Ionic Liquids as Solvents for SN2 Processes. Demonstration of the Complex Interplay of Interactions Resulting in the Observed Solvent Effects

Article abstract of DOI:10.1002/cplu.201800510

Bimolecular nucleophilic substitution reactions between triphenylphosphine and benzylic electrophiles have been examined in an ionic liquid to probe interactions with species along the reaction coordinate. Trends in the rate constant were found on both varying the leaving group and the electronic nature of the aromatic ring. In all the cases considered, interactions between the components of the ionic liquid and the transition state were shown to be more significant in determining reaction outcome than previously observed for this class of reaction. This demonstrates the importance of considering interactions of the ionic liquid components with all species along the reaction coordinate when investigating the origin of ionic liquid solvent effects, along with how such effects might be exploited.

Full text of DOI:10.1002/cplu.201800510

Accepted Article
Title: Ionic liquids as solvents for SN2 processes. Demonstration of the
complex interplay of interactions resulting in the observed solvent
effects
Authors: Karin S Schaffarczyk McHale, Ronald S Haines, and Jason
Brian Harper
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPlusChem 10.1002/cplu.201800510
Link to VoR: http://dx.doi.org/10.1002/cplu.201800510
A Journal of
www.chempluschem.org

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
Ionic liquids as solvents for SN2 processes. Demonstration of the
complex interplay of interactions resulting in the observed
solvent effects
Karin S. Schaffarczyk McHale,[a] Ronald S. Haines[a] and Jason B. Harper*,[a]
Abstract: Bimolecular nucleophilic substitution reactions between
triphenylphosphine and benzylic electrophiles have been examined
in an ionic liquid to probe interactions with species along the reaction
coordinate. Trends in the rate constant were found on both varying
the leaving group and the electronic nature of the aromatic ring. In
all of the cases considered, interactions between the components of
the ionic liquid and the transition state were shown to be more
significant in determining reaction outcome than previously observed
for this class of reaction. This demonstrates the importance of
considering interactions of the ionic liquid components with all
species along the reaction coordinate when investigating the origin
of ionic liquid solvent effects, along with how such effects might be
exploited.
solvents for synthetic purposes is hindered by the lack of a
general understanding as to how ionic liquids affect reaction
outcome compared to conventional organic solvents.[14]
Given this current limitation to the application of these
solvents, there has been a recent focus on quantitatively
analysing ionic liquid solvent effects on well-known reactions.
Particularly, detailed kinetic analyses and determination of the
microscopic interactions that are responsible for ionic liquid
solvent effects[15] have paved the way for initial design of
solvents in efforts to control reaction outcome.[16] An example of
the latter was a study focussing on the bimolecular nucleophilic
substitution reaction (SN2) between pyridine 1 and benzyl
bromide 2d (Scheme 1).[17] For this process, the bimolecular rate
constant (k2) increased with the amount of the representative
ionic
liquid
1-butyl-3-methylimidazolium
([Bmim][N(SO2CF3)2], 4)
bis(trifluoromethanesulfonyl)imide
Introduction
present in the reaction mixture, with the rate constant
enhancement due to an entropically favourable interaction
between the cation of the ionic liquid 4 and the lone pair of
pyridine 1.[17-18] Understanding this interaction allowed a degree
of control of reaction outcome through rational selection of the
components of the ionic liquid for this[19] and related nucleophilic
processes.[20] It should be noted that in related systems
involving an imidazole as the nucleophile, there the authors
focused primarily on ionic liquid-transition state interactions
being responsible for the ionic liquid solvent effect observed,
though no activation parameter data was presented.[21]
Ionic liquids are defined as salts with melting points below 100
ºC,[1] with their low melting points frequently due to the bulky,
asymmetric and charge diffuse nature of the constituent ions.
Being a solvent composed of charged species, ionic liquids have
negligible vapour pressures and low flammability, compared to
organic solvents,[1a,
thus making them ‘safer’[3] and
1c, 2]
4]
‘greener’.[2,
There are a plethora of cation and anion
combinations possible, resulting in ionic liquids with a wide
range of physicochemical properties;[2, 4b, 5] for this reason, they
are often described as ‘designer solvents’.[2, 4b, 6]
Given the potential benefits, ionic liquids have been
considered for a broad variety of applications; these range from
biomass pretreatment[7] through energy storage[8] and batteries[9]
to biomedical applications[10] and separation technologies.[11] In
all of these cases, along with analyses of potential impacts of
these salts on the environment,[12] a key feature is the ability to
choose an ionic liquid with the properties appropriate for the
application. If ionic liquids are to be used as solvents for
preparative chemistry, the equivalent understanding is required
in terms of choosing the appropriate ionic liquid to get the
desired reaction outcome.
Br
acetonitrile
N
Br
N
N
O
N
3
1
2d
O
4
N
S
S
CF3
F3C
O
O
Scheme 1. The bimolecular nucleophilic substitution reaction between
pyridine 1 and benzyl bromide 2d to give the pyridinium salt 3, which has been
examined in reaction mixtures containing the ionic liquid 4.[17-18]
Ionic liquids have been considered as solvents for a wide
variety of reaction types and the potential extension of 'solvent
design' has been shown in some cases where reaction outcome
can be manipulated depending upon the ionic liquid used.[13]
However, such reaction control is typically observed rather than
planned on a rational basis; the application of ionic liquids as
It is of interest to extend this understanding to other nucleophilic
centres. For the reaction between triphenylphosphine 5 and
each of the benzyl halides 2d and 6d (Scheme 2), it was shown
that there was a markedly different effect of the salt 4 on
reaction outcome depending on the nature of the leaving group;
in the case of the bromide 2d, there was an initial increase in the
rate constant on addition of ionic liquid 4, followed by a decrease
and subsequent increase as more salt was added.[18b]
effect of the leaving group was not observed for the equivalent
reactions involving pyridine 1. Importantly, these studies
This
[a]
K. S. Schaffarczyk McHale, Dr R. S. Haines, Assoc. Prof. J. B.
Harper
School of Chemistry, University of New South Wales,
UNSW Sydney, NSW 2052, Australia
E-mail: j.harper@unsw.edu.au
suggest that there are additional interactions between the ionic
liquid 4 and species along the reaction coordinate in cases
involving phosphorus nucleophiles that are not present in the
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
acetonitrile
XPh3P
X
PPh3
+
ionic liquid 4
R
R
9 X = Cl
10 X = Br
11 X = I
6 X = Cl
2 X = Br
7 X = I
5
12 X = OAc
8 X = OAc
Scheme 2. The bimolecular nucleophilic substitution reaction between
triphenylphosphine 5 and each of the benzyl derivatives 2, 6-8 to give the
corresponding salts 9-12, which have been examined in reaction mixtures
containing the ionic liquid 4 (a R = 3,5-Me-4-MeO, b R = MeO, c R = Me, d R
= H, e R = Br, f R = CF3, g R = NO2).
pyridine 1 case. These additional interactions compete with the
previously noted interaction between the cation of the ionic liquid
4 and the nucleophile,[18a] resulting in the difference in mole
fraction dependence of reaction outcome.
Figure 1. The normalised bimolecular rate constants (k2), relative to the
acetonitrile, for the reactions between triphenylphosphine 5 and either benzyl
iodide 7d (purple) or benzyl acetate 8d (green) in different proportions of the
ionic liquid [Bmim][N(SO2CF3)2] 4 in acetonitrile. Reported uncertainties are
the standard deviation of triplicate results compounded by division, some
uncertainties fall within the size of the markers used.
The work described herein focuses on elucidating these
additional interactions, through investigating further how varying
the electrophile varies the ionic liquid solvent effect observed in
these SN2 processes. The reactions considered are between
triphenylphosphine 5 and each of a range of benzyl derivatives
with different substituents and differing substituents on the
aromatic ring (Scheme 2) in the well described ionic liquid 4.
The use of this ionic liquid allows direct comparison to previous
work, though it is noted that changing the components of the
ionic liquid would also affect the interactions being considered;
for example, related studies on the reaction in Scheme 1 have
suggested increased interactions with the transition state in salts
containing more coordinating anions.[22]
While the shape of the plots are slightly different (the greatest
change occurs at low proportions of salt 4 in the acetate 8d case,
compared to a more linear trend for the iodide 7d) and the
extent of the rate constant decrease also varies (ca. 30% for the
iodide 7d and ca. 90% for the acetate 8d), these differences are
small when compared to the equivalent solvent effects for
reactions involving the halides 2d and 6d; in those cases, only
rate constant enhancements were observed.[23] (A normalized
plot showing the effects of the salt on reactions of all of the
benzyl derivatives 2d, 6d, 7d and 8d is given in the Supporting
Information, Figure S1.) This trend in the ionic liquid solvent
effect on related SN2 processes involving nitrogen nucleophiles
has been observed with different benzyl halides used as the
electrophiles.[17, 21a] In those studies, in the presence of an ionic
liquid reaction with benzyl iodide generally led to a decrease in
the rate constant of the reaction, the benzyl chlorides always
resulted in a rate enhancement and benzyl bromides yielded
comparable (if slightly higher) rate constants relative to the
molecular solvent, though it should be noted that only one
solvent composition was considered in these cases.
Results and Discussion
Effect of varying the leaving group on the electrophile
Initial focus was on extending the series that began with the
reactions of triphenylphosphine 5 with each of the halides 2d
and 6d.[18b] To do so, the iodide 7d and the acetate 8d were
considered. The iodide 7d forms a series with the halides 2d
and 6d, whilst the acetate 8d has a poor, though charge diffuse,
leaving group compared to, say, chloride.[14] The solvent effects
of the salt 4 on the reaction of triphenylphosphine 5 with
electrophiles 7d and 8d were initially examined by determining
the bimolecular rate constant (k2) for the reaction over a range of
solvent compositions of [Bmim][N(SO2CF3)2] 4 in acetonitrile.
These rate data are presented in Figure 1, having been
normalized to the value for the rate constant in acetonitrile to
allow comparison (equivalent plots for the absolute rate data are
available in the Supporting Information). It should be noted that
the order of the reactivity in acetonitrile is as would be expected
(8d < 6d < 2d < 7d).
Before continuing, it is worth noting that the
physicochemical properties of the mixture will change
dramatically across the range of solvent compositions used, and
such changes must be considered. For example, while the
viscosity of the mixture will change, and as a result diffusion of
solutes will change,[24] the reactions are not under diffusion
control so this does not affect the rate constants. Likewise, the
apparent polarity (based on interactions with a given probe
molecule) of the mixture would be expected to change, though
not greatly given the similar polarity measurements of
acetonitrile and the salt 4;[25] importantly, other solute-solvent
interactions would be expected to change and might be
expected to dominate in terms of solvent effects.[26] As such, it
is perhaps not surprising that it is difficult to ascribe the
Immediately apparent is that the effect of the salt 4 on the
reactions of species 7d and 8d is different to that observed
previously with the same nucleophile.[18b] There is a general
decrease in the bimolecular rate constant (k2) for the reaction of
both these electrophiles with triphenylphosphine 5 as the
amount of the ionic liquid 4 in the reaction mixture increases.
observed rate constant changes here (and earlier[18b]
) to
changes in polarity of the medium and that interactions between
the solute and the species along the reaction coordinate need to
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
Table 1. Activation parameters for the reaction between triphenylphosphine 5 and each of the benzyl derivatives 2d, 6d, 7d and 8d in mixtures of
[Bmim][N(SO2CF3)2] 4 in acetonitrile at the mole fraction specified
Acetonitrile (χ4 = 0)
Intermediate proportion of salt 4
High proportion of salt 4
Electrophile
[a]
[a]
[a]
∆H‡ / kJ mol-1 [a]
∆S‡ / J K-1 mol-1
χ4
∆H‡ / kJ mol-1 [a]
∆S‡ / J K-1 mol-1
χ4
∆H‡ / kJ mol-1 [a]
∆S‡ / J K-1 mol-1
6d[b
51.7 ± 1.9
-237.9 ± 5.5
0.20
59.7 ±1.5
-201.3 ± 4.4
0.78
60.3 ± 1.9
-197.2 ± 5.8
2d[b]
7d
44.8 ± 3.7
52.5 ± 1.1
71.7 ± 2.6
-210 ± 13
-160.3 ± 3.9
-206.4 ± 7.9
0.36
0.30
0.29
57.1 ±1.5
41.9 ±1.7
43.4 ± 3.7
-163.1 ± 5.1
-203.2 ± 6.1
-300 ± 11
0.93
0.96
0.90
57.2 ± 1.5
47.0 ± 2.0
38.0 ± 2.9
-162.8 ± 5.3
-189.6 ± 6.9
-325.2 ± 8.7
8d
[a] Errors reported are from the fit of the linear regression. [b] Data reproduced from Schaffarczyk McHale et al.[18b]
be considered. Further, whilst changes in the nanostructure of
these solvent have been reported on changing solvent
composition, in related bimolecular condensation reactions
nanodomains of the solvent did not significantly affect reaction
outcome and consideration of interactions involving components
of the ionic liquid are more important.[27]
studies were undertaken and rate constants determined at key
solvent compositions; acetonitrile (χ4 = 0), an intermediate
solvent composition by which the majority of the rate decrease
had occurred (χ4 ca. 0.3), and 'neat' ionic liquid (χ4 ca. 0.9,
where the ionic liquid is diluted by reagents only). Using an
Eyring plot (see Figures S15 and S17, Supporting Information)
the activation parameters for the reactions of triphenylphosphine
5 with each of the electrophiles 7d and 8d were determined for
each solvent system specified; these data are presented in
Table 1, with previously reported data for reactions involving the
halides 2d and 6d included for comparison.[18b]
The activation parameters for reaction of each of the
species 7d and 8d with triphenylphosphine in acetonitrile were
as would be expected. The values for case 7d are similar to the
other halide cases considered, with a balance of enthalpic and
entropic effects being responsible for the order of rate constants
observed in molecular solvent. The enthalpy of activation for
reaction of the acetate 8d with nucleophile 5 is notably larger
than the other cases considered, accounting for the significant
rate constant decrease observed in that case.
For both the iodide 7d and the acetate 8d cases, addition
of the ionic liquid 4 results in a decrease in both of the activation
parameters of the reaction with triphenylphosphine 5 compared
to the acetonitrile case (irrespective of the amount of salt 4
added). That is, the rate constant for the reaction decreases on
addition of the ionic liquid 4 are the result of an entropic cost
outweighing an enthalpic benefit. This outcome is different to
that observed in the cases of the chloride 6d and the bromide 2d,
where rate constant increases were the result of a dominant
beneficial entropic effect;[18b] these data were consistent with the
pyridine 1 case[17] and suggest interactions between reagents
and the salt 4.[18a]
Whilst previously the interactions of the cation of the salt 4
with the nucleophile has been used to explain (at least in part)
changes in rate constant for reactions involving both pyridine
1[17-18] and triphenylphosphine 5,[18b] these interactions cannot be
dominant in the work described here since the observed rate
constant effect is different. The leaving group ability of the
benzyl substituent also cannot be solely responsible; the best
(iodide) and worst (acetate) give the same effect. It is perhaps
useful rather to consider the nature of the charge development
in the transition state (Figure 2); chloride would be the ‘hardest’
or most charge dense leaving group, iodide and acetate being
the ‘softest’ or most charge diffuse, and bromide sits somewhere
in between.[14] Such differences might be expected to affect
interactions with the components of the ionic liquid 4, and hence
affect reaction outcome.
δ+
δ-
H H
Ph3P
X
R
Figure 2. The transition state for the reaction shown in Scheme 2 – sites of
charge development represent key sites for potential interaction with the ionic
liquid 4. Note that this representation is general; the extent of charge
development along with the extent to which bond breakage and bond cleavage
occur (and whether it is symmetrical) will depend on the nature of the reacting
species.
A decrease in both activation parameters for reactions
involving the iodide 7d and the acetate 8d could be due to the
dominant interactions involving either stabilisation of, and
organisation about, the transition state complex or
a
destabilisation of, and lesser organisation about, the reagents, in
the ionic liquid 4 than in the acetonitrile. The former argument is
considered more likely given the charge development in the
In order to further elucidate the trends in rate constant seen on
changing solvent composition, temperature dependent kinetic
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
transition state (Figure 2) and the potential for either the cation
or the anion of the ionic liquid to interact with that charge
development.[28] Considering the interactions implicit from
activation parameters for the bromide 2d and chloride 6d
cases[18b] adds further weight to this argument. Significantly, the
dominance of these interactions (which have been demonstrated
by the difference in effects of the salt in the bromide 2d and
chloride 6d cases) is novel for an SN2 process with a neutral
nucleophile, though some previous studies have suggested
similar interactions result in rate constant enhancement.[21]
Rather, such activation parameter changes have been seen
before for SN1 processes where a greater degree of charge
development in the transition state would be expected.[29] For the
reactions studied here these results demonstrate that such
transition state-ionic liquid interactions must be taken into
account when considering solvent effects in these systems
before using ionic liquids to promote the rate of reaction in a
synthetic scenario.
reaction mixture for the substrates 2a-c,e-g are presented as
Figures S8-S13, Supporting Informattion. In order to allow
viewing of all of the rate data together, they are presented in
Figure 3 having been normalized to the value for the rate
constant in acetonitrile.
Figure 3. The normalised bimolecular rate constants (k2) for the reaction
between triphenylphosphine 5 and each of the benzyl bromides 2a-g (light
blue 2a, red 2b, blue 2c, green 2d, orange 2e and purple 2f, pink 2g) in
different proportions of the ionic liquid [Bmim][N(SO2CF3)2] 4 in acetonitrile at
24.4 ˚C. Reported uncertainties are propagated from the standard deviation of
triplicate results and some fall within the size of the markers used.
The changes in activation parameters on moving from a
low proportion of salt
4
in the reaction mixture to
[Bmim][N(SO2CF3)2] 4 diluted only by reagents are small (if
measurable) for the electrophiles 7d and 8d. This small change
suggests only a slight change in the balance of the interactions
responsible for the solvent effect of the ionic liquid 4 once a
given proportion of salt had been added, despite the changes in
rate constant that are observed.[30]
It is worth briefly returning to a comment earlier regards
the nature of the leaving group; differences in the degree of
change density on the leaving group, might be expected to affect
interactions with the components of the ionic liquid 4, and hence
affect reaction outcome. The trends observed in the activation
parameters support this argument, with the most charge diffuse
systems (the iodide 7d and acetate 8d cases) showing the
proposed transition state – ionic liquid interaction dominating,
the comparatively charge dense chloride 6d case being
dominated by ionic liquid – nucleophile interactions and the
benzyl bromide case 2d being the intermediate example where
While the shapes of these plots varied slightly with the substrate
2, there was typically a rate constant increase for each of the
reactions considered except the nitro case 2g (discussed further
below) with the amount of the ionic liquid 4 present. Further, in
general, the magnitude of the rate constant enhancement varied
with the electronic nature of the substituent; it was greater for
electron donating substituents and less for electron withdrawing
substituents.[32] The exception is the bromide 2a, which has
three electron donating groups but shows rate constant
enhancements comparable to the protio 2d and bromide 2e
cases across the range of mole fractions of the salt 4 considered.
This comparison, along with the relative rate constants for the
reactions in acetonitrile, suggests that the effects of the
substituents are not simply additive in this case.[33] Irrespective,
in this case, practical use of an ionic liquid might be evaluated
simply in terms of the electronic nature of the phenyl group with
electron rich systems gaining the most benefit.
These changes in rate constant enhancement are
consistent with what has been observed previously in reactions
of pyridine 1 with substituted benzyl bromides (albeit studied at
only one proportion of ionic liquid in the reaction mixture); the
greatest rate constant enhancements were seen with electron
donating groups.[17] The same trends have been seen in the
reaction of N-methylimidazole with benzyl chlorides, once again
studied at only one proportion of ionic liquid in the reaction
mixture.[21a] Each of these reports also used Hammett analysis
to compare the nature of the transition state for different
substrates. Equivalent analyses were carried out for each
composition of the reaction mixture (see Figures S23-S31,
Supporting Information) at the two solvation extremes shown in
Figure 4.
both interactions contribute significantly.
suggests that a charge diffuse transition state is beneficial for
ionic liquid transition state interactions, though these
Importantly, this
–
interactions might not necessarily result in an improvement in
reaction outcome. In fact, ionic liquid solvents should only be
considered in cases where such interactions might be expected
to be limited.
Effect of varying the sigma donating ability on the
electrophile
In order to further understand how variation of the electronics of
the electrophile affects ionic liquid solvent effects, the reactions
of triphenylphosphine 5 with each of the benzyl bromides 2a-c,e-
g across a range of solvent compositions of the salt 4 were
considered. The substituents on the phenyl ring in the
electrophiles 2a-g cover a wide range of electronic properties
(as assessed by, for example, Hammett σ values[31]) and thus
might be expected to systematically vary the extent of charge
development in the transition state of these reactions. The plots
of rate constant against the proportion of the salt 4 in the
The curved nature of the plots shown in Figure 4 (and in
Figures S23-S31) is consistent with that seen previously[17, 21a]
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
Table 2. Changes in activation parameters for the reaction between
triphenylphosphine 5 and each of the benzyl derivatives 2b-f in mixtures of
[Bmim][N(SO2CF3)2] 4 in acetonitrile between the mole fractions specified.
Changes in activation enthalpy are given in kJ mol-1, changes in activation
entropy are given in
responsible for the rate constant changes seen.
J
K-1 mol-1. Parameters highlighted in red are
Change in χ4
0 è 0.3
Change in χ4
0.3 è 0.9
Change in χ4
0 è 0.9
Reagent
∆(∆H‡) = -2.4 ±
3.2
∆(∆H‡) = 9.8 ± 2.9
∆(∆S‡) = 31 ± 9
∆(∆H‡) = 7.4 ± 3.5
∆(∆S‡) = 30 ± 12
2b
∆(∆S‡) = -1 ± 10
∆(∆H‡) = 5.2 ± 1.3
∆(∆S‡) = 23 ± 4
∆(∆H‡) = -2.6 ± 1.9
∆(∆S‡) = -9.4 ± 5.9
∆(∆H‡) = 2.6 ± 1.8
∆(∆S‡) = 14 ± 6
2c
∆(∆H‡) = 12 ± 4
∆(∆S‡) = 47 ± 14
∆(∆H‡) = 0.1 ± 2.1
∆(∆S‡) = 0.3 ± 7.9
∆(∆H‡) = 12 ± 4
∆(∆S‡) = 47 ± 14
Figure 4. Hammett plots constructed from the relative rate data obtained for
the bimolecular nucleophilic substitution reaction between triphenylphosphine
5 and each of the benzyl bromides 2b-g carried out in acetonitrile (blue, χX 0)
and [Bmim][N(SO2CF3] 4 (red, χ4 ca. 0.95). Uncertainties are propagated from
the standard deviation of triplicate results for the k2 values obtained. Some
uncertainties fall within the size of the markers.
2d[b]
2e
∆(∆H‡) = -2.1 ± 3.2
∆(∆S‡) = -9 ± 11
[c]
[c]
-
-
∆(∆H‡) = 1.9 ± 2.1
∆(∆S‡) = 8.3 ± 7.1
[c]
[c]
2f
-
-
and can be ascribed to the change from an 'open' or 'loose'
transition state for electron donating groups to a 'closed' or 'tight'
transition state for electron withdrawing groups. In general, on
increasing the proportion of the ionic liquid 4 in the reaction
mixture, the proportion of the plot that has a large slope
[a] Errors reported are compounded from the fit of the linear regression. [b]
Calculated based on data from Schaffarczyk McHale et al.[18b] [c] As
discussed in text, activation parameters were not determined at intermediate
proportions of the salt 4 in the reaction mixture.
increases, consistent with
development being favoured by the salt 4. This is exemplified in
Figure 4, noting the large uncertainties as
compounding errors and the logarithmic scale.[34]
It should be noted that during this analysis, the nitro case
2g was a particular outlier, being faster than might be expected
at low proportions of the salt 4 in the reaction mixture. It was
considered that alternative mechanisms may also be
contributing, likely a radical mechanism based on literature
precedent[35] as well as EPR studies (Figure S34).[36] As such, it
was not considered during subsequent analysis.
All of the above analyses indicate that changing the nature
of the substituent on the phenyl ring changes the ionic liquid
effects on the substitution process. Once again, in order to
determine the microscopic origins of these effects, temperature
dependent kinetic studies were undertaken at the same
compositions of the reaction mixture considered above (χ4 = 0,
ca. 0.3 and ca. 0.9, where the ionic liquid is diluted by reagents
only). From these studies Eyring plots were constructed (Figures
S19-S22) and the activation parameters determined for each
solvent system (Table S25); while the absolute activation
parameters are necessary, it is the changes in these parameters
on changing the solvent composition that requires discussion
and these are presented in Table 2. Importantly, the activation
parameter responsible for the rate constant change has been
highlighted where it is clear. It should be noted that the given
the similarity in rate constants observed across the range of
solvent compositions for reactions involving species 2e,f,
activation parameters were determined only at the extreme
solvent compositions (and these were found to be virtually
identical).[37] Immediately apparent from the data in Table 2 is
a
greater degree of charge
that the rate constant change seen for the reaction of
triphenylphosphine 5 with each of the species 2b-f is, where
measurable, due to an increase in the entropy of activation.
This outcome suggests that the dominant interaction involving
the ionic liquid is with a starting material, likely the nucleophile 5
based on previous studies. Several things must be noted
however.
a result of
The proportion of the ionic liquid 4 in the reaction mixture
required to see the dominant interaction varies; it is clearly
observable by χ4 ca. 0.3 in the case of substrates 2c,d but not in
the case of substrate 2b. This result suggests that other
interactions, likely with the transition state, are significant at low
proportions of salt 4 in the reaction mixture in the latter case,
consistent with a greater degree of charge development in (and
perhaps the more ‘open’ nature of) the transition state in that
example. Immediately related to this is that ionic liquid -
transition state interactions are more significant at the higher
proportions of the salt 4 in the reaction mixture for cases 2c and
2d. The former case is shown here with negative changes in the
activation parameters on going from χ4 = 0.3 è χ4 = 0.9; the
case of the protio example 2d is more complex though has been
discussed extensively previously.[18b] Irrespective, the presence
of ionic liquid – transition state interactions, their balance with
ionic liquid – starting material interactions and the importance of
the amount of salt 4 required for them to be observed, is
highlighted.
Unlike the changes seen when the leaving group on the
electrophile was changed, no case is seen where interactions
between the transition state and the ionic liquid dominate (as
interpreted by decreases in the activation parameters). It is
worth noting that changes in these parameters are very small (if
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
precursor. Full details for this preparation can be found in the
Supporting Information.
measurable) in the cases of the electron withdrawing
substituents bromo and trifluoromethyl on the phenyl ring
(reagents 2e and 2f, respectively), suggesting that transition
state interactions might be more significant in these cases
though it is not clear how these cases (which are the 'closed' or
'tight' transition state) necessarily correlate with the charge
diffuse nature of the leaving group in the iodide 7d and acetate
7d cases. Once again though, the importance of interactions
involving the components of the ionic liquid 4 and the transition
state is demonstrated and merits consideration when using ionic
liquids as solvents for this reaction and others where these types
of interactions are also present.
Reaction mixtures for the kinetic studies contained
triphenylphosphine 5 and at least a ten-fold excess of the
electrophile at a given proportion of the ionic liquid 4, with the
remainder of the solvent being acetonitrile. The reaction
progress was monitored using 31P{1H} NMR spectroscopy
following the signal corresponding to the starting material,
triphenylphosphine 5 at δ ca. -5. This allowed determination of
kobs for the reaction, with k2 obtained using the known
concentration of the reagent in excess, the electrophile. These
reactions were either monitored in situ whereby the reaction was
followed until > 95% completion or by initial rates methodology
where signals corresponding to both reagent and product are
monitored, with appropriate t1 times, until 7-10% completion. The
reactions were monitored using either a Burker Avance III 400 or
Burker Avance III 500 spectrometer equipped with either a TBI
or BBFO probe using ca. 0.5 mL of the reaction mixture in a 5
mm NMR tube. Results were shown to be reproducible between
different probes and spectrometers. Activation parameters,
where they appear, were determined in a similar manner to that
described above except that the bimolecular rate constants were
determined at four temperatures across a range of at least 30 °C
and then fitted to the bimolecular Eyring equation.[46]
Conclusions
The work described has shown that on varying the nature of the
electrophile in a bimolecular substitution process, that the
solvent effects of an ionic liquid change. All of the kinetic
analyses presented show that these changes in ionic liquid
solvent effects are due to the extent of interaction of the
components of the ionic liquid with the transition state of the
process. While suggested previously for related systems, this
work represents the first time for bimolecular substitution
processes that such interactions have been shown to be
dominant (for example, in the iodide 7d and acetate 8d cases)
and in which the proportion of the salt 4 present could be related
to the balance of interactions.
Further details for all of the kinetic analyses as well as the
temperatures, rate constants, and compositions of the stock
solutions (masses, concentrations of electrophile, mole fraction
of ionic liquid) for each system studied can be found in the
Supporting Information.
This observation[38] of the importance of ionic liquid –
transition state interactions is significant as any design of ionic
liquids to control reaction outcome in cases such as those Acknowledgements
demonstrated must take these interactions into account; this is
in contrast to previous examples where only the interaction of
the cation of the ionic liquid with the nucleophile was
considered.[16b] Rather, studies need to consider both
components of the ionic liquid and both interactions, akin to
studies on the SNAr reaction reported recently.[16a, 39] This being
said, there remains ample opportunity for the design of ionic
liquids to control these interactions and hence reaction outcome.
JBH acknowledges financial support from the Australian
Research Council Discovery Project Funding Scheme (Projects
DP130102331, DP180103682). KSSM acknowledges the
support of the Australian government through the receipt of an
Research Training Programme Stipend. The authors would like
to acknowledge the NMR Facility within the Mark Wainwright
Analytical Centre at the University of New South Wales for NMR
support.
Experimental Section
Keywords: bimolecular substitution • Hammett plot • ionic
liquids • kinetics • solvent effects
Triphenylphosphine 5 was recrystallized from acetonitrile.[40]
Benzyl bromides 2c-g were commercially available and were
[1] a) C. L. Hussey, Pure Appl. Chem. 1988, 60, 1763-1772;
b) C. Chiappe, D. Pieraccini, J. Phys. Org. Chem. 2005, 18,
275-297; c) F. Endres, S. Z. El Abedin, Acc. Chem. Res.
2007, 40, 1106-1113.
used without further purification. Benzyl bromides 2a[41],b[42]
,
benzyl iodides 7c, 7d and 7g,[43] and benzyl acetate 8d[44] were
synthesised through modified literature procedures, details of
which can be found in the Supporting Information. The ionic
liquid [Bmim][N(SO2CF3)2] 4 was prepared according to literature
methods;[45] 1-bromobutane was added to 1-methylimidazole to
yield the bromide salt precursor which was then treated with
[2] K. N. Marsh, A. Deev, A. Wu, E. Tran, A. Klamt, Kor. J.
Chem. Eng. 2002, 19, 357-362.
[3] E. T. Kimura, D. M. Ebert, P. W. Dodge, Toxicol. Appl.
Pharmacol. 1971, 19, 699-704.
[4] a) R. A. Sheldon, Green Chem. 2005, 7, 267-278; b) A.
Maia, Mini-Rev. Org. Chem. 2011, 8, 178-185.
[5] M. T. Clough, C. R. Crick, J. Gräsvik, P. A. Hunt, H.
Niedermeyer, T. Welton, O. P. Whitaker, Chem. Sci. 2015,
6, 1101-1114.
lithium bis(trifluoromethanesulfonyl)imide through
a
salt
metathesis reaction to yield [Bmim][N(SO2CF3)] 4. Before use,
the ionic liquid 4 was dried under reduced pressure (< 0.1 mbar)
at room temperature and was found to contain < 200 ppm water
using Karl Fischer titrimetry. Ion chromatography showed < 1%
bromide content, confirming the absence of the bromide salt
[6] M. Freemantle, Chem. Eng. News 1998, 76, 32-37.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
[7] A. George, A. Brandt, K. Tran, S. M. S. N. S. Zahari, D.
Klein-Marcuschamer, N. Sun, N. Sathitsuksanoh, J. Shi, V.
Stavila, R. Parthasarathi, S. Singh, B. M. Holmes, T.
Welton, B. A. Simmons, J. P. Hallett, Green Chem. 2015,
17, 1728-1734.
[25] T. Welton, in Ionic Liquids in Synthesis, 2nd ed. (Eds.: P.
Wasserscheid, T. Welton), Wiley-VCH, Weinheim,
Germany, 2008, pp. 130-175.
[26] L. Crowhurst, P. R. Mawdsley, J. M. Perez-Arlandis, P. A.
Salter, T. Welton, Phys. Chem. Chem. Phys. 2003, 5,
2790-2794.
[8] D. R. MacFarlane, M. Forsyth, P. C. Howlett, M. Kar, S.
Passerini, J. M. Pringle, H. Ohno, M. Watanabe, F. Yan, W. [27] S. T. Keaveney, T. L. Greaves, D. F. Kennedy, J. B.
Zheng, S. Zhang, J. Zhang, Nat. Rev. Mater. 2016, 1,
15005.
[9] Q. Yang, Z. Zhang, X.-G. Sun, Y.-S. Hu, H. Xing, S. Dai,
Chem. Soc. Rev. 2018, 47, 2020-2064.
[10] K. S. Egorova, E. G. Gordeev, V. P. Ananikov, Chem. Rev.
2017, 117, 7132-7189.
[11] A. Berthod, M. J. Ruiz-Ángel, S. Carda-Broch, J.
Chromatogr. A 2018, 1559, 2-16.
Harper, J. Phys. Chem. B 2016, 120, 12687-12699.
[28] Whilst their may be synergistic interactions involving the
acetonitrile and the ionic liquid, the simplest argument is
that the changes in interactions in solution on increasing
the proportion of ionic liquid to the reaction mixture are
due to the introduction of interactions involving the ionic
liquid.
[29] a) B. Y. W. Man, J. M. Hook, J. B. Harper, Tetrahedron
Lett. 2005, 46, 7641-7645; b) H. M. Yau, S. A. Barnes, J.
M. Hook, T. G. A. Youngs, A. K. Croft, J. B. Harper, Chem.
Commun. 2008, 3576-3578; c) S. T. Keaveney, B. P.
White, R. S. Haines, J. B. Harper, Org. Biomol. Chem.
2016, 14, 2572-2580.
[30] The comparatively small changes in the rate constants
observed means that activation parameters do not
necessarily distinguish the origin of the rate constant
change, particularly given the uncertainties in activation
parameter data.
[31] N. S. Isaacs, Physical Organic Chemistry, 2nd ed.,
Addison Wesley Longman Limited, Harlow, 1998.
[32] Similar mole fraction dependence data is available for the
chlorides 6b,d,g and the iodides 7b,d,g. The trends in
data are similar in all cases (see Supporting Information,
Figures S2-S6).
[12] M. Amde, J.-F. Liu, L. Pang, Environ. Sci. Technol. 2015,
49, 12611-12627.
[13] a) T. Welton, Chem. Rev. 1999, 99, 2071-2083; b) P.
Wasserscheid, W. Keim, Angew. Chem., Int. Ed. 2000, 39,
3772-3789; c) J. P. Hallett, T. Welton, Chem. Rev. 2011,
111, 3508-3576; d) R. L. Vekariya, Journal of Molecular
Liquids 2017, 227, 44-60.
[14] M. B. Smith, J. March, March's Advanced Organic
Chemistry, Wiley-Interscience, New York, 2001.
[15] a) H. M. Yau, S. T. Keaveney, B. J. Butler, E. E. L. Tanner,
M. S. Guerry, S. R. D. George, M. H. Dunn, A. K. Croft, J.
B. Harper, Pure Appl. Chem. 2013, 85, 1979-1990; b) S. T.
Keaveney, R. S. Haines, J. B. Harper, in Encyclopedia of
Physcial Organic Chemistry, Vol. 2 (Ed.: U. Wille), Wiley,
2017; c) A. Gilbert, R. S. Haines, J. B. Harper, in Elsevier
Reference Modules in Chemistry, Molecular Sciences and
Chemical Engineering (Ed.: J. Reedijk), Elsevier, Waltham,
MA, 2018, doi.org/10.1016/B978-0-12-409547-2.14212-X;
d) R. R. Hawker, R. S. Haines, J. B. Harper, Adv. Phys.
Org. Chem. 2018, 52, in press.
[16] a) R. R. Hawker, R. S. Haines, J. B. Harper, Org. Biomol.
Chem. 2018, 16, 3453-3463; b) R. R. Hawker, R. S.
Haines, J. B. Harper, Chem. Commun. 2018, 54, 2296-
2299; c) R. R. Hawker, R. S. Haines, J. B. Harper, J. Phys.
Org. Chem., In press.
[33] a) Given this lack of additivity of substituent effects, this
reagent was not used in subsequent Hammett analysis; b)
Once again, the physicochemical properties of the solvent
system will change with the composition of the mixture,
however they cannot be used to explain the changes seen.
[34] Similar but limited data for the chloride 6 and iodide 7
series are presented in the Supporting Information, Figures
S14-S16 and Tables S26-S27.
[35] C. Costentin, P. Hapiot, M. Médebielle, J.-M. Savéant, J.
[17] H. M. Yau, A. G. Howe, J. M. Hook, A. K. Croft, J. B.
Harper, Org. Biomol. Chem. 2009, 7, 3572-3575.
[18] a) H. M. Yau, A. K. Croft, J. B. Harper, Faraday Discuss.
2012, 154, 365-371; b) K. S. Schaffarczyk McHale, R. R.
Hawker, J. B. Harper, New J. Chem. 2016, 40, 7437-7444.
[19] E. E. L. Tanner, H. M. Yau, R. R. Hawker, A. K. Croft, J. B.
Harper, Org. Biomol. Chem. 2013, 11, 6170-6175.
[20] a) S. T. Keaveney, K. S. Schaffarczyk McHale, R. S.
Haines, J. B. Harper, Org. Biomol. Chem. 2014, 12, 7092-
7099; b) S. T. Keaveney, R. S. Haines, J. B. Harper, Org.
Biomol. Chem. 2015, 13, 3771-3780.
[21] a) R. Bini, C. Chiappe, C. S. Pomelli, B. Parisi, J. Org.
Chem. 2009, 74, 8522-8530; b) X. Mu, N. Jiang, C. Liu, D.
Zhang, J. Phys. Chem. A 2017, 121, 1133-1139.
[22] S. T. Keaveney, D. V. Francis, W. Cao, R. S. Haines, J. B.
Harper, Aust. J. Chem. 2014, 68, 31-35.
[23] It should be noted that due to vast differences in absolute
rate constants, these mole fraction dependence data were
determined at different temperatures. Whilst changing
temperature might be expected to subtly change the
magnitude of the effects, the gross effects seen would be
the same.
[24] S. T. Keaveney, K. S. Schaffarczyk Mc Hale, J. W.
Stranger, B. Ganbold, W. S. Price, J. B. Harper,
ChemPhysChem 2016, 17, 3853–3862.
Am. Chem. Soc. 1999, 121, 4451-4460.
[36] It should be noted that there was no change in the rate
constant on addition of radical initiators or radical inhibitors.
Irrespective, the uncertainty in the underlying mechanism
meant it was not considered further
[37] Similar but limited data for the chloride 6 and iodide 7
series are presented in the Supporting Information, Figures
S14-S16 and Tables S26-S27.
[38] The latter effect has been shown for nucleophilic aromatic
substitution processes, see references 16a and 39.
[39] R. R. Hawker, M. J. Wong, R. S. Haines, J. B. Harper, Org.
Biomol. Chem. 2017, 15, 6433-6440.
[40] W. L. F. Armarego, C. L. L. Chai, Purification of Laboratory
Chemicals, Seventh ed., Butterworth-Heinemann, Oxford,
2013.
[41] a) S. M. Johnson, S. Connelly, I. A. Wilson, J. W. Kelly, J.
Med. Chem. 2008, 6348-6358; b) P. Dhankher, T. D.
Sheppard, Synlett 2014, 381-384; c) B. J. Littler, M.
Aizenberg, N. B. Ambhaikar, T. A. Blythe, T. T. Curran, V.
Dvornikovs, Y. C. Jung, V. Jurkauskas, E. C. Lee, A. R.
Looker, H. Luong, T. A. Martinot, D. B. Miller, B. J.
Neubert-Langille, P. A. Otten, P. J. Rose, P. L. Ruggiero,
Org. Process Res. Dev. 2015, 270-283; d) W. Li, C. Yang,
G.-L. Gao, W. Xia, Synlett 2016, 1391-1396.
[42] K. K. Mayer, S. Prior, W. Wiegrebe, Montash. Chem.
Chem. Mon. 1986, 511.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
[43] C. T. Hoang, V. Alezra, R. Guillot, C. Kouklovsky, Org. Lett.
2007, 9, 2521-2524.
[44] S. Zhang, X. Zhang, X. Ling, C. He, R. Huang, J. Pan, J. Li,
Y. Xiong, RSC Adv. 2014, 4, 30768-30774.
Kazarian, P. A. Salter, T. Welton, Phys. Chem. Chem.
Phys. 2001, 3, 5192-5200; c) S. V. Dzyuba, R. A. Bartsch,
ChemPhysChem 2002, 3, 161-166.
[46] H. Eyring, J. Chem. Phys. 1935, 3, 107-115.
[45] a) S. Steines, P. Wasserscheid, B. Drießen-Hölscher, J.
Prakt. Chem. 2000, 342, 348-354; b) L. Cammarata, S. G.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800510
ChemPlusChem
FULL PAPER
Entry for the Table of Contents
FULL PAPER
The interactions responsible for ionic
liquid effects vary with the amount of
ionic liquid in the reaction mixture, as
does the microscopic origin of the
effects observed.
Karin S. Schaffarczyk McHale, Ronald
S. Haines] and Jason B. Harper*
acetonitrile
ionic liquid
X
XPh3P
PPh3
+
Page No. – Page No.
X = Cl
Ionic liquids as solvents for SN2
processes. Demonstration of the
complex interplay of interactions
resulting in the observed solvent
effects
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm))
X = Br
X = I
X = OAc
This article is protected by copyright. All rights reserved.