
properties are available.3,4,6,7 Further, since the forces between the
components of the solvent are electrostatic, ionic liquids have very
low vapour pressures1,7 and are thus often proposed as alternatives
to environmentally damaging organic solvents.8
One of the challenges to the widespread application of ionic
liquids is that differences in the outcomes of reactions, both in
terms of rates and selectivities, are often observed on changing
from molecular to ionic solvents, but there has been little
Scheme 1 The Menschutkin reaction between the benzyl bromides 1a–g
investigation of the origin of these changes.9,§ In order for ionic
and pyridine 2 carried out in either acetonitrile or the ionic liquid
liquids to become broadly applicable, their effect on reaction
outcome needs to be understood in a fashion akin to that available
for molecular solvents.10
[Bmim][N(CF3SO2)2].14
The key difference between molecular solvents and ionic liquids
is the significant electrostatic interactions between the components
of the liquid. Along with resulting in the low vapour pressures so
characteristic of ionic liquids, these electrostatic interactions may
result in increased stabilisation of charged species and significant
ordering in the solution.
Our research effort aims to understand the origin of the
effects of ionic liquids on organic processes and to examine
whether electrostatic interactions between components of the
solution might be used to explain the outcomes observed. We
have demonstrated the effect of ionic liquid solvents on nitrile
oxide cycloadditions,11 both in terms of rate and regioselectivity,
with the changes being attributed to compression of, and hence
increased steric demand in, the transition state. We have also
examined the SN1 substitution of a tertiary chloride in mixtures
containing various proportions of the ionic liquid, 1-butyl-3-
methylimidazolium ([Bmim]+) bis(trifluoromethanesulfonyl)imide
([N(CF3SO2)2]-) and shown that the variation in rate observed12
was the result of enthalpic stabilisation of the incipient charges in
the transition state, which was offset by the necessary ordering of
the components of the solution.13
The work described herein aims to extend our observations on
substitution processes by considering the Menschutkin reaction
of substituted benzyl bromides and pyridine (Scheme 1).14 This
reaction was chosen because the extent of charge development is
less than for the SN1 process and both this and the extent of bond
formation and bond breakage in the transition state, whether it is
either ‘loose’ or ‘tight’,15 can be varied with the electronic nature
of the substituent.
aSchool of Chemistry, University of New South Wales, Sydney, NSW 2052,
Australia. E-mail: j.harper@unsw.edu.au; Fax: +61 2 9385 6141; Tel: +61
2 9385 4692
bAnalytical Centre, University of New South Wales, Sydney, NSW 2052,
Australia
cSchool of Chemistry, University of Wales Bangor, Bangor, Gwynedd LL57
2UW, United Kingdom
† Electronic supplementary information (ESI) available: Hammett plot for
the reaction shown in Scheme 1 at 300.8 K, second-order rate constants
for reactions shown in Scheme 1 at 300.8 K and 311.5 K (acetonitrile) and
278.1 K and 300.8 K ([Bmim][N(CF3SO2)2]), exemplar Eyring plots for
the data shown in Tables 2 and 3, second-order rate constants and Eyring
plots for the reactions of pyridine with benzyl chloride and with benzyl
iodide. See DOI: 10.1039/b909171h
‡ Visiting from the Department of Chemistry and Chemical Biology,
Harvard University, 12 Oxford Street, Cambridge, MA 02138, USA.
§ There are specific cases where the reactivity of the components of an ionic
liquid,23 particularly through the use of protic cations24 and basic anions25
alters the reaction outcome. These cases do not rely on the fundamental
properties of the ionic liquid and will not be referred to further here.
3572 | Org. Biomol. Chem., 2009, 7, 3572–3575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Second-order rate constants for the reactions described in
Scheme 1 at 289.6 K with either acetonitrile or [Bmim][N(CF3SO2)2] as
the solvent
Experimental
The benzyl halides 1a–g, 4 and 5 were either commercially available
or prepared from commercially available starting materials using
literature methods.16 [Bmim][N(CF3SO2)2] was prepared from
the corresponding chloride4 and dried to constant weight over
phosphorus pentoxide. Pyridine (2) was purified using a literature
method17 and analytical grade acetonitrile was used as received.
1H NMR spectra were recorded either on a Bruker Avance 500
spectrometer (500 MHz), a Bruker Avance 400 spectrometer
(400 MHz) or on a Bruker DPX 300 spectrometer (300 MHz)
using ca. 0.7 mL of reaction mixture (details below) in a 5 mm
NMR tube. Results were demonstrated to be independent of the
instrument used.
k2 (acetonitrile)/
k2 ([Bmim][N(CF3SO2)2])/
a
a
10-4 mol L-1 s-1
10-4 mol L-1 s-1
1a
1b
1c
1d
1e
1f
21.0 0.61
8.22 0.15
4.73 0.25
4.55 0.23
3.68 0.11
3.05 0.38
2.76 0.15
53.2 0.82
18.3 0.54
12.2 0.57
9.44 0.35
7.05 0.22
5.68 0.45
3.89 0.12
1g
a Uncertainties quoted are standard deviations.
Kinetic analyses of the Menschutkin reactions were carried
out in solutions containing the appropriate benzyl halide (ca.
0.05 mol L-1) and pyridine 2 (ca. 5–20 equiv.) over a range
of temperatures between 278 K and 341 K. In each case the
1
reaction was followed using H NMR spectroscopy until more
than 95% of the starting material was consumed. Spectra were
taken at regular intervals during the reaction and at least 20
spectra were obtained for each kinetic run. The extent of reaction
was deduced by integration of the signals corresponding to the
benzyl protons in the starting materials 1a–g (ca. d = 4.5) and the
products 3a–g (ca. d = 5.8). From this information, the pseudo
first-order rate constants for the reaction of each of the species
1a–g under these conditions were calculated and, consequently,
from the concentration dependence of these values, the second-
order rate constants at each temperature were also calculated. The
activation parameters were then determined using the bimolecular
form of the Eyring equation.18
Fig. 1 Hammett plot for the second-order rate constants for the reaction
outlined in Scheme 1 carried out at 289.6 K in either acetonitrile (ꢀ) or
the ionic liquid [Bmim][N(CF3SO2)2] (᭹).
is a ‘tight’ transition state.15 As the electron-donating ability of the
substituent increases, and a positive charge can be more readily
stabilised at the benzylic position, preferential bond breaking over
bond formation is observed. That is, there is a ‘loose’ transition
state.15
The effect of the ionic liquid on the Hammett plot is to increase
the slope of the tangent to the curve at any given point; this was
consistent across a range of temperatures (see the ESI†). While
this indicates an increased charge development on going to the
transition state compared to the case in acetonitrile, it is not
apparent simply from this data alone whether it is stabilisation
of the incipient charges by the components of the ionic liquid
which is responsible for the observed rate increases.
The activation parameters DH‡ and DS‡ were calculated for the
reactions of the bromides 1a–g from the temperature dependent
kinetic data using Eyring plots18 (Tables 2 and 3). In all cases,
the observed activation enthalpy is positive and there is little
change based on the electronic nature of the substituent. The
latter is important as it indicates that while the degree of charge
development at the benzylic position in the transition state changes
with the electronic nature of the substituent, this is not reflected in
the activation enthalpy. Similarly, for both solvent systems there
is little change in the entropy of activation with the nature of the
substituent, despite marked changes in the nature of the transition
state. This is likely to be the result of ordering of the solvent shell
as the transition state becomes more ‘loose’.
Results and discussion
Initially, the reaction of the benzyl bromides 1a–g was examined
using 79Br NMR spectroscopy19 by following the formation of
a single signal due to the bromide ion in acetonitrile. How-
ever, line broadening of the bromide signal in the ionic liquid
[Bmim][N(CF3SO2)2] made such analysis impractical and thus 1H
NMR spectroscopy was used.
The first thing to note is that for each of the substituted bromides
1a–g, kinetic analysis showed that the reaction proceeded through
a bimolecular rate-determining step with no measurable contribu-
tion from the corresponding unimolecular process. These second-
order rate constants for the Menschutkin reactions at 289.6 K
are tabulated below (Table 1) and the Hammett plot given (Fig. 1);
the latter represents the first such investigation undertaken in ionic
liquid media. Under the conditions used, the mole fraction of ionic
liquid present in the reaction mixture was ca. c = 0.9.
From this data, it is clear that the effect of moving from
acetonitrile to the ionic liquid [Bmim][N(CF3SO2)2] was to increase
the rate of reaction of the bromides 1a–g by 1.5–3 times, with the
greater change observed for electron-donating substituents. The
Hammett plot is non-linear in each instance, though it is important
to note that the data used are the second-order rate constants;
that is, the curvature is not due to a change in mechanism from an
SN2 to an SN1 process as the substituents become more electron-
donating. Rather, the extent of bond formation and bond breakage
is comparable for electron-withdrawing substituents. That is, there
The effect on the activation parameters of changing solvent
from acetonitrile to the ionic liquid [Bmim][N(CF3SO2)2] are best
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3572–3575 | 3573
©

View Article Online
Table 2 Activation parameters for the Menschutkin reaction of pyridine
2 with the benzyl bromides 1a–g in acetonitrile
Table 5 Activation parameters for the Menschutkin reaction of pyridine
2 with benzyl chloride 4 in either acetonitrile or [Bmim][N(CF3SO2)2]
a
a
a
DH‡/kJ mol-1
DS‡/J K-1 mol-1
DH‡/kJ mol-1
DS‡/J K-1 mol-1
1a
1b
1c
1d
1e
1f
40.9 1.2
43.2 1.1
42.6 1.3
43.4 0.8
47.7 1.0
50.0 2.8
44.2 1.0
-220
-219
-226
-224
-210
-205
-225
4
4
4
3
3
9
3
Acetonitrile
[Bmim][N(CF3SO2)2]
Difference
62.8 2.4
62.7 1.9
-0.1 3.0
-203
-190
13
7
6
9
a Uncertainties quoted are standard deviations.
1g
a Uncertainties quoted are standard deviations.
Table 6 Activation parameters for the Menschutkin reaction of pyridine
2 with benzyl iodide 5 in either acetonitrile or [Bmim][N(CF3SO2)2]
a
DH‡/kJ mol-1
DS‡/J K-1 mol-1
Table 3 Activation parameters for the Menschutkin reaction of pyridine
2 with the benzyl bromides 1a–g in [Bmim][N(CF3SO2)2]
Acetonitrile
[Bmim][N(CF3SO2)2]
Difference
41.7 0.3
51.3 0.3
9.6 0.4
-220
-186
34
1
1
1
a
a
DH‡/kJ mol-1
DS‡/J K-1 mol-1
a Uncertainties quoted are standard deviations.
1a
1b
1c
1d
1e
1f
47.7 1.1
48.8 0.9
53.5 2.0
49.9 0.8
51.5 1.2
53.1 2.1
48.7 0.7
-188
-193
-181
-195
-191
-187
-207
4
3
7
3
4
4
2
was the leaving group used. As such, kinetic analyses, equivalent
to those described above, were carried out for benzyl chloride 4 and
benzyl iodide 5 and the activation parameters calculated (Tables 5
and 6).
1g
a Uncertainties quoted are standard deviations.
The first thing to note is that there is a trend in the effect of
the ionic liquid on the activation parameters. Both the change
in enthalpy of activation (DH‡) and the change in the entropy
of activation (DS‡) increase on going from the chloride 4 to the
bromide 1d to the iodide 5. This is consistent with a decrease in
stabilisation of the incipient charges in the transition state, along
with less ordering, relative to the starting material, as the size
of the halogen increases and the electronegativity decreases. This
can be rationalised by decreased ordering in, and stabilisation of,
the transition state as the charge density on the incipient anion
decreases.
While the trend in these effects is clear, the most critical
observation from this data is that the effect of changing the
leaving group was small. Irrespective of the halide used, the rate
enhancements observed over the range of temperatures considered
on changing the solvent to an ionic liquid are primarily as a result
of an increase in the entropy of activation.
It is also worth considering whether the effect of the ionic liquid
varies with the electronic nature of the aromatic substrate (and
hence the nature of the transition state). The fact that the effect of
the ionic liquid on the enthalpy of activation does not change with
the nature of the substituent on the aromatic ring suggests that
the charge development, which is shown in Fig. 1 to be greater
at the benzylic position in an ionic liquid, is not stabilised to
a greater extent in [Bmim][N(CF3SO2)2]. This may be the result
of a congested transition state where interaction of the cationic
reaction centre with the anion of the ionic liquid is restricted. Such
congestion would account for the fact that the effect of the ionic
liquid is the same for all substituents, though in order to explain the
increase in the activation enthalpy it requires that the ionic liquid
is less able to interact with the incipient charges than acetonitrile.
This is considered unlikely given the significant cation–halogen
interactions observed for an SN1 process previously,13 and the
relative accessibility of the incipient negative charge on the halogen
atom.
Table 4 Changes in the activation parameters for the Menschutkin
reaction of pyridine 2 with the benzyl bromides 1a–g on going from
acetonitrile to [Bmim][N(CF3SO2)2] as the solvent
a
a
D(DH‡)/kJ mol-1
D(DS‡)/J K-1 mol-1
1a
1b
1c
1d
1e
1f
6.9 1.6
5.6 1.4
10.9 2.3
6.6 1.2
3.9 1.6
3.2 3.5
4.5 1.2
32
26
45
29
19
17 12
18
5
5
8
4
5
1g
4
a Uncertainties quoted are standard deviations.
seen in Table 4. The most striking point is that these changes are
very different from those which have been observed previously
for the substitution of an alkyl chloride that occurs through an
SN1 mechanism, where a dramatic enthalpic stabilisation was
counterbalanced by an entropic cost.13 In this case there is very
little change in the enthalpy of activation, with the effect ranging
from zero, within the limits of the experiment, to ca. 11 kJ mol-1 in
the case of the benzyl bromide 1c. While the values vary slightly,
the key point is that enthalpic effect of the ionic liquid is small
and unfavourable in the present case. In contrast, the entropy of
activation increases on moving to the ionic liquid and, again, this
is observed for all substituents. To reiterate, for the Menschutkin
reaction of each of the benzyl bromides 1a–g, the rate enhancement
observed on changing from acetonitrile to an ionic liquid is due
to an increase in disorder on going to the transition state. It is
worthwhile considering the origin of these changes as they differ
so significantly from those observed previously for a process that
proceeds through an SN1 mechanism.13
One of the potential sources of the differences in the activation
parameters between this system and that reported previously13
3574 | Org. Biomol. Chem., 2009, 7, 3572–3575
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
A similar argument also holds with respect to the observed
entropies of activation. Whilst the congested transition state might
limit the extent to which organisation of the ionic liquid changes
with different electron-donating groups, it is still not clear as to
why the entropy of activation decreases on going to an ionic liquid;
simply considering the transition state, this implies there is less
ordering in the ionic liquid case than for acetonitrile, which is
considered unlikely for the same reasons outlined above.
Until this point, the interaction of the solvents with the starting
materials has not been discussed. In the SN1 process studied
previously12,13 interactions with the starting material were limited.
By comparison, the charge separation in the transition state of
the SN1 process is considerable and the interactions with the ionic
solvent significant. In this case, as the reaction proceeds through
an SN2 mechanism, the charge development is not as significant
in the transition state and hence the interaction with the solvent
might be expected to be less than in the SN1 case. However, this
change would only decrease the magnitude of the entropic effect;
the activation entropy would still be expected to be negative. As
such, an increased interaction of the solvent with the starting
materials is implied.
The introduction of aromatic components to the starting
material in the work presented here needs to be considered.
Aromatic compounds have anomalously high solubilities in ionic
liquids20 and this has been demonstrated, using molecular dynam-
ics simulations for simple systems, to be due to the interaction of
the quadrupole of the aromatic compound with the components
of the ionic liquid.21 Importantly, in these simulations, significant
ordering of the components of the solution about the aromatic
solute was observed.21,22
AGH would like to acknowledge the support provided by
a Herchel Smith Harvard Summer Undergraduate Research
Fellowship. AKC is grateful to the Royal Society for the award
of a Travel Grant. JBH acknowledges financial support from
the University of New South Wales Faculty Research Grants
Programme.
Notes and references
1 C. L. Hussey, Pure Appl. Chem., 1988, 60, 1763.
2 R. H. Dubois, M. J. Zaworotko and P. S. White, Inorg. Chem., 1989, 28,
2019; J. S. Wilkes and M. J. Zawarotko, J. Chem. Soc., Chem. Commun.,
1992, 965.
3 P. Bonhoˆte, A. Das, N. Papageorgiou, K. Kalanasundram and M.
Gra¨tzel, Inorg. Chem., 1996, 35, 1168.
4 J. G. Huddleston, H. D. Willauer, R. P. Swatloski, A. E. Visser and
R. D. Rogers, Chem. Commun., 1998, 1765.
5 S. Z. El Abedin and F. Endres, Acc. Chem. Res., 2007, 40,
1106.
6 H. Stegemann, A. Rhode, A. Reiche, A. Schnittke and H. Fu¨llbier,
Electrochim. Acta, 1992, 37, 379; A. Elaiwi, P. B. Hitchcock, K. R.
Seddon, N. Srinivasan, Y.-M. Tan and T. Welton, J. Chem. Soc., Dalton
Trans., 1995, 3467; K. R. Seddon, J. Chem. Technol. Biotechnol., 1997,
68, 351.
7 K. R. Seddon, Kinet. Catal., 1996, 37, 693.
8 R. D. Rogers and K. R. Seddon, Science, 2003, 302, 792.
9 J. B. Harper and M. N. Kobrak, Mini-Rev. Org. Chem., 2006, 3,
253.
10 M. B. Smith, and J. March, March’s Advanced Organic Chemistry,
Wiley-Interscience, Chichester, UK, 2001.
11 C. E. Rosella and J. B. Harper, Tetrahedron Lett., 2009, 50,
992.
12 B. Y. W. Man, J. M. Hook and J. B. Harper, Tetrahedron Lett., 2005,
46, 7641.
13 H. M. Yau, S. A. Barnes, J. M. Hook, T. G. A. Youngs, A. K. Croft
and J. B. Harper, Chem. Commun., 2008, 3576.
If there is significant organisation of the solvent about the aro-
matic starting materials it is likely that this will require significant
disruption on going to the transition state. This disruption may
simply be a result of bringing the two reactants together but there
may also be some contribution from a change in the quadrupole
of each aromatic system (as a result of some charge development).
Experimental efforts to deconvolute this contribution are currently
underway.
14 P. Haberfield, A. Nudelman, A. Bloom, R. Romm and H. Ginsberg,
J. Org. Chem., 1971, 36, 1792.
15 C. Lim, S.-H. Kim, S.-D. Yoh, M. Fujio and Y. Tsuno, Tetrahedron
Lett., 1997, 38, 3243.
16 G. H. Daub and R. N. Castle, J. Org. Chem., 1954, 19, 1571; R. P.
Hanzlik, A. R. Schaefer, J. B. Moon and C. M. Judson, J. Am. Chem.
Soc., 1987, 109, 4926; T. Yokoyama, G. R. Wiley and S. I. Miller, J. Org.
Chem., 1969, 34, 1859.
17 W. L. F. Armarego, and C. Chai, Purification of Laboratory Chemicals,
Butterworth–Heinemann, Oxford, UK, 5th edn, 2003.
18 H. Eyring, J. Chem. Phys., 1935, 3, 107.
19 S. J. Chan, A. G. Howe, J. M. Hook and J. B. Harper, Magn. Reson.
Chem., 2009, 47, 342.
Conclusions
20 L. A. Blanchard and J. F. Brennecke, Ind. Eng. Chem. Res., 2001, 40,
The ionic liquid [Bmim][N(CF3SO2)2] accelerates the Menschutkin
reaction of benzyl halides with pyridine. The origin of this rate
enhancement is an increase in the entropy of activation, which
offsets a small increase in the enthalpy of activation, in contrast
to the effects on substitution processes reported previously. The
rate of reaction has been shown to be independent of the nature of
both the aromatic substrate and the leaving group, and is proposed
to be governed by the organisation of the ionic liquid about the
aromatic starting materials.
287.
21 C. G. Hanke, A. Johansson, J. B. Harper and R. M. Lynden-Bell, Chem.
Phys. Lett., 2003, 374, 85.
22 J. B. Harper and R. M. Lynden-Bell, Mol. Phys., 2004, 102, 85; R. M.
Lynden-Bell, M. G. Del Po´polo, T. G. A. Youngs, J. Kohanoff, C. G.
Hanke, J. B. Harper and C. C. Pinilla, Acc. Chem. Res., 2007, 40, 1138.
23 J. R. Harjani, S. J. Nara and M. M. Salunkhe, Tetrahedron Lett., 2002,
43, 1127; V. K. Aggarwal, I. Emme and A. Mereu, Chem. Commun.,
2002, 1612; H.-P. Nguyen, H. Matondo and M. Baboule`ne, Green
Chem., 2003, 5, 303; J. Gui, Y. Deng, Z. Hu and Z. Sun, Tetrahedron
Lett., 2004, 45, 2681.
24 S. S. Palimkar, S. A. Siddiqui, T. Daniel, R. J. Lahoti and K. V.
Srinivasan, J. Org. Chem., 2003, 68, 9371; H. Zhang, F. Xu, X. Zhou,
G. Zhang and C. Wang, Green Chem., 2007, 9, 1208.
25 J.-M. Xu, Q. Wu, Q.-Y. Zhang, F. Zhang and X.-F. Lin, Eur. J. Org.
Chem., 2007, 1798; B. C. Ranu, R. Jana and S. Sowmiah, J. Org. Chem.,
2007, 72, 3152.
Acknowledgements
HMY acknowledges the support of the Australian govern-
ment through the receipt of an Australian Postgraduate Award.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 3572–3575 | 3575
©