Ionicity and proton transfer in protic ionic liquids

Article abstract of DOI:10.1039/c0cp00239a

Proton transfer in protic ionic liquids is poorly understood. Some acid/base proton transfer reactions do not proceed to the extent that is expected from ΔpK_a^aq data from aqueous solutions, yet some do. In this work we have investigated protic ionic liquids obtained by proton transfer from a common acid, acetic acid, to a range of amine bases of similar pK_a^aq values. Probe indicator observations, transport property data allowing the construction of Walden plots and computational studies all suggest that there is a clear distinction between the behaviour of simple primary vs. tertiary amines, the proton transfer being more complete in the former case than the latter. The origins of this seem to be related to the hydrogen bonding ability of the ammonium ions in providing a good solvating environment for the ions produced by the proton transfer.

Full text of DOI:10.1039/c0cp00239a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Ionicity and proton transfer in protic ionic liquids
Jelena Stoimenovski,* Ekaterina I. Izgorodina and Douglas R. MacFarlane
Received 16th April 2010, Accepted 8th June 2010
DOI: 10.1039/c0cp00239a
Proton transfer in protic ionic liquids is poorly understood. Some acid/base proton transfer
aq
reactions do not proceed to the extent that is expected from DpK
a
data from aqueous
solutions, yet some do. In this work we have investigated protic ionic liquids obtained by proton
aq
transfer from a common acid, acetic acid, to a range of amine bases of similar pK
a
values.
Probe indicator observations, transport property data allowing the construction of Walden plots
and computational studies all suggest that there is a clear distinction between the behaviour of
simple primary vs. tertiary amines, the proton transfer being more complete in the former case
than the latter. The origins of this seem to be related to the hydrogen bonding ability of the
ammonium ions in providing a good solvating environment for the ions produced by the proton
transfer.
11–13
esterification reactions.
Introduction
PILs have also been found to be
suitable electrolytes in fuel cells due to the presence of the
14
mobile proton.
Over the past decade much attention has been given to ionic
liquids (ILs), a class of compounds comprised only of ions,
However, the properties exhibited by PILs depend largely
on their degree of ionization, i.e. the degree of proton transfer
from acid to base. In principle, since ILs are composed entirely
of ions, the equilibrium of eqn (1) should lie predominantly to
the right, for a ‘‘pure’’ salt to be formed. Some of the literature
1
with melting points below 100 1C. ILs have become of
interest in many fields (chemistry, biochemistry and materials
engineering to name a few) as their unique properties
offer a basis for new and improved technologies. Typical
characteristics, such as non-flammability, non-volatility,
increased thermal stability and high ionic conductivity, are
just some of the many distinctive properties that can be
found amongst the huge IL family, which can all be
suggests that the extent of proton transfer should be at least
15,16
99% for the PIL to be considered a ‘‘pure’’ salt.
Yoshizawa
et al. demonstrated the use of a Walden plot to qualitatively
10
assess ionicity and the degree of proton transfer in ILs.
2
,3
customized through careful choice of cation and anion.
However, it has become clear that the Walden plot is unable
to distinguish between low ionicity in an IL, that can arise
for example from ion pairing, and a low degree of proton
They can be utilized as solvents in a variety of organic and
inorganic reactions, as electrolytes in batteries, fuel cells and
solar cells, as protein solubilising and stabilising media, as
pharmaceutical delivery aids and as pharmaceuticals in some
17
transfer.
Yoshizawa and co-workers have also used aqueous pKa
4
–9
cases.
values as the basis for estimates of the degree of proton
10
transfer in PILs. By finding a simple relationship between
aq
ILs are divided into two major subclasses: aprotic and
protic. In aprotic systems the parent base molecule is ionised
the observed boiling points and the difference in pK
aq
a
values
(
thus becoming a cation) by accepting any group other than a
for the acid and the base, DpK
a
, Yoshizawa et al. observed
1
proton to its basic site. Protic ionic liquids (PILs), on the
0
aq
that DpK
transfer to occur in the PILs studied. This result is somewhat
a
> 10 seemed to be required for the full proton
other hand, are compounds which are formed upon proton
transfer from a Bro
¨
1
¨
nsted acid (HA) to a Bronsted base (B)
18
surprising given that simple equilibrium calculations indicate
aq
that, in aqueous solution, DpK
0
(
eqn (1)).
a
= 4 is sufficient to produce
9
9% proton transfer. This suggests that the IL solvation
+
HA + B " BH + A
À
environment can have a very strong effect on the energy of
aq
proton transfer, i.e. that DpKa is a relatively poor estimate of
(1)
the effective DpK in the ionic liquid environment. The work of
a
1
9
PILs possess notably different chemical properties to
conventional aprotic ILs, including in some cases an
ability to distil due to high volatility, although distillation
takes place via the neutral acid and base. Possessing an
Johansson et al. appears to support these observations.
However, the situation becomes more perplexing when one
2
considers the work of Banerjee and co-workers who present
0
crystal structures for a number of compounds formed by
aq
exchangeable proton, PILs exhibit Bro
¨
nsted acidity, which
combination of acids and bases of varying DpKa and show
that proton transfer has substantially taken place in the
aq
allows these compounds to be used as solvents for a number
of acid catalysed reactions including Diels–Alder, Mannich and
crystalline salt when DpKa is greater than 3. Below this value
the crystal structure is that of a co-crystal of the acid and
2
base. Thus this work seems to indicate that proton transfer
0
School of Chemistry, Monash University, Clayton, Victoria 3800,
Australia. E-mail: Jelena.Stoimenovski@sci.monash.edu.au
in the solid state of an ionic compound can be substantial for
This journal is ꢀc the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10341–10347 | 10341

View Article Online
aq
a
DpK
> 3 and this is considered a rule of thumb in the
co-crystal field. Since the ‘‘solvent’’ environments can be expected
to have many similarities in the liquid and solid states of salts,
these observations seem to be in conflict with those of Yoshizawa
et al. This paper explores the proton transfer in these liquids in
more depth in an attempt to probe these issues.
It is difficult to assess proton transfer using standard
2
methods such as pH measurements. Burrell et al. have
1
1
5
recently illustrated the use of N NMR to distinguish between
neutral and ionised amine bases. However, NMR literature
has shown that the effect of proton transfer on the chemical
shifts of nitrogens cannot be distinguished from that of
22,23
hydrogen bonding
and hence this technique is not applicable
to the present work. An alternative approach involves the use
of an indicator acid existing in various protonated forms
depending on the pH of the medium. In this study, phenol
red (also known as phenolsulfonphthalein or PSP) was chosen
as a typical probe indicator acid due to the strong degree of
colouration in both its acid and base forms. In the crystalline
form and in solution under very acidic conditions, PSP exists
in its zwitterionic form, appearing as an orange-red coloured
solution (Fig. 1). As the pH increases (pK
from the ketone group is lost and the solution becomes yellow.
At an even higher pH (pK = 8.00) a second proton is lost
from the phenolic group, resulting in the bright pink (fuchsia)
a
= 1.03) the proton
a
aq
a
aq
Fig. 2 pK
and DpK
a
values of acetic acid and amines used in the
2
4
study.
colour.
Owing to sufficient difference in the pKa1 and pKa2
aq
aq
values of the PSP indicator, it is possible to visually assess
the ionic state of PSP in a variety of different media. In this
Experimental
work, PSP was used to determine its degree of protonation in a
aq
General
group of amines (Fig. 2) of relatively similar values of pK
aq
a
and hence of similar DpK
a
values (B6) when paired with
All amines were purchased from Aldrich Chemical Company
and were further purified by distillation over KOH. Glacial
acetic acid (99.9+%) was purchased from Sigma-Aldrich
(USA) and was used without further purification.
acetic acid. These were chosen to exemplify different structural
features, in particular the hydrogen bonding ability that is
present in the primary amines but not in the tertiary. The
amines were subsequently used to prepare acetate-containing
PILs for which transport properties were determined and were
compared to each other on a Walden plot.
NMR data were recorded on a Bruker Avance400 (9.4 Tesla
magnet) with a 5 mm broadband autotunable probe with
Z-gradients and BACS 60 tube autosampler.
Fig. 1 Different forms of PSP, and the corresponding colours, as pH is increased.
0342 | Phys. Chem. Chem. Phys., 2010, 12, 10341–10347
This journal is ꢀc the Owner Societies 2010
1

View Article Online
Density data were collected using an Anton Paar DMA
000 density meter. The density of the compounds was
N-heptylamine, were fully conformationally screened at this
level of theory to ensure the global (rather than merely local)
minima were obtained. The screening was achieved by
exploring all conformations arising from the rotations around
single chemical bonds. The conformation of the lowest energy
was selected as the global minimum for each species. In the
case of N-heptylamine the ethyl fragment was added to
the lowest energy conformation of N-butylamine and the
conformational search was restricted to the rotation around
5
determined using the ‘‘oscillating U-tube principle’’.
The viscosity measurements were performed using a falling
ball technique on an Anton Paar AMVn viscosity meter.
Conductivity measurements were carried out on a locally
designed dip cell probe containing two platinum wires covered
in glass. 0.01 M KCl solution at 25 1C was used to determine
the cell constant. The conductivities were obtained by measuring
the complex impedance spectra between 1 MHz and 0.01 Hz
on a Solatron SI 1296 dielectric interface.
the single C
4 5
–C bond. Quantum chemical free energies of
solvation in aqueous solution were calculated using the
2
8,29
COSMO-RS approach
set of programs.
as implemented in the ADF 2009
Synthesis
N-Butylammonium acetate. N-Butylamine (7.14 g, 97.60 mmol)
was added dropwise to glacial acetic acid (5.86 g, 97.60 mmol).
The mixture was stirred together in an ice bath for three hours,
Molecular volumes were computed using the HF/aug-cc-
pVTZ electron density with tight criteria (an increased density
of points for more accurate integration). The effective ionic
radii were calculated assuming that the ions studied have a
spherical shape:
after which a clear viscous liquid remained.
1
H NMR (400 MHz, D O) d (ppm) 0.92–0.96 (t, J 7.6, 3H),
2
1
.36–1.45 (sextet, J 7.2 2H), 1.61–1.68 (p, J 7.6, 2H), 1.92
4
3
3
V
calc = pRion
(2)
(
s, 3H), 2.98–3.02 (t, J 7.6, 2H).
N-Heptylammonium acetate. N-Heptylamine (7.56 g,
Results
6
5.61 mmol) was added dropwise to glacial acetic acid (3.94 g,
5.61 mmol). The mixture was stirred together in an ice bath
6
Initial studies of the extent of protonation using the PSP
indicator are presented in Fig. 3. Simple addition of a small
quantity of the indicator acid to the pure liquid amine resulted
in the various changes in colour of the otherwise colourless
compounds shown in Fig. 3. There was a clear distinction
observed between the primary amines, in which the PSP was a
pink (fuchsia) colour, and the tertiary amines in which the
indicator became bright yellow in colour.
for three hours, after which a clear viscous liquid remained.
The product crystallized after approximately 2 h standing at
room temperature.
1
H NMR (400 MHz, D O) d (ppm) 0.88–0.92 (t, J 6.9, 3H),
.32–1.44 (m, 8H), 1.65–1.72 (p, J 7.5, 2H), 1.94 (s, 1H),
2
1
3
.00–3.04 (t, J 7.5, 2H).
Triethylammonium acetate. Triethylamine acetate (6.90 g,
8.22 mmol) was added dropwise to glacial acetic acid (4.10 g,
8.22 mmol). The mixture was stirred together in an ice bath
It appears that the indicator acid is entirely deprotonated to
its dibasic form in the primary amines, while in the tertiary
amines the PSP has only been deprotonated to its monobasic
form. It is not surprising that all of the amines were able to
deprotonate the PSP acid to at least the monobasic form since
6
6
forming an emulsion.
1
9
N-Methylpyrrolidinium acetate . N-Methylpyrrolidine
7.92 g, 92.98 mmol) was added dropwise to glacial acetic acid
5.58 g, 92.98 mmol). The mixture was stirred together in an
aq
aq
pKa for this proton is B1.0 and hence DpKa > 9 for all of
(
(
the amines involved. However, only the primary amines were
aq
able to deprotonate the second proton (DpK
a
E 2.5) of the
ice bath for three hours, after which a slightly yellow clear
indicator. Several secondary amines (dipropylamine and
pyrrolidine) were also studied (not shown) and found either
to deprotonate the second proton of the indicator, and hence
behave as primary amines do, or to deprotonate only the first
proton of PSP and therefore act more like tertiary amines.
Thus the secondary amines as a class seem to be transitional
depending on the details of their chemical structure.
liquid remained.
1
H NMR (400 MHz, D O) d (ppm) 1.94 (s, 3H), 2.00–2.10
2
(
m, 2H), 2.14–2.21 (m, 2H), 2.94 (s, 3H), 3.05–3.12 (m, 2H),
3
.64–3.68 (m, 2H).
2
5
N,N-Dimethylbutylammonium acetate . N,N-Dimethyl-
butylamine (7.22 g, 71.32 mmol) was added dropwise to glacial
acetic acid (4.28 g, 71.32 mmol). The mixture was stirred
together in an ice bath for three hours, after which a clear
The density data for the synthesised PILs, displayed in
Fig. 4, show that the densities of the primary amine based
PIL, butylammonium acetate, and a tertiary amine based PIL,
methylpyrrolidinium acetate, are very similar. Dimethyl-
butylammonium acetate exhibits the lowest density of all of
the synthesised PILs. It was very clear in handling these
compounds during these measurements that the tertiary amine
based salts were quite volatile (strong amine smell), whereas
the primary amine salts were not.
liquid remained.
1
H NMR (400 MHz, D O) d (ppm) 0.87–0.90 (t, J 7.4, 3H),
2
1
2
.28–1.37 (sextet, J 7.4, 2H), 1.60–1.68 (m, 2H), 1.86 (s, 3H),
.81 (s, 6H), 3.06–3.10 (m, 2H).
Theoretical procedures
Standard ab initio molecular orbital theory and density functional
theory (DFT) calculations were carried out using the
2
Typical viscosity and conductivity data are presented in
Fig. 5 and 6 as a function of temperature.
6
27
GAUSSIAN 09 and ADF 2009 sets of programs. All
geometry optimizations were performed at the B3LYP/
The viscosity data presented illustrate the immense difference
between the viscosities of the primary and tertiary amine based
6
-31+G(d) level of theory. Individual amines, except for
This journal is ꢀc the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10341–10347 | 10343

View Article Online
Fig. 3 PSP in acidic aqueous solution (1), in neat N-butylamine + PSP (2), N-heptylamine + PSP (3), triethylamine + PSP (4),
N-methylpyrrolidine + PSP (5), N,N-dimethylbutylamine + PSP (6).
Fig. 4 Density data for the synthesised PILs.
Fig. 5 Viscosity data for the synthesised PILs as a function of
temperature.
PILs; this difference is immediately obvious in the handling of
these PILs. The primary amine based PILs possess viscosities
more than an order of magnitude higher than the tertiary
amine-based PILs. In addition, the viscosities of the latter
change only slightly over the temperature range studied, while
the former show a significant increase. As discussed below, this
may be the result of disruption of the hydrogen bonding
between the ions as the temperature increases. Note that in
the case of methylpyrrolidinium acetate, the viscosity data
could not be obtained for the whole range of temperatures
chosen for this study (i.e. 70–90 1C) due to volatility of the
sample at elevated temperatures.
can be observed due to the melting of the sample at approximately
30 1C. The primary amine based PILs have significantly lower
conductivities than the tertiary amine based PILs. However,
with the temperature increase, the difference in the conductivity
between the primary and tertiary amine based PILs becomes
smaller, whereby at 70 1C the conductivities of butylammonium
acetate and dimethylpyrrolidinium acetate are comparable. At
even higher temperatures, butylammonium acetate has higher
conductivity than that of dimethylbutylammonium acetate.
Methylpyrrolidinium acetate exhibits the highest conductivity,
while the primary amine based PIL, heptylammonium acetate,
possesses lowest conductivity of all the synthesised PILs. It
might seem that the tertiary amine based PILs are appearing
more ‘‘ionic’’ on the basis of these data, however as we show
below, this is an incorrect conclusion.
The conductivities of the synthesised PILs (Fig. 6) differ
substantially from one another at room temperature. In the
case of heptylammonium acetate, a distinct step in conductivity
1
0344 | Phys. Chem. Chem. Phys., 2010, 12, 10341–10347
This journal is ꢀc the Owner Societies 2010

View Article Online
Table 1 Deviations from the Walden plot, DW, with and without an
adjustment for ion size
Adjusted
DW (Æ0.1)
DW (Æ0.1)
Synthesised PILs
@ 313 K
@ 313 K
Butylammonium acetate
0.49
0.72
1.78
2.33
0.32
0.57
1.62
2.15
Heptylammonium acetate
Methylpyrrolidinium acetate
Dimethylbutylammonium acetate
might be expected on the basis of its viscosity, according to the
Walden rule. In the case of PILs, a large DW may also indicate
incomplete proton transfer, i.e. the presence of the parent acid
and base molecules.
Fig. 7 shows two distinct groups of PILs. One observes a
large deviation (Table 1) from the ‘‘ideal line’’ for tertiary
amine based PILs, while the primary amine based PILs lie
1
7
close to the ideal line. As discussed recently the Walden plot
axis should be adjusted to allow for the effect of ion size.
Doing so in this case produces small adjustments to DW,
which are also listed in Table 1. If anything, this accentuates
the difference between the two families of compounds.
The data suggest that the primary amine based PILs are
predominantly proton transferred, i.e. the equilibrium in
eqn (1) lies predominantly towards the ionic species, while
the tertiary amine based PILs, with DW values of well over 1,
demonstrate the presence of neutral non-conducting species,
possibly the parent acid and base (though we reiterate that
large values of DW can also arise from neutral ion pairs and
aggregates).
Fig. 6 Conductivity data for the synthesised PILs as a function of
temperature.
Discussion
The reactions of the primary amines with acetic acid resulted
in highly viscous liquids, much more viscous and much less
volatile than the parent acid and base. Both acid and base have
readily accessible hydrogen bond donor and acceptor sites and
therefore there is no obvious reason why mixing them would
produce such a significant increase in viscosity via additional
hydrogen bonding alone. These observations indicate that
some reaction has occurred between the acid and the base.
These PILs exhibit molar conductivities close to that expected
on the basis of the Walden rule (small DW). As shown in
Fig. 3, the indicator tests tended to show that the indicator
acid rapidly displayed a colouration typical of the fully
deprotonated form. These pieces of evidence taken together
indicate that the extent of proton transfer in these PILs is
relatively close to complete.
Fig. 7 Walden plot for the synthesised PILs.
A Walden plot encompassing all of the density, viscosity
and conductivity data is presented in Fig. 7. The Walden rule:
LZ = constant
(3)
where, L is the molar conductivity and Z the viscosity, predicts
a simple inverse relationship between these two key transport
properties, which can be tested using a plot of log L versus log
Z, as described by Yoshizawa et al., in which data for a 0.1 M
On the other hand, upon the addition of the tertiary amines
to acetic acid, very fluid liquids were produced, the fluidity
being similar to that of the original starting materials. The
mixtures remained quite volatile. It seemed that only partial
reaction occurred. Reaction of triethylamine and acetic acid
resulted in an emulsion, indicating an inability of the two
liquids to mix at all; presumably, therefore, no reaction has
occurred. The ‘‘PILs’’ resulting from these amines tended to
show large deviations from the Walden rule and the indicator
acid appeared to remain in the mono-protonated state in the
pure base. This evidence taken as a whole seems to indicate
1
0
aqueous KCl solution are used as a point of reference.
Hence, an ‘‘ideal’’ line representing the behaviour of an ideal
electrolyte (in which ions act in an independent fashion) runs
1
0,17,30
directly through the KCl point with unity slope.
Any
negative deviation from this reference line is thought to
indicate a lack of complete proton transfer, or formation of
neutral clusters that cannot conduct. The vertical deviation,
DW, from the ‘‘ideal line’’ can be taken as a measure of this
phenomenon. For example, DW = 1 represents the situation
where the IL is exhibiting only 10% of the conductivity that
This journal is ꢀc the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10341–10347 | 10345

View Article Online
that relatively little proton transfer has taken place in
these cases.
case of the tertiary amine system studied in that work, the
formation of these ions in compositions which are acid rich,
thus providing a further hydrogen bond donor to assist in
However, the amines selected for this study all have
aq
aq
À
comparable pKa values, such that the difference, DpKa
,
stabilising A , was observed to support much stronger proton
for the various combinations is very similar (B6). These
compounds were chosen in order to allow an examination of
aq
transfer than at the stoichiometric composition.
the hypothesis that DpK
a
> 10 is required for strong proton
Computational studies
transfer. Thus it appears, from the various observations
The free energy of the proton transfer process, DGPT, can be
expressed as follows:
described above, that proton transfer equilibria in PILs based
on simple primary amines behave, in fact, much as expected on
aq
the basis of aqueous pK
a
values, i.e. that DpK
a
> 2 is
+
DGPT = Gtot(BH ) À Gtot(B) + Gtot(A ) À Gtot(AH) (4)
À
sufficient to produce predominant proton transfer (B90%)
aq
and DpK
a
> 4 will produce >99% ionisation. On the other
where Gtot = Ggas + DGsolv for each solvated species and is
obtained from the sum of the free energy of the gas phase
species plus its free energy of solvation in the solvent. Eqn (5)
can then be re-written in the following form:
hand, in this work values substantially greater than those
aq
around DpKa = 6 are required to produce strong proton
transfer with the simple tertiary amines and hence it is these
cases that are behaving unusually. Secondary amines seem to
1
lie between these extremes. The work of Yoshizawa et al.
0
+
DGPT = DGgas + DG (BH ) À DGsolv(B)
solv
À
mostly focussed on tertiary amines and hence the conclusions
reached here are quite consistent with their observations.
The molecular origins of these differences almost certainly
relate to differences in the hydrogen bonding environments
offered by the amine/ammonium ions. In the primary amine
case (and the corresponding alkyl ammonium ions) the cation
has multiple hydrogen bond donors available to interact with
and solvate the carboxylate anion that is formed by the proton
transfer reaction. One might expect its solvation in this respect
to be relatively similar to that of water. On the other hand, the
hydrogen bonding frustrated environment offered by the
tertiary amine/ammonium ion is distinctly different to that
of water. The hydrogen bonding acceptor sites on the acetate
anion (two sites on each of two oxygens) have only the single
proton on any given neighbouring cation with which to
hydrogen bond. This could be expected to provide a less
stabilising solvent environment for the anion (and also the
cation) and hence proton transfer to form these ions is
energetically less favoured. The case of ‘‘triethylammonium
acetate’’ presents an interesting extreme in which the acid and
base do not even mix.
+
DGsolv(A ) À DGsolv(AH)
(5)
where DGgas is the free energy of proton transfer in the gas
phase. This equation makes explicit the role of solvation
energies in the observed proton transfer processes in any given
solvent.
Quantum chemical calculations of DGsolv in aqueous
solutions were performed for the protonated amines. The
COSMO-RS approach was employed to extract the free
energies of solvation, as it explicitly treats contributions from
hydrogen bonding sites in the molecule to the total free energy
of solvation. The results are given in Table 2. Analysis of the
results reveals, as expected, that primary ammonium ions have
substantially larger solvation energies in water compared to
those of the tertiary ammonium ions, indicating that the
protonated primary amines are strongly stabilised by the
hydrogen bonding environment that water represents. While
this type of calculation cannot directly probe the IL environ-
ment it is our hypothesis here that similar hydrogen bond
stabilisation is available for both cation and anion by the
protonated primary amine in the IL. On the other hand
the hydrogen bonding ‘‘starved’’ environment represented by
the protonated tertiary ammonium cation ionic liquid presents
far less stabilisation of the ions and therefore the proton
transfer reaction is relatively discouraged in that medium. These
observations are all also consistent with the crystal structure
observations of Banerjee et al.; however, recognising that
hydrogen bonding is a key element in the phenomena here
suggests that crystal structures in which hydrogen bonding is
strongly enabled, by the fixed arrangement of the ions, might
exhibit stronger proton transfer than the liquid state of the
This distinction between primary and tertiary amine based
PILs assists in understanding the various effects that can be
observed in the Walden plot. The case of butylammonium
acetate, with an adjusted DW of 0.32, probably represents
an example of a close-to-ideal PIL (‘‘ideal’’ indicating fully
proton transferred and all ions independently mobile). The
two tertiary amine based ‘‘PILs’’, which are in fact very poorly
proton transferred, sit low on the Walden plot for this reason.
Heptylammonium acetate illustrates the further phenomenon
that influences the Walden plot in that it is likely to be fully
proton transferred but sits lower in the Walden plot because of
association between the ions. In this case hydrogen bonding is
clearly the origin of the association. Any other functional
groups on either ion that can add further basis for hydrogen
bonding or ion association will exacerbate this effect. This
is true of many of the pharmacologically active ILs under
À1
Table
2
Free energies of solvation (DGsolv
,
in kJ mol
)
of
the primary and tertiary ammonium cations. DDGsolv expresses the
differences relative to the dimethylbutylammonium ion
À1
À1
DGsolv/kJ mol
DDGsolv/kJ mol
N,N-Dimethylbutylammonium À228.4
0.0
À6.3
9
investigation in our laboratories.
N-Methylpyrrolidinium
N-Heptylammonium
N-Butylammonium
Acetate
À234.7
À304.1
À307.5
À357.2
A further factor in the proton transfer process in these neat
liquids is the possibility of more complex species forming, of
À75.7
À79.1
À129.0
À
19
the type A–H–A , as described by Johansson et al. In the
1
0346 | Phys. Chem. Chem. Phys., 2010, 12, 10341–10347
This journal is ꢀc the Owner Societies 2010

View Article Online
same material. This would suggest that crystallisation might
encourage proton transfer in some cases.
12 Y. Du and F. Tian, J. Chem. Res., 2006, 486–489.
1
1
1
3 Z. Duan, Y. Gu, J. Zhang, L. Zhu and Y. Deng, J. Mol. Catal. A:
Chem., 2006, 250, 163–168.
4 M. A. B. H. Susan, A. Noda, S. Mitsushima and M. Watanabe,
Chem. Commun., 2003, 938–939.
5 B. Nuthakki, T. L. Greaves, I. Krodkiewska, A. Weerawardena,
M. I. Burgar, R. J. Mulder and C. J. Drummond, Aust. J. Chem.,
2007, 60, 21–28.
Conclusions
In this work we have shown that primary and tertiary amines
aq
of similar pKa values lead to a different extent of proton
1
6 D. R. MacFarlane and K. R. Seddon, Aust. J. Chem., 2007, 60,
transfer when paired with a common acid, acetic acid. The
Walden plot for four synthesised PILs reveals that PILs
obtained from primary amines exhibit a ‘‘close-to-ideal’’
behaviour, whereas those from tertiary amines sit far below
the ideal line (DW > 1). Based on this and a range of physical
property observations we conclude that simple primary amines
produce highly ionised ILs, whereas tertiary amines tend to
form mixtures with a low degree of proton transfer. COSMO-RS
free energies of solvation of protonated amines indicate that
the solvation of primary ammonium ions is substantially
stronger due to the increased hydrogen bonding ability of
the former. It is important to note that these trends can be
easily disturbed by other solvation effects, for example longer
chain alkyl groups bring hydrophobic interactions into play
and other functional groups elsewhere in either molecule,
especially –OH or other hydrogen bonding sites, will have a
major effect on the outcome.
3–5.
17 D. R. MacFarlane, M. Forsyth, E. I. Izgorodina, A. P. Abbott,
G. Annat and K. Fraser, Phys. Chem. Chem. Phys., 2009, 11,
4
8 D. R. MacFarlane, J. M. Pringle, K. M. Johansson, S. A. Forsyth
962–4967.
1
and M. Forsyth, Chem. Commun., 2006, 1905–1917.
19 K. M. Johansson, E. I. Izgorodina, M. Forsyth, D. R. MacFarlane
and K. R. Seddon, Phys. Chem. Chem. Phys., 2008, 10, 2972–2978.
0 R. Banerjee, P. M. Bhatt, N. V. Ravindra and G. R. Desiraju,
2
Cryst. Growth Des., 2005, 5, 2299–2309.
21 G. L. Burrell, I. M. Burgar, F. Separovic and N. F. Dunlop, Phys.
Chem. Chem. Phys., 2010, 12, 1571–1577.
2
2 Z. Dega-Szafran, M. Szafran, J. Sitkowski and L. Stefaniak,
J. Phys. Org. Chem., 1996, 9, 746–750.
2
3 M. Witanowski, Z. Biedrzycka, W. Sicinska and Z. Grabowski,
J. Magn. Reson., 1998, 131, 54–60.
2
2
2
4 Z. Tamura and M. Maeda, Yakugaku Zasshi, 1997, 117, 764–770.
5 A. Walker, WO Pat., 2009034329, 2009.
6 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, T. V. Jr.,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz,
Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, A.
L. G. Liu, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 09,
Revision A.02, Gaussian, Inc., Wallingford, CT, 2009.
Acknowledgements
We gratefully acknowledge generous allocations of computing
time from the National Facility of the National Computational
Infrastructure and the Monash Sun Grid Cluster at the e-research
centre of Monash University. EII gratefully acknowledges the
support of the Australian Research Council for her postdoctoral
fellowship, as does DRM for his Federation Fellowship.
Notes and References
1
P. Walden, Z. Phys. Chem., Stoechiom. Verwandtschaftsl., 1906,
5, 207–249.
5
27 E. J. Baerends, J. Autschbach, J. A. Berger, A. Berces,
´
2
3
K. R. Seddon, J. Chem. Technol. Biotechnol., 1997, 68, 351–356.
Ionic Liquids in Synthesis, ed. P. Wasserscheid, T. Welton,
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany,
F. M. Bickelhaupt, C. Bo, P. M. B. P. L. de Boeij, L. Cavallo,
D. P. Chong, L. Deng, R. M. Dickson, D. E. Ellis, M. van Faassen,
L. Fan, T. H. Fischer, C. Fonseca Guerra, S. J. A. van Gisbergen,
2nd edn, 2008.
T. Welton, Chem. Rev., 1999, 99, 2071–2083.
Electrodeposition from Ionic Liquids, ed. F. Endres,
D. MacFarlane, A. Abbott, Wiley-VCH, 2008.
K. Fujita, D. R. MacFarlane and M. Forsyth, Chem. Commun.,
¨
¨
A. W. Gotz, J. A. Groeneveld, O. V. Gritsenko, M. Gruning,
4
5
F. E. Harris, P. van den Hoek, C. R. Jacob, H. Jacobsen,
L. Jensen, E. S. Kadantsev, G. van Kessel, R. Klooster,
F. Kootstra, M. V. Krykunov, E. van Lenthe, J. N. Louwen,
D. A. McCormack, A. Michalak, J. Neugebauer, V. P. Nicu,
V. P. Osinga, S. Patchkovskii, P. H. T. Philipsen, D. Post,
C. C. Pye, W. Ravenek, J. I. Rodriguez, P. Romaniello, P. Ros,
6
7
8
2
005, 4804–4806.
W. L. Hough and R. D. Rogers, Bull. Chem. Soc. Jpn., 2007, 80,
262–2269.
2
P. R. T. Schipper, G. Schreckenbach, J. G. Snijders, M. Sola,
`
W. L. Hough, M. Smiglak, H. Rodriguez, R. P. Swatloski,
S. K. Spear, D. T. Daly, J. Pernak, J. E. Grisel, R. D. Carliss,
M. D. Soutullo, J. J. H. Davis and R. D. Rogers, New J. Chem.,
2007, 31, 1429–1436.
J. Stoimenovski, D. R. MacFarlane, K. Bica and R. D. Rogers,
Pharm. Res., 27, 521–526.
M. Swart, D. Swerhone, G. te Velde, P. Vernooijs, L. Versluis,
L. Visscher, O. Visser, F. Wang, T. A. Wesolowski, E. M. van
Wezenbeek, G. Wiesenekker, S. K. Wolff, T. K. Woo,
A. L. Yakovlev and T. Ziegler, ADF 2009.01b, Scientific Comput-
ing & Modelling NV (SCM), 2009.
9
28 F. Eckert and A. Klamt, J. Comput. Chem., 2006, 27, 11–19.
29 A. Klamt, F. Eckert, M. Diedenhofen and M. E. Beck, J. Phys.
Chem. A, 2003, 107, 9380–9386.
30 K. J. Fraser, E. I. Izgorodina, M. Forsyth, J. L. Scott and
D. R. MacFarlane, Chem. Commun., 2007, 3817–3819.
1
0 M. Yoshizawa, W. Xu and C. A. Angell, J. Am. Chem. Soc., 2003,
25, 15411–15419.
1
1 E. Janus, I. Goc-Maciejewska, M. Lozynski and J. Pernak,
Tetrahedron Lett., 2006, 47, 4079–4083.
1
This journal is ꢀc the Owner Societies 2010
Phys. Chem. Chem. Phys., 2010, 12, 10341–10347 | 10347