Nitrogen: Versus phosphorus nucleophiles-how changing the nucleophilic heteroatom affects ionic liquid solvent effects in bimolecular nucleophilic substitution processes

Article abstract of DOI:10.1039/c6nj00721j

A series of nitrogen and phosphorus nucleophiles have been investigated to determine whether the previously established ionic liquid solvent effects on a bimolecular nucleophilic substitution (S_N2) reaction vary with the nature of the nucleophilic centre. Reaction of group 15 triphenyl nucleophiles with benzyl bromide showed a different trend in the rate constant with increasing proportions of ionic liquid in the reaction mixture than was observed with pyridine. This result suggests additional interactions are important; a supposition supported by differences in reaction outcome observed when the electrophile was varied in reactions with triphenylphosphine. A novel ionic liquid solvent effect was observed in the reaction of tributylamine with benzyl bromide, with the position of equilibrium varying with the proportions of the ionic liquid present in the reaction mixture. Overall, the work presented demonstrates the importance of considering all possible interactions between an ionic liquid solvent and species along the reaction coordinate and has expanded upon our current predictive framework for ionic liquid solvent effects. Such understanding is important as it allows further development of a predictive framework for the application of ionic liquids in preparative chemistry.

Full text of DOI:10.1039/c6nj00721j

NJC
View Article Online
View Journal
PAPER
Nitrogen versus phosphorus nucleophiles – how
changing the nucleophilic heteroatom affects
ionic liquid solvent effects in bimolecular
nucleophilic substitution processes†
Cite this:DOI: 10.1039/c6nj00721j
Karin S. Schaffarczyk McHale, Rebecca R. Hawker and Jason B. Harper*
A series of nitrogen and phosphorus nucleophiles have been investigated to determine whether the previously
established ionic liquid solvent effects on a bimolecular nucleophilic substitution (SN2) reaction vary with the
nature of the nucleophilic centre. Reaction of group 15 triphenyl nucleophiles with benzyl bromide showed a
different trend in the rate constant with increasing proportions of ionic liquid in the reaction mixture than
was observed with pyridine. This result suggests additional interactions are important; a supposition supported
by differences in reaction outcome observed when the electrophile was varied in reactions with
triphenylphosphine. A novel ionic liquid solvent effect was observed in the reaction of tributylamine with benzyl
bromide, with the position of equilibrium varying with the proportions of the ionic liquid present in the reaction
mixture. Overall, the work presented demonstrates the importance of considering all possible interactions
between an ionic liquid solvent and species along the reaction coordinate and has expanded upon our current
predictive framework for ionic liquid solvent effects. Such understanding is important as it allows further
development of a predictive framework for the application of ionic liquids in preparative chemistry.
Received (in Nottingham, UK)
7th March 2016,
Accepted 27th June 2016
DOI: 10.1039/c6nj00721j
www.rsc.org/njc
To address this limitation, significant effort has been
devoted to investigating ionic liquid solvent effects on well-
Introduction
Ionic liquids typically consist of a bulky, asymmetric cation,1 with known substitution reactions.13–31 Notable are those studies
an accompanying anion of varying coordination strength.2 The involving quantitative analysis of the differences in reaction
diffuse nature of the charge on the component ions frustrates outcome between ionic liquids and molecular solvents in an
crystallisation, lowering the melting points of these salts; ionic attempt to establish a predictive model of ionic liquid solvent
liquids are arbitrarily defined as those salts with normal melting effects.13–16,18,20,22,23,25–30,32 Of particular relevance to the work
points below 100 1C.3–6 The electrostatic interactions present in detailed here is the SN2 reaction between pyridine 1 and each
ionic liquids result in their low vapour pressures1,3,6 which, of the benzyl halides 2 and 3 (Scheme 1) which has been
coupled with their good thermal stabilities,2,5 leads to ionic extensively considered in ionic liquids.13–16
liquids being considered as potential alternatives to conventional
Part of these initial studies considered how the rate constant
organic solvents.5,7–9 The main drawback to the application of of the reaction was affected by varying the proportion of the
ionic liquids as solvents for synthetic processes is that there is ionic liquid 1-butyl-3-methylimidazolium bis(trifluoromethane-
currently a limited understanding as to how ionic liquids affect sulfonyl)imide ([Bmim][N(SO2CF3)2], 6) in the reaction mixture;
reaction outcome compared to conventional solvents.2,10–12
the co-solvent was acetonitrile. This type of study is of interest
as ionic liquids are often used in mixtures with molecular
solvents.33 It was shown that there was an increase in the
School 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
† Electronic supplementary information (ESI) available: Preparation of the ionic
liquid 6; details of the NMR signals used to follow reaction progress in each case;
the mathematical forms from which the rate data was derived for each system;
mole fraction kinetic dependence plot for the reaction between species 1 and 3;
Eyring plots from which activation parameters were derived; the ratio between the
forward and reverse rate constants for the reaction between reagents 3 and 10a;
the compositions of all the stock solutions prepared; rate constant data, includ-
ing the data on which Fig. S1 and S2–S6 are based, including temperature
dependent kinetic data from which the activation parameters shown in
Tables 1 and 5 are derived. See DOI: 10.1039/c6nj00721j
Scheme 1 The Menschutkin reaction between pyridine 1 and the benzyl
halides 2 and 3 to give the corresponding salts 4 and 5.13–16
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
Scheme 2 The SN2 reaction between the group 15 triphenyl nucleophiles
7 and the benzyl halides 2 and 3 to give the products 8 and 9.
Fig. 1 The bimolecular rate constant of the reaction between pyridine 1
and benzyl chloride 2 in various mole fractions of the ionic liquid 6 in
acetonitrile at 40.0 1C.‡ Reproduced from Keaveney et al.14
Scheme 3 The SN2 reaction between the group 15 tributyl compounds
10 and benzyl halides 2 and 3 to give the products 11 and 12.
bimolecular rate constant (k2) of the reaction between species 1 how aromatic and aliphatic substituents on the nucleophile
and 2 with increasing concentrations of the ionic liquid 6 in the can be influential.
reaction mixture (Fig. 1).14
The activation parameters for this process were determined
in both acetonitrile and the ionic liquid 6. These data showed
Experimental
that with the ionic liquid 6 as the solvent (w6 0.85) there was an
Benzyl bromide 2 and benzyl chloride 3 were commercially available
increase in the enthalpy of activation accompanied by an
and were purified through distillation then stored over activated
increase in the entropy of activation relative to acetonitrile.
molecular sieves at 4 1C prior to use. Triphenylamine 7a was
The contribution of the entropic benefit outweighed that of the
recrystallized from ethanol.38 Tributylamine 10a was distilled from
enthalpic cost resulting in the rate enhancement observed. The
calcium hydride and stored at 4 1C.39 Tributylphosphine 10b was
cause of this trend in the activation parameters was deter-
distilled under reduced pressure and stored under nitrogen. The
mined, through deconvolution studies and molecular dynamics
ionic liquid [Bmim][N(SO2CF3)2] 6 was prepared according to litera-
simulations, to be an interaction between the nitrogen lone
ture methods,40,41 in which 1-bromobutane was added to 1-methyl-
pair of pyridine 1 and the cation of the ionic liquid 6.15
imidazole to afford the bromide salt precursor which was then
The understanding developed has been extended to con-
treated with lithium bis(trifluoromethanesulfonyl)imide through a
sider the effect of changing the nature of the cation of the ionic
salt metathesis reaction to yield [Bmim][N(SO2CF3)2] 6. The ionic
liquid, with reduction in the key interaction corresponding
liquid 6 was dried under reduced pressure (o0.5 mbar) at room
temperature before use and was found to contain o300 ppm
water using Karl Fischer titrimetry. Absence of the bromide
salt precursor was confirmed using ion chromatography. Full
details for this preparation can be found in the ESI.†
to a decrease in the observed solvent effects.16 The utility of
this rationale has been exploited on another reaction, the
condensation of an aldehyde and an amine, in which similar
interactions are present.34,35 It was therefore of interest to
determine whether this model could be extended to include
other nucleophilic heteroatoms. The work detailed here is an
investigation of the effects of an ionic liquid solvent on SN2
reactions in which the nucleophilic heteroatom is either nitrogen
or phosphorus.
Phosphabenzene would have been the ideal nucleophile,
allowing direct comparison to the pyridine 1 ionic liquid
solvent effect studies,14–16,28 however the nucleophilic site
of phosphabenzene is not the heteroatom.36,37 Instead SN2
reactions involving triphenyl and tributyl group 15 nucleo-
philes reacting with benzyl halides 2 and 3 were investigated
(Schemes 2 and 3, respectively) as these species are readily
available and have reactive heteroatoms.
Either 1H or 31P{1H} NMR spectroscopy was used to monitor
reaction progress, dependent on the species present, on either
a Bruker Avance III 400, Bruker Avance III 500 or Bruker Avance
III 600 spectrometer with either a TBI or BBFO probe using
ca. 0.5 mL of reaction mixture (detailed below) in a 5 mm NMR
tube. Results were shown to be reproducible between different
probes and spectrometers. The location and nature of the
signals used for each case are detailed in the ESI.†
Details pertaining to kinetic data analysis as well as sample
preparation and conditions including; the exact mole fractions,
stock solution composition by mass, temperatures and rate
constants for the systems used can be found in the ESI.†
This work allows for an increased understanding of how
variation of the nucleophilic heteroatom impacts the observed
ionic liquid solvent effects but also permits a comparison of
Results and discussion
Kinetic studies involving triphenylamine 7a
Initially, the reaction between triphenylamine 7a and benzyl
bromide 3 was investigated to allow comparison of the ionic
liquid solvent effects with the pyridine 1 case; both cases involve
‡ Some uncertainties fall within the size of the markers used. Uncertainties are
reported as either half the range (Fig. 1) or the standard deviation (Fig. 2–6) of
triplicate results.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
Table 1 The activation parameters obtained for the reaction between
a nitrogen nucleophile. The reaction involving benzyl bromide 3
proved sufficiently fast to monitor using initial rate methodology;§
the reaction of triphenylamine 7a with benzyl chloride 2 was
too slow to be practically monitored.
triphenylamine 7a and benzyl bromide
[Bmim][N(SO2CF3)2] 6 (w6 0.78)
3
in acetonitrile (w6 0) and
w6
DH‡ a/kJ molÀ1
DS‡ a/J KÀ1 molÀ1
Due to the practical constraints placed on the choice of
electrophile it was important to consider whether variation of
the leaving group might affect the reaction; previously benzyl
chloride 2 was used as the electrophile.14 The effect of varying
the proportion of the salt 6 in the reaction mixture on the rate
0
0.78
74.8 Æ 2.9
92.0 Æ 7.8
À242 Æ 9
À170 Æ 23
a
Errors reported are from the fit of the linear regression.
of the SN2 reaction involving pyridine 1 was repeated using The trend in k2 for the triphenylamine 7a and pyridine 1 reactions
benzyl bromide 3. The results of these experiments (Fig. S1, as the proportion of the salt 6 in the reaction mixture varies are
ESI†) match the trends shown previously for the chloride case different; in the case of nucleophile 7a, the rate constant increases
(Fig. 1),14 showing that for the pyridine 1 system (Scheme 1) dramatically at the highest mole fraction whereas the pyridine 1
there is no notable leaving group effect. As such, comparison of case displays a consistent increase (Fig. S1, ESI†). The only
the triphenylamine 7a and benzyl bromide 3 system (Scheme 2) similarity is that for both SN2 reactions a rate enhancement is
with that of the pyridine 1 system (Scheme 1) is a reasonable observed on moving from acetonitrile to the highest mole
starting point for this study. The dependence of the bimolecu- fraction of the ionic liquid 6 examined.13,14,16,28
lar rate constant of the reaction between species 3 and 7a on
To further understand the microscopic origins of these ionic
the proportion of the salt 6 in the reaction mixture was liquid solvent effects, temperature dependent kinetic studies
investigated (Fig. 2). For the kinetic studies presented here were undertaken. These were conducted at w6 0 and 0.78 (where
the salt 6 was chosen as it was the ionic liquid used in the acetonitrile and the salt 6 are diluted by reagents only) Eyring
pyridine 1 studies discussed above.13–16,28
plots constructed and activation parameters for these solvent
From Fig. 2 it can be seen that when moving from w6 0 to ca. compositions determined (Fig. S2, ESI,† and Table 1).
0.4 there is an increase in the rate constant, followed by a It is worth noting that the uncertainties in the activation
decrease in the rate constant upon moving to w6 ca. 0.6. There is parameters for this process are large compared to those deter-
then an increase in k2 when moving to the highest mole mined for the reaction of reagents 1 and 3.13 These large
fraction examined, w6 0.78; the greatest change is seen on uncertainties result from the comparatively large uncertainties
moving from w6 0.70 to 0.78. Of particular interest is the in the rate data, particularly for the ionic liquid case, seen
difference in the rate constants between w6 0 and 0.78 being in Fig. 2. However these uncertainties do not affect the conclu-
ca. 12-fold, this is greater than that observed for the pyridine 1 sions drawn given the differences between the activation para-
system (Table S1 and Fig. S1, ESI†).
meters determined for the acetonitrile and ionic liquid 6 cases
Whilst direct comparison of absolute rate changes cannot be considered.
made between the triphenylamine 7a and pyridine 1 systems
The activation parameters presented in Table 1 show an
(as an elevated temperature was required to monitor the former increase in both the enthalpy and entropy of activation when
case in a reasonable time), trends in the rate constants upon moving from w6 0 to 0.78; this was observed for the pyridine 1
increasing the proportion of the salt 6 present can still be compared. SN2 reaction.13,14,16,28 The increase in activation parameters
may be due to either an increase in interactions between the
ionic liquid 6 and starting materials or a decrease in inter-
actions between the ionic liquid 6 and the transition state
complex, relative to acetonitrile. The former interaction has been
used to rationalise outcomes for the pyridine 1 case;13–16,28 the
latter is considered unlikely in both cases given the charge
development in the transition state. Yet, changing the nucleo-
phile can change the manner in which the ionic liquid interacts
with species along the reaction coordinate. Hence, changes in
either the nature or the magnitude of these interactions may be
responsible for the difference in the mole fraction dependence
of the salt 6 on the rate constant.
A number of factors might contribute to the differences in
solvation of the nucleophile 7a and species along the reaction
coordinate, relative to the reaction considered previously
(Scheme 1). The presence of three, bulky phenyl rings around
Fig. 2 The bimolecular rate constant of the reaction between triphenyl-
amine 7a and benzyl bromide 3 in different mole fractions of the ionic
liquid 6 in acetonitrile at 75.0 1C.‡
the nucleophilic centre might be expected to limit cation–
nucleophile interactions. Introduction of these aromatic species
§ The product 9a was characterised as a mixture with the starting materials 3 and
may also result in quadrupole–cation interactions with the cation
7a after heating at reflux in acetonitrile for 1 month. The reaction had proceeded
to ca. 5%; this may represent the equilibrium position under these conditions.
of the salt 6.42,43 The nature of the electrophile, particularly
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
Table 2 The activation parameters obtained for the reaction between
reagents 3 and 7b in acetonitrile (w6 0), mixtures of acetonitrile and
[Bmim][N(SO2CF3)2] 6 (w6 0.36 and w6 0.70) and [Bmim][N(SO2CF3)2] 6 (w6 0.96)
variation of the leaving group, may influence interactions
between the salt 6 and the electrophile; given this is not observed
in the pyridine 1 system suggests that such interactions are not
dominant. Changing the reagents, will also change the nature of
the transition state (and hence interactions with it), particularly
the extent of charge development (the transition state may
change from being more ‘closed’, or ‘tight’, to being more ‘open’,
or ‘loose’44) and accessibility (due to the steric bulk of the
reagents). These additional interactions are likely present in
the pyridine 1 system (Scheme 1) but do not significantly
contribute to the solvent effects observed.13 Another contributor
may be differences in organisation of the solvent as the amount
of salt 6 in the reaction mixture changes, an effect previously
noted in the literature.25,31 As appropriate, each of the above
factors are investigated and discussed further below.
w6
DH‡ a/kJ molÀ1
DS‡ a/J KÀ1 molÀ1
0
44.8 Æ 3.7
57.1 Æ 1.5
46.2 Æ 3.0
57.2 Æ 1.5
À210 Æ 13
À163 Æ 5
À202 Æ 10
À163 Æ 5
0.36
0.70
0.96
a
Errors reported are from the fit of the linear regression.
Once again, temperature dependent kinetic studies were
undertaken to examine the origin of these solvent effects. These
studies were undertaken at w6 0, 0.36, 0.70 and 0.96 (the
acetonitrile case, local extremes of k2 and the ionic liquid case,
respectively) with the resultant Eyring plots (Fig. S3, ESI†)
allowing calculation of activation parameters in each solvent
mixture (Table 2).
Kinetic studies involving triphenylamine 7b
Given the main aim of this work, the equivalent phosphorus
nucleophile 7b was then investigated. In a similar fashion to
the above triphenylamine 7a system a mole fraction depen-
dence kinetic study was undertaken for the reaction between
triphenylphosphine 7b and benzyl bromide 3 (Scheme 2). The
results of these studies are shown in Fig. 3.
From these data it can be seen that there is ca. two-fold increase
in k2 between the acetonitrile case (w6 0) and the highest mole
fraction examined (w6 0.96).¶ A similar rate enhancement is also
observed at w6 ca. 0.4 with a significant decrease in k2 observed on
moving to w6 ca. 0.7. The observed trend is similar to that shown
for the above triphenylamine 7a reaction, noting once more that
the different temperatures used limit direct comparison of rate
constant data. The main difference between the systems involving
either nucleophile 7a or 7b is the relative difference of the rate
constants between the minimum and the highest mole fraction
used (w6 0.60 - 0.78 and w6 0.79 - 0.96, respectively).
An increase in the enthalpy and entropy of activation were
observed upon moving from w6 0 to 0.36, followed by a decrease
in both activation parameters between w6 0.36 and 0.70 and
then an increase in both activation parameters again when
moving from w6 0.70 to 0.96. It should be noted that the
activation parameters at w6 0.36 and 0.96 are the same within
error suggesting the microscopic origins of the rate enhance-
ment observed at these two mole fractions in the mole fraction
dependence studies may be similar. The local minimum rate
constant and decreased activation parameters observed at w6
0.70 suggest that the factors that are causing the rate enhance-
ment observed at w6 0.36 are counteracted as the amount of the
ionic liquid 6 is increased to w6 ca. 0.7.
There is an increase in both activation parameters for each
of the amine 7a and phosphine 7b cases; the entropic compo-
nent dominates, causing the observed rate increases on moving
to w6 0.78 and 0.96, respectively (cf. acetonitrile).8 Given this
and the similarities in the dependence of the reaction rate
for both nucleophiles 7a and 7b on the amount of the salt 6 in
the mixture, it is reasonable to suggest that the interactions
causing these changes are the same.
At the points where the largest k2 values are observed (w6 0.36
and 0.96), the changes in activation parameters relative to
acetonitrile are likely due to the cation–nucleophile interaction
implicated in other SN2 processes. Conversely, for the local
minimum of k2 for the triphenylphosphine 7b system (w6 0.70)
the change in activation parameters is likely due to other inter-
actions becoming more significant at this solvent composition.
These competing interactions are likely between the ionic liquid
6 and the transition state complex.
Fig. 3 The bimolecular rate constant of the reaction between triphenyl-
phosphine 7b and benzyl bromide 3 in different mole fractions of the ionic
liquid 6 in acetonitrile at 24.4 1C.‡
Investigating the importance of the electrophile
As the triphenylphosphine 7b nucleophile is more reactive than
triphenylamine 7a it is possible to investigate how varying the
¶ The highest mole fraction of the salt 6 in the reaction mixture examined for the
triphenylphosphine 7b system was higher than that examined for triphenylamine
7a system due to the faster rate of the triphenylphosphine 7b reaction thus
resulting in smaller concentrations of reagents being required to allow practical
measurement of rate constants. This does not impact on the comparisons made.
8 Comparison of data at the highest mole fractions examined, despite their
different values, for both the triphenyl nucleophiles 7 is possible as there is a rate
increase observed in both cases, relative to the equivalent reactions in
acetonitrile.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
construction of Eyring plots (Fig. S4, ESI†) and calculation of
the activation parameters in each solvent system (Table 3).
Upon moving from acetonitrile (w6 0) to the highest mole
fraction of the salt 6 examined (w6 0.78) there is an increase in
both the enthalpy and entropy of activation; the values
obtained at w6 0.20 are the same as those at w6 0.78 given
uncertainties in the measurements (Table 3). Both the depen-
dence of the rate constant on the proportion of salt 6 in the
reaction mixture and the activation parameter data match that
for reactions involving pyridine 1 as the nucleophile.13,14
In contrast to the reaction of triphenylphosphine 7b with
bromide 3, no local minimum exists in the mole fraction
dependence kinetic study for the chloride 2 case. This suggests
that any microscopic interactions responsible for this feature
are not present in the chloride 2 case. The only change is the
nature of the leaving group, suggesting interaction of the ionic
liquid 6 and either the electrophilic species 2/3 or the incipient
charge on the halogen in the transition state is causing this
effect. It is unlikely that interaction between the salt 6 and the
benzylic reagent (2 or 3) is causing the difference in ionic liquid
solvent effects observed for the triphenylphosphine 7b system
as no such difference is noted for the pyridine 1 systems
(Scheme 1). Therefore it is more likely that interaction between
the transition state complex and the ionic liquid 6 is the key
interaction causing this change.
Fig. 4 The bimolecular rate constant of the reaction between triphenyl-
phosphine 7b and benzyl chloride 2 in different mole fractions of the ionic
liquid 6 in acetonitrile at 63.6 1C.‡
electrophile may affect the observed ionic liquid solvent effects.
Kinetic studies were undertaken in which values of k2 for the
reaction between triphenylphosphine 7b and benzyl chloride 2
were determined in varying solvent mixtures of acetonitrile and
the salt 6 (Fig. 4). Due to the differences in reactivity of the
benzylic species 2 and 3, different temperatures were required
for practical monitoring of the reaction. Whilst this limits
comparisons somewhat the trends in the dependence of k2
on the amount of the salt 6 in the reaction mixture can still be
compared.
These kinetic studies show that there is a general increase in
the rate constant of the reaction of species 2 and 7b as the mole
fraction of the salt 6 in the reaction mixture increases. Between
w6 0 and ca. 0.6 there is a ca. five-fold increase in the rate
constant observed; increasing the proportion of the salt 6 further
does not significantly affect the rate constant. It should be noted
that the trend in the rate constant on varying the concentration
of salt 6 is similar to that observed for the reaction between
pyridine 1 and benzyl chloride 2 (Fig. 1).14
There are notable differences in the mole fraction depen-
dence of the rate constant of the triphenylphosphine 7b system
with different benzyl halides 2 and 3 (Fig. 3). These differences
suggest a change in the interactions of the solvent with the
components along the reaction coordinate when the halide
leaving group for this SN2 reaction is varied.
The above demonstrates that, in a general sense, ionic liquid
solvent effects are similar for SN2 reactions involving nitrogen
and phosphorus nucleophiles. However, it is clear that the
established framework for the SN2 reaction cannot simply be
applied to the triphenyl group 15 systems examined here. There
are differences in the mole fraction dependence of the rate
constants compared to the pyridine 1 system (Scheme 1).14
Further, changing the leaving group of the electrophile affected
the reaction outcome. These differences suggest that different
interactions, particularly between the solvent and the transition
state complex, are important for SN2 reactions involving triphenyl
group 15 nucleophiles.
Kinetic studies involving tributylamine 10a
Temperature dependent kinetic studies were undertaken
to further understand the origins of these solvent effects in
which w6 0, 0.20 and 0.78 (the acetonitrile case, w6 by which the
majority of the rate enhancement had occurred and the ionic
liquid case, respectively) were examined. These studies allowed
Due to the differences in the ionic liquid solvent effects
observed in the reaction of the triphenyl group 15 nucleophiles
7 with benzyl halides 2 and 3 compared to pyridine 1, it was of
interest to examine similar reactions using tributyl group 15
nucleophiles 10. These trialkyl group 15 compounds allowed
further examination of the factors that may have caused the
ionic liquid effects observed above (particularly removal of any
quadrupole–cation interactions).
The reaction of tributylamine 10a with benzyl bromide 3
(Scheme 3) was investigated across a range of proportions of
the salt 6 in the reaction mixture (Fig. 5). It should be noted
that for the tributylamine 10a system (Scheme 3) a shift in the
position of equilibrium of the reaction was observed for w6
ca. 0.5 and above; potential manipulation of the position of the
equilibrium is not a solvent effect previously observed for SN2
processes in the presence of an ionic liquid.13–15,18,19
Table 3 The activation parameters for the reaction between triphenyl-
phosphine 7b and benzyl chloride 2 in acetonitrile (wX 0), a mixture of
acetonitrile and [Bmim][N(SO2CF3)2] 6 (w6 0.20) and [Bmim][N(SO2CF3)2] 6
(w6 0.78)
w6
DH‡ a/kJ molÀ1
DS‡ a/J KÀ1 molÀ1
0
0.20
0.78
51.7 Æ 1.9
59.7 Æ 1.5
60.3 Æ 1.9
À238 Æ 6
À201 Æ 4
À197 Æ 6
a
Errors reported are from the fit of the linear regression.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
Table 4 The activation parameters for the reaction between tributyl-
amine 10a and benzyl bromide 3 in acetonitrile (w6 0), a mixture of
acetonitrile and [Bmim][N(SO2CF3)2] 6 (w6 0.22) and [Bmim][N(SO2CF3)2]
6 (w6 0.87)
w6
DH‡ a/kJ molÀ1
DS‡ a/J KÀ1 molÀ1
0
0.22
0.87
25.6 Æ 2.2
32.9 Æ 1.6
28.8 Æ 1.9
À284 Æ 8
À256 Æ 5
À277 Æ 6
a
Errors reported are from the fit of the linear regression.
the same within error suggesting that the interactions for these
two cases are similar. This is consistent with the rate constants
observed (Fig. 5) but this trend is dissimilar to that observed for
the SN2 process.14,16,28
Fig. 5 The bimolecular rate constant of the reaction between tributyl-
amine 10a and benzyl bromide 3 in different mole fractions of the ionic
liquid 6 in acetonitrile at 45.0 1C.‡
The different ionic liquid solvent effects observed here for
the tributylamine 10a system (Scheme 3) are at odds with
the established framework for the ionic liquid solvent effect
for SN2 processes, suggesting that additional interactions are
significant in this system other than the cation–nucleophile
interaction. This difference is further highlighted by the change
in the position of equilibrium at w6 Z ca. 0.5, an effect not
previously observed for the SN2 reactions mentioned above.
The factors introduced previously, particularly the potential for
interactions of the salt 6 with the transition state complex to
become more significant, are clearly relevant in this case also.
As shown in Fig. 5 there is an initial increase in the rate
constant for this reaction at low mole fractions of salt 6, with
a two-fold increase in the rate constant occurring between w6 0
and ca. 0.2. This is then followed by a decrease in the rate
constant as the amount of the ionic liquid 6 is increased
further. At the highest mole fraction examined (w6 0.87) the k2
values are lower than those for the acetonitrile case making
these ionic liquid effects more akin to those seen previously for
unimolecular processes20,23,29 rather than the bimolecular pro-
cesses examined above and previously.13,15,16,26,28
As mentioned, an equilibrium was noted for all mixtures
w6 Z ca. 0.5, with the rate constant for the reverse process
increasing with increasing concentrations of the salt 6. This
change is likely due to an increase in the nucleophilicity of
bromide 3 in salt 6 compared to acetonitrile, as has been
reported previously.19 The effect of a decreasing rate constant
for the forward reaction, with an increasing rate constant for
the reverse reaction, as the mole fraction of the salt 6 increases
results in the position of equilibrium shifting towards the
reagents 3 and 10a (for more details see Fig. S5, ESI†).**
Once again, to further understand these solvent effects
temperature dependent kinetic studies were undertaken in
which w6 0, 0.22 and 0.87 (the acetonitrile case, w6 by which
the majority of the rate enhancement had occurred and the
ionic liquid case, respectively) were examined. These studies
allowed construction of Eyring plots (Fig. S6, ESI†) and calcula-
tion of the activation parameters in each solvent system
(Table 4). For the Eyring plot and the activation parameters of
the reverse process see the ESI† (Fig. S7 and Table S20, ESI†).
From these data there is an increase in the enthalpy and
entropy of activation when moving from w6 0 to 0.22, which is
consistent with the rate enhancement observed between these
cases. The activation parameters for the w6 0 and 0.87 cases are
Kinetic studies involving tributylphosphine 10b
Given the above data and the different ionic liquid solvent effect
observed for the tributylamine 10a system (Scheme 3), reaction of
the equivalent phosphorus nucleophile 10b with benzyl chloride
2 was investigated.†† Comparison between the tributyl nucleo-
philes 10 will however be limited given the halide effect observed
for the triphenylphosphine 7b reaction above (Scheme 2). The
rate constant for reaction between species 2 and 10b was
investigated in mixtures of the ionic liquid 6 and acetonitrile
(Fig. 6). It should be noted that no equilibrium was observed in
this case.
The rate constant of the reaction of species 2 and 10b
increases with increasing mole fractions of the ionic liquid 6.
This trend is similar to that for the reactions of both pyridine 1
and triphenylphosphine 7b with benzyl chloride 2 (Fig. 1
and 4).13,15,16,26,28 It should be noted that the rate constant
increase for this reaction was more consistent across the range
of mole fractions of the salt 6 used than in previous cases.
The origins of the rate enhancement observed in Fig. 6 were
investigated through temperature dependent kinetic studies at
w6 0 and 0.83 (the acetonitrile and ionic liquid cases, respectively)
allowing Eyring plots (Fig. S8, ESI†) and activation parameters
(Table 5) for each solvent system to be obtained.
** There may also be an equilibrium present in the triphenylamine 7a system
(Scheme 2) however, due to the slow rate of this reaction, monitoring beyond ca.
8% completion was not undertaken (see footnote ‡). Preliminary studies of
amines with varying substituents on the nucleophilic centre suggests that
substituents that are branched or with more than four carbon atoms result in
an equilibrium being established when reacting with bromide 3 in the salt 6. This
is likely due to the stability of the product in each case, with the more sterically
congested derivatives being less stable.
Upon changing the solvent from w6 0 to 0.83 the enthalpy of
activation remains the same within uncertainty, while the
entropy of activation increases (Table 5). These data are similar
†† Benzyl chloride 2 was chosen as the electrophile for this study as benzyl
bromide 3 provided a reaction too fast (t1/2 ca. s) to be monitored using
conventional NMR spectroscopic methods.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016

View Article Online
NJC
Paper
This is supported by the different effects observed when the
leaving group of the electrophile was varied.
Another significant outcome of this work is that the position
of the equilibrium for the reaction between tributylamine
10a and benzyl bromide 3 changes with the proportion of
[Bmim][N(SO2CF3)2] 6 in the reaction mixture. This is a novel
ionic liquid effect for the SN2 reaction and demonstrates that
ionic liquids may be able to influence not only the kinetics of a
process, but also its thermodynamics.
Overall, the work presented here has expanded the predic-
tive framework for ionic liquid solvent effects on SN2 reactions;
it has also highlighted additional factors that need to be
considered. In the future, investigation of the nature of these
additional, competing interactions and incorporation of more
nucleophilic heteroatoms into the predictive framework will
increase the potential utility of ionic liquids as solvents for
synthetic processes.‡‡
Fig. 6 The bimolecular rate constant of the reaction between tributyl-
phosphine 10b and benzyl chloride 2 in different mole fractions of the
ionic liquid 6 in acetonitrile at 15.0 1C.‡
Table 5 The activation parameters obtained for the reaction between
tributylphosphine 10b and benzyl chloride 2 in acetonitrile (w6 0) and
[Bmim][N(SO2CF3)2] 6 (w6 0.83)
w6
DH‡ a/kJ molÀ1
DS‡ a/J KÀ1 molÀ1
Acknowledgements
0
0.83
45.1 Æ 2.5
49.2 Æ 2.0
À215.7 Æ 8
À190.3 Æ 7
JBH acknowledges financial support from the Australian
Research Council Discovery Project Funding Scheme (Project
DP130102331). RRH acknowledges the support of the Australian
government through the receipt of an Australian Postgraduate
Award. The authors would like to acknowledge Sinead T. Keaveney
for advice in manuscript preparation and the NMR Facility
within the Mark Wainwright Analytical Centre at the University
of New South Wales for NMR support.
a
Errors reported are from the fit of the linear regression.
to the activation parameters observed for the reaction of
triphenylphosphine 7b and benzyl chloride 2 (Scheme 2) as
well as the pyridine 1 cases (Scheme 1), suggesting that the
main interaction causing the observed rate enhancements is
between the cation of the ionic liquid 6 and the nucleophile.
Changes in the activation parameters are different to the
tributylamine 10a system (Scheme 3) and there was no equili-
brium established for the tributylphosphine 10b system
(Scheme 3). This suggests different interactions are causing
the ionic liquid solvent effects observed between the group 15
tributyl nucleophiles 10, though with the caveat that changing
the leaving group may contribute in these cases.
Notes and references
1 K. N. Marsh, A. Deev, A. Wu, E. Tran and A. Klamt, Korean
J. Chem. Eng., 2002, 19, 357–362.
2 A. Maia, Mini-Rev. Org. Chem., 2011, 8, 178–185.
3 C. L. Hussey, Pure Appl. Chem., 1988, 60, 1763–1772.
4 M. B. Smith and J. March, March’s Advanced Organic Chem-
istry, Wiley-Interscience, New York, 2001.
5 C. Chiappe and D. Pieraccini, J. Phys. Org. Chem., 2005, 18,
275–297.
6 F. Endres and S. Z. El Abedin, Acc. Chem. Res., 2007, 40,
1106–1113.
7 E. T. Kimura, D. M. Ebert and P. W. Dodge, Toxicol. Appl.
Pharmacol., 1971, 19, 699–704.
8 G. Huddleston, D. Jonathan and R. Rogers, Chem. Commun.,
1998, 1765–1766.
9 R. A. Sheldon, Green Chem., 2005, 7, 267–278.
10 J. B. Harper and M. N. Kobrak, Mini-Rev. Org. Chem., 2006, 3,
253–259.
Conclusions
This work has probed how variation of the nucleophilic
heteroatom may affect ionic liquid solvent effects observed
for SN2 processes. Using nitrogen and phosphorus nucleophiles
resulted in similar ionic liquid solvent effects, notably for the
triphenyl group 15 nucleophiles, which is advantageous in
terms of being able to predict ionic liquid solvent effects on
reaction outcome on changing the nucleophilic centre.
Importantly the dependences of the rate constants on the
proportion of the salt 6 present are different to that observed 11 T. Welton, Chem. Rev., 1999, 99, 2071–2083.
for the pyridine system. The activation parameters suggest that
the rate enhancement is due to the key interaction proposed
previously (between the cation and the nucleophile). The
differences observed in the mole fraction dependence studies
‡‡ It might be suggested that variation of the components of the ionic liquid,
particularly the cation, may be used to elucidate these additional interactions.
Previous work from our group16 and that of Welton45 suggests that simple
methylation at the C-2 position of imidazolium is not likely to be sufficient to
suggest additional competing interactions of the ionic liquid, such
cause a change in the interactions and that a cation which was significantly more
as with the transition state, are also significant in these cases. sterically hindered was required.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016
New J. Chem.

View Article Online
Paper
NJC
12 J. P. Hallett and T. Welton, Chem. Rev., 2011, 111, 3508–3576. 29 S. T. Keaveney, B. P. White, R. S. Haines and J. B. Harper,
13 H. M. Yau, A. G. Howe, J. M. Hook, A. K. Croft and
J. B. Harper, Org. Biomol. Chem., 2009, 7, 3572–3575.
14 S. T. Keaveney and J. B. Harper, RSC Adv., 2013, 3,
15698–15704.
15 H. M. Yau, A. K. Croft and J. B. Harper, Faraday Discuss.,
2012, 154, 365–371.
16 E. E. L. Tanner, H. M. Yau, R. R. Hawker, A. K. Croft and
J. B. Harper, Org. Biomol. Chem., 2013, 11, 6170–6175.
17 M. J. Earle, S. P. Katdare and K. R. Seddon, Org. Lett., 2004,
6, 707–710.
Org. Biomol. Chem., 2016, 14, 2572–2580.
30 E. D. Kochly, N. J. Lemon and A. M. Deh-Lee, Molecules,
2016, 21, 60.
31 M. A. Crosio, N. M. Correa, J. J. Silber and R. D. Falcone,
Org. Biomol. Chem., 2016, 14, 3170–3177.
32 R. R. Hawker, J. Panchompoo, L. Aldous and J. B. Harper,
ChemPlusChem, 2016, 81, 574–583.
33 P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000,
39, 3772–3789.
34 S. T. Keaveney, K. S. Schaffarczyk McHale, R. S. Haines and
J. B. Harper, Org. Biomol. Chem., 2014, 12, 7092–7099.
35 S. T. Keaveney, R. S. Haines and J. B. Harper, Org. Biomol.
Chem., 2015, 13, 3771–3780.
36 I. W. Scanlan and D. T. Clark, J. Chem. Soc., 1973, 70,
1222–1232.
37 Y. Zhang, F. S. Tham, J. F. Nixon, C. Taylor, J. C. Green and
C. A. Reed, Angew. Chem., Int. Ed., 2008, 120, 3861–3864.
38 W. L. F. Armarego and C. L. L. Chai, Purification of Labora-
tory Chemicals, Butterworth-Heinemann, Oxford, 7th edn,
2013.
´
18 L. Crowhurst, N. L. Lancaster, A. Perez, M. Juan and
T. Welton, J. Am. Chem. Soc., 2004, 126, 11549–11555.
19 N. L. Lancaster and T. Welton, J. Org. Chem., 2004, 69,
5986–5992.
20 B. Y. W. Man, J. M. Hook and J. B. Harper, Tetrahedron Lett.,
2005, 46, 7641–7645.
21 I. Newington, J. M. Perez-Arlandis and T. Welton, Org. Lett.,
2007, 9, 5247–5250.
22 G. Ranieri, J. P. Hallett and T. Welton, Ind. Eng. Chem. Res.,
2008, 47, 638–644.
23 H. M. Yau, S. A. Barnes, J. M. Hook, T. G. A. Youngs, 39 K. Zou, A. Horhota, B. Yu, J. W. Szostak and
A. K. Croft and J. B. Harper, Chem. Commun., 2008,
3576–3578.
L. W. McLauglin, Org. Lett., 2005, 7, 1485–1487.
40 S. Steines, P. Wasserscheid and B. Drießen-Holscher,
¨
24 S. G. Jones, H. M. Yau, E. Davies, J. M. Hook, T. G. A. Youngs,
J. Prakt. Chem., 2000, 342, 348–354.
J. B. Harper and A. K. Croft, Phys. Chem. Chem. Phys., 2010, 41 L. Cammarata, S. G. Kazarian, P. A. Salter and T. Welton,
12, 1873–1878. Phys. Chem. Chem. Phys., 2001, 3, 5192–5200.
25 C. C. Weber, A. F. Masters and T. Maschmeyer, Angew. 42 C. G. Hanke, A. Johansson, J. B. Harper and R. M. Lynden-Bell,
Chem., Int. Ed., 2012, 51, 11483–11486. Chem. Phys. Lett., 2003, 374, 85–90.
26 E. E. L. Tanner, R. R. Hawker, H. M. Yau, A. K. Croft and 43 J. B. Harper and R. M. Lynden-Bell, Mol. Phys., 2004, 102,
J. B. Harper, Org. Biomol. Chem., 2013, 11, 7516–7521. 85–94.
27 H. M. Yau, S. T. Keaveney, B. J. Butler, E. E. L. Tanner, 44 C. Lim, S.-H. Kim, S.-D. Yoh, M. Fujio and Y. Tsuno,
M. S. Guerry, S. R. D. George, M. H. Dunn, A. K. Croft and
J. B. Harper, Pure Appl. Chem., 2013, 85, 1979–1990.
28 S. T. Keaveney, D. V. Francis, W. Cao, R. S. Haines and
J. B. Harper, Aust. J. Chem., 2014, 68, 31–35.
Tetrahedron Lett., 1997, 38, 3243–3246.
45 R. P. Matthews, I. J. Villar-Garcia, C. C. Weber, J. Griffith,
F. Cameron, J. P. Hallett, P. A. Hunt and T. Welton, Phys.
Chem. Chem. Phys., 2016, 18, 8608–8624.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2016