Electroanalysis of Neutral Precursors in Protic Ionic Liquids and Synthesis of High-Ionicity Ionic Liquids

Article abstract of DOI:10.1021/acs.langmuir.7b02294

Protic ionic liquids (PILs) are ionic liquids that are formed by transferring protons from Br?nsted acids to Br?nsted bases. While they nominally consist entirely of ions, PILs can often behave as though they contain a significant amount of neutral species (either molecules or ion clusters), and there is currently a lot of interest in determining the degree of "ionicity" of PILs. In this contribution, we describe a simple electroanalytical method for detecting and quantifying residual excess acids in a series of ammonium-based PILs (diethylmethylammonium triflate [dema][TfO], dimethylethylammonium triflate [dmea][TfO], triethylammonium trifluoroacetate [tea][TfAc], and dimethylbutylammonium triflate [dmba][TfO]). Ultra-microelectrode voltammetry reveals that some of the accepted methods for synthesizing PILs can readily result in the formation of nonstoichiometric PILs containing up to 230 mM excess acid. In addition, vacuum purification of PILs is of limited use in cases where nonstoichiometric PILs are formed. Although excess bases can be readily removed from PILs under ambient conditions, excess acids cannot be removed, even under high vacuum. The effects of excess acid on the electrocatalytic oxygen reduction reaction (ORR) in PILs have been studied, and the onset potential of the ORR in [dema][TfO] increases by 0.8 V upon addition of acid to PIL. On the basis of the results of our analyses, we provide some recommendations for the synthesis of highly ionic PILs.

Full text of DOI:10.1021/acs.langmuir.7b02294

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES
Article
Electroanalysis of Neutral Precursors in Protic Ionic
Liquids and Synthesis of High-Ionicity Ionic Liquids
Sean E. Goodwin, Daniel E. Smith, Joshua S. Gibson, Robert George Jones, and Darren Anthony Walsh
Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b02294 • Publication Date (Web): 07 Aug 2017
Downloaded from http://pubs.acs.org on August 9, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 34
Langmuir
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Electroanalysis of Neutral Precursors in Protic Ionic Liquids and
Synthesis of High-Ionicity Ionic Liquids
Sean E. Goodwin,1 Daniel E. Smith,1 Joshua S. Gibson,2 Robert G. Jones,2 and Darren A. Walsh1
1School of Chemistry and GSK Carbon Neutral Laboratory for Sustainable Chemistry
The University of Nottingham, Jubilee Campus
Nottingham NG7 2TU, UK
2Department of Physical Chemistry
School of Chemistry
The University of Nottingham, University Park
Nottingham NG7 2RD, UK
*darren.walsh@nottingham.ac.uk; Tel: +44 115 8467495; Fax: +44 115 9513562
1
ACS Paragon Plus Environment

Langmuir
Page 2 of 34
1
2
3
4
Abstract
5
6
7
8
9
Protic ionic liquids (PILs) are ionic liquids that are formed by transferring protons from Brønsted
acids to Brønsted bases. While they nominally consist entirely of ions, PILs can often behave as
though they contain a significant amount of neutral species (either molecules or ion clusters), and
there is currently a lot of interest in determining the degree of “ionicity” of PILs. In this
contribution, we describe a simple electroanalytical method for detecting and quantifying residual
excess acids in a series of ammoniumꢀbased PILs (diethylmethylammonium triflate,
[dema][TfO], dimethylethylammonium triflate, [dmea][TfO], triethylammonium trifluoroacetate,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[tea][TfAc], and dimethylbutylammonium triflate [dmba][TfO]).
Ultramicroelectrode
voltammetry reveals that some of the accepted methods for synthesising PILs can readily result in
the formation of nonꢀstoichiometric PILs containing up to 230 mM excess acid. In addition,
vacuum purification of PILs is of limited use in cases where nonꢀstoichiometric PILs are formed.
While excess bases can be readily removed from PILs, even under ambient conditions, excess
acids cannot, even under high vacuum. The effects of excess acid on the electrocatalytic oxygen
reduction reaction (ORR) in PILs have been studied, and the onset potential of the ORR in
[dema][TfO] increases by 0.8 V upon addition of excess acid to PIL. Based on the results of our
analyses, we provide some recommendations for the synthesis of highlyꢀionic PILs.
2
ACS Paragon Plus Environment

Page 3 of 34
Langmuir
1
2
3
4
Introduction
5
6
7
8
9
Room temperature ionic liquids (RTILs) are defined as species consisting entirely of ions
and which are liquid below 100 °C.1 Due to their low vapor pressures, wide electrochemical
windows, and inherent conductivities, RTILs are being used in a wide range of applications, such
as catalysis,2 synthesis,3 electrodeposition,4 sensing,5 and energy conversion and storage.6ꢀ7 In
each case, the physicochemical properties of the RTILs are the main drivers for their selection,
and understanding the structureꢀproperty relationships of these media is an extremely active area
of research. A major theme in this research is understanding the degree of “ionicity” of RTILs
(or how much of the RTILs actually comprises free ions). It is clear that some RTILs behave as
though they are not composed entirely of free ions, exhibiting relatively high vapor pressures and
low ionic conductivities.8ꢀ15 Such phenomena are often particularly evident in protic ionic liquids
(PILs), which are RTILs formed by transferring protons from Brønsted acids to Brønsted bases.
Depending on the degree of proton transfer during synthesis, PILs can actually comprise a
mixture of the salt, the parent acid, and the parent base, reducing the ionicity of the liquids. It has
also been proposed that the formation of longꢀlived ion pairs and ion clusters within the liquids
can lower the degree of ionicity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
At present, there are no simple methods for unambiguously determining the degree of
ionicity of PILs. However, qualitative estimates based on measurements of ionic conductivity
are extremely common, and often involve use of the Walden rule:
Λη = k
(1)
3
ACS Paragon Plus Environment

Langmuir
Page 4 of 34
1
2
3
4
5
6
7
8
9
where Λ is the molar conductivity, η is the viscosity, and k is a temperatureꢀdependent constant.
The data for fullyꢀdissociated, strong electrolytes such as dilute aqueous KCl lie along the ideal
Walden line of log Λ vs. log η−1, where η−1 is the “fluidity” of the medium, but the conductivities
of PILs tend to fall significantly below the ideal line, providing a qualitative estimate of
ionicity.16 However, conductivity measurements cannot distinguish between the presence of
neutral ion pairs, neutral ion clusters, and neutral molecules in the liquids. Due to such
difficulties, Macfarlane has suggested that the ionicity of PILs be simply defined as the “effective
fraction of ions available to participate in conduction.”17
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A number of alternatives to conductimetry have been proposed for studying the ionicity
of PIL, including the analysis of PIL vapors and NMR spectroscopy. Analysis of PIL vapors
demonstrates that back proton transfer from the protonated base to the deprotonated acid can
occur at elevated temperatures18ꢀ19 but cannot provide insights into the concentration of
precursors in an “asꢀsynthesised” liquid. 1H NMR spectroscopy can reveal the degree of
shielding of the PIL cations’ exchangeable protons but measuring the extent of protonation of the
basic site is complicated by the presence of protic impurities such as H2O, and fast exchange of
protons between the components of the PILs.20ꢀ21 15N NMR spectroscopy is another option for
studying the extent of protonation of amine basic sites, but is complicated by interference from
hydrogen bonding.16,22 As such, qualitative methods such as use of Walden plots remain popular
in studies of the ionicity of PILs. The usefulness and popularity of such qualitative methods are
in part due to the frequent correlation between the conductivity of PILs and differences in the
aqueous pKas of the parent acids and bases (ꢁpKa) used to make PILs. ꢁpKa represents the
driving force for proton transfer and, while ꢁpKa > 3 is sufficient for nearꢀcomplete proton
transfer in aqueous solutions, it is generally believed that ꢁpKa values greater than about 10 lead
to highlyꢀionic (or good) PILs.14ꢀ15,23 However, Macfarlane and coꢀworkers have demonstrated
4
ACS Paragon Plus Environment

Page 5 of 34
Langmuir
1
2
3
4
5
6
7
8
9
the weakness of the use of aqueous pKas as a guide for designing PILs.22 They found that the
required ꢁpKa for effectively complete proton transfer actually depends on the nature of the
amine used to form ammoniumꢀbased PILs. A series of primary amines deprotonated an
indicator acid at ꢁpKa = 3 but ꢁpKa = 10 was required to deprotonate tertiary amines, a factor that
was attributed to the different Hꢀbonding abilities of the species, due to steric hindrance of
hydrogen bonding to tertiary amines.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Complicating studies of ionicity of PILs, and the identification of methods for
synthesising highlyꢀionic PILs, is the range of methods that are used for their synthesis. PILs
have been synthesised using stoichiometric addition of neat acid to neat base, either with or
without an excess of base,15ꢀ16,20ꢀ22,24ꢀ31 and addition of acid to base with at least one component
in solution.15,31ꢀ33 To mitigate the risk of losing volatile components from reaction mixtures, the
importance of keeping reaction mixtures cold using waterꢀice baths or dry iceꢀacetone baths is
often stressed.15,34 It has been claimed that nonꢀstoichiometric PILs can be purified by
evaporating excess neutral molecules under vacuum,28 but others have warned against purifying
PILs by vacuum drying, in case it exacerbates nonꢀstoichiometry.11,16,35 Compounding the
problem is the absence of a reliable method for detect and quantify the concentration of unreacted
precursor species in PILs, making it difficult to compare different methods for synthesising PILs.
In this paper, we describe a simple voltammetric method that can be used to rapidly assess
the quality of a series of ammoniumꢀbased PILs, a class of PILs that is often used in
electrochemical studies and devices.6,36ꢀ37
We show significant differences in the
ultramicroelectrode (UME) voltammetry of the PILs diethylmethylammonium triflate
[dema][TfO], dimethylethylammonium triflate [dmea][TfO], triethylammonium trifluoroacetate
[tea][TfAc], and dimethylbutylammonium triflate [dmba][TfO] when prepared using different
5
ACS Paragon Plus Environment

Langmuir
Page 6 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Structures of the protic ionic liquids [dema][TfO], [dmea][TfO], [tea][TfAc], and
[dmba][TfO].
literature methods. We demonstrate that this difference is caused by the presence of a significant
concentration of excess parent acids in some PILs, particularly those formed by addition of neat
parent acids to neat bases. Using our method, we quantify the excess acids present in some PILs
and show that the acid can be extremely stable in PILs. For example, excess triflic acid, TfOH,
persists in [dema][TfO] even when the PIL is held under high vacuum (~10ꢀ7 mbar) for several
weeks. In contrast, quantification of excess bases in PILs is complicated by evaporation of the
bases from the PILs within hours, even at atmospheric pressure. We also demonstrate the
1
limitations of H NMR spectroscopy and conductimetry for measuring the concentration of
excess acids in these media and, based on our observations, we make a number of
recommendations for the synthesis of PILs with high ionicity. Finally, we show that the excess
acids can have a significant effect on the chemical and electrochemical properties of the liquids.
6
ACS Paragon Plus Environment

Page 7 of 34
Langmuir
1
2
3
4
5
6
Experimental
7
8
9
Materials and Apparatus. All reagents were from Alfa Aesar or Sigma Aldrich and were
used as received. H2 (99.995%), Ar (99.998%), O2 (99.998%), and N2 (99.998%) were from
BOC gases (Nottingham, UK). Au wire (0.5ꢀmm diameter, 99.9985%), Pt wire (0.5ꢀmm
diameter, 99.997%, and 0.025 mm diameter, 99.95%), and Pd wire (0.1 mm diameter, 99.99%)
were from Alfa Aesar. Electrochemical measurements were carried out using model 910B and
CH1140 potentiostats (CH Instruments, Austin, TX)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis and Characterization of Ionic Liquids. PILs were synthesised by adding the
appropriate parent acids to bases, while the bases were kept cold in ice baths. Three methods
were used to synthesise PILs: (A) neat acids were slowly added to a 5% molar excess of neat
base, (B) neat acids were added to a 15% molar excess of neat base, and (C) 1.0 M aqueous acids
were added to a 5% molar excess of neat base, and excess H2O was then removed using rotary
evaporation. After synthesis, all PILs were heated to 70 °C under vacuum (3 × 10−2 mbar) for 48
1
hr to remove excess water and base. PILs were characterised by H NMR spectroscopy, and
residual water contents were determined using Karl Fischer titrations (Mitsubishi CA100). In all
instances, residual water contents were below 200 ppm.
Electrode Fabrication and Electrochemical Measurements. The use of Pd/H reference
electrodes for electrochemistry in PILs was introduced by Angell.31 Pd/H electrodes provide
stable potentials in PILs, which are defined by the hydrogenꢀredox equilibrium at the interface
between the Pd and the PIL. Pd/H reference electrodes were fabricated by sealing Pd wires (~2
cm) into borosilicate glass using a butane flame. ~1 cm of the Pd wires was left protruding from
the glass sheaths to form Pd rod electrodes. Cu contact wires were attached to the Pd wires
inside the glass sheath using Ag epoxy (Chemtronics, Kennesaw, GA). The Pd rods were then
7
ACS Paragon Plus Environment

Langmuir
Page 8 of 34
1
2
3
4
5
6
7
8
9
annealed in a butane flame, then immersed in water, and H2 was bubbled over the metal surfaces
for 30 min. The rods were then rinsed with water and then dried under a flow of N2. 25ꢀꢂm Ptꢀ
disk UMEs were prepared by sealing 25ꢀꢂm diameter Pt wires in soft glass, and then grinding
and polishing the end of the UMEs until smooth Pt disks were exposed (using the methods
described in Reference 38). All PILs were saturated with Ar, H2, or O2 prior to carrying out
voltammetric measurements by bubbling the gas through the PILs for 10 minutes, and a blanket
of the appropriate of gas was maintained above the PILs during the measurements.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Voltammetry In Vacuo. Voltammetry was performed in vacuo using an ultraꢀhigh
vacuum (UHV) chamber with a base pressure of 2 × 10−8 mbar. An inꢀline mass spectrometer
(Vacuum Generator SX200, 1ꢀ200 amu quadrupole mass spectrometer, operated with an ESS
European Spectrometry Systems power supply and software) was positioned directly above the
electrochemical cell. The electrochemical cell contained a Pd/H reference electrode and Pt wire
working and counter electrodes that were connected to an external potentiostat. The Pd/H
reference electrodes gave a stable potential over the course of the experiments (52 days). A piece
of tubing, with a valve that could be externally operated, was placed over one of the electrodes to
collect gases emanating from the electrode. Calibration experiments were undertaken with the
valve either open, or closed then opened, in order to determine from which electrode H2 was
evolved (see Figure S1 in the Supporting Information for a schematic of the experimental setup).
Results and Discussion
Electrochemistry of PILs and Electroreduction of Excess Acids in PILs. Figure 1
shows cyclic voltammograms (CVs) recorded in three PILs synthesised using different methods.
In each case, the CVs were recorded using a 25ꢀꢂm diameter Pt UME, and the PIL was saturated
with Ar to remove any O2 from the liquids. In Figure 1B and 1F, waves due to Pt oxide
8
ACS Paragon Plus Environment

Page 9 of 34
Langmuir
1
2
3
4
5
6
7
8
9
formation can be seen at potentials, E, > 1.0 V and 0.6 V (these waves cannot be clearly resolved
in the other CVs due to the scale of the current).39 At the negative potential limit, a cathodic
current flowed in each case, due to reduction of the PIL cations.14,31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. CVs of a series of Arꢀsaturated PILs recorded using a 25ꢀꢂm diameter Pt UME. The
applied potentials were cycled between (A) −0.9 V (initial) and 1.6 V at 10 mV s−1, (B) −0.4 V
(initial) and 1.6 V at 50 mV s−1, (C) −0.35 V (initial) and 1.8 V at 50 mV s−1, (D) −0.45 V
(initial) and 1.8 V at 50 mV s−1, (E) −0.6 V (initial) and 1.6 V at 50 mV s−1, (F) −0.4 V (initial)
9
ACS Paragon Plus Environment

Langmuir
Page 10 of 34
1
2
3
4
5
and 1.2 V at 50 mV s−1. The insets in E and F show photographs of the PILs used to record the
respective data.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
However, Figure 1A, 1C, and 1E clearly show a peak labelled “a”, which does not appear in
Figure 1B, 1D, or 1F, so there was some difference between liquids that were nominally the
same, but which were synthesised using different methods. In general, we found that the use of
Synthesis A (addition of neat acid to 5% excess neat base) resulted in the appearance of peak a in
the voltammograms. As well as yielding this additional peak in the CVs, it is notable that the
PILs synthesised using different methods had different appearances. The insets of Figures 1E
and 1F show that samples of [dmea][TfO] synthesised using Syntheses A and C were brown and
colorless, respectively. In all of our syntheses, we found that deep coloration of our PILs
correlated with the presence of peak a during voltammetry. It is important to note that this work
is not the first to show the anomalous reduction current in CVs of similar PILs at Pt electrodes at
0 V (note that some workers use Pt “reversible hydrogen electrodes” consisting of H2 gas flowing
over Pt electrodes immersed in the PIL, but the potential of this reference electrode is within
about 50 mV of that of our Pd/H electrode, so the results are comparable).40ꢀ41 However, this
peak has thus far not been conclusively attributed to a specific electrochemical process. Angell
and coꢀworkers did observe these peaks previously during voltammetry of PILs but the PILs had
been saturated with H2.31 They suggested that the peak resulted from reduction of the acid
formed by H2 oxidation in H2ꢀsaturated PILs; when H2 is oxidised, the PIL anions are protonated
to reform the parent acid (Equation 2), which is reduced at E < 0 V (Equation 3):
H2 + 2A− ꢀ2HA + 2e−
(2)
10
ACS Paragon Plus Environment

Page 11 of 34
Langmuir
1
2
3
4
5
6
2HA + 2e− ꢀH2 + 2A−
(3)
7
8
9
where A− and HA represent the anion and parent acid of a generalised PIL, respectively.
However, while this description is reasonable for voltammetry in H2ꢀsaturated media, it does not
explain why cathodic waves are sometimes observed during voltammetry of H2ꢀfree PILs.
To identify the origin of the anomalous peak in our voltammograms, we first posited that
electroreduction of undissociated acid remaining in some asꢀsynthesised liquids may have been
responsible. Our reasoning was based on Angell’s observation of the peak in H2ꢀsaturated
PILs;31 given that the product of H2 oxidation (Equation 2) is the acid that is used to synthesise
the PIL in the first place, it seems reasonable that excess acid in the PILs should also be reduced
at this potential. To test whether reduction of excess acid in the PILs was responsible for the
appearance of the peak in some of the voltammograms, those liquids that exhibited the peak
during voltammetry were spiked with additional acid and reanalysed voltammetrically. Figures
2AꢀC show steadyꢀstate voltammograms of each PIL. Clearly, increasing the concentration of
the parent acid in each PIL caused an increase in the cathodic current flowing negative of E = 0
V, indicating that the currents were due to reduction of excess acid in each PIL (note that the
increase in cathodic current at −0.4 V in Figure 2C is due to the reduction of the PIL cations,
which occurred at a lower potential in [tea][TfEt] than in the [TfO]−ꢀbased PILs. The
concentration of acid in each PILs can be related to the steadyꢀstate current, iss, flowing through
the UME using Equation 4:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
iss = 4nFaDC
(4)
11
ACS Paragon Plus Environment

Langmuir
Page 12 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2.
Steadyꢀstate voltammograms of Arꢀsaturated PILs containing increasing
concentrations of parent acids. Applied potentials were swept from (A) 0.2 V to −0.6 V at 10
mV s−1, (B) 0.2 V to −0.4 V at 10 mV s−1 and (C) 0.2 V to −0.5 V at 5 mV s−1. The
voltammograms were recorded using a 12.5ꢀꢂm radius Pt UME at (A and C) room temperature
and (B) 55 °C. Figures D−F show plots of the steadyꢀstate cathodic current, iss, vs. the
concentration of added acid, with a dashed line of best fit.
12
ACS Paragon Plus Environment

Page 13 of 34
Langmuir
1
2
3
4
5
6
7
8
where n is the stoichiometric number of electrons transferred, F is the Faraday constant, a is the
radius of the UME, D is the diffusion coefficient of the acid, and C is its concentration. iss was
measured at each concentration of added acid and the resulting relationships are shown in Figures
2DꢀF. iss increased linearly as the concentration of added acid increased (in agreement with
Equation 4), and nonꢀzero intercepts are visible in each figure, due to the presence of acid in each
asꢀsynthesised PIL. Extrapolation of the lines of best fit in Figure 2DꢀF to iss = 0.0 A gives initial
concentrations of parent acid of 160 mM in [dema][TfO], 20 mM in [dmea][TfO], and 230 mM
in [tea][TfAc]. Consideration of Equation 4 demonstrates that D can be determined from the
slopes of the lines of best fit and the calculated values are DTfOH = 4.2 × 10−7 cm2 s−1 in
[dema][TfO], DTfOH = 9.2 × 10−7 cm2 s−1 in [dmea][TfO], and DTfAcH = 5.2 × 10−7 cm2 s−1 in
[tea][TfAc]. These values can be compared with values between 0.226 × 10−7 cm2 s−1 and 3.13 ×
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10−7 cm2 s−1 determined during
a
study of mass transport of hydrogen
bis(trifluoromethanesulfonyl)imide (which is larger than the acids used here) in series of aprotic
ionic liquids.42
Chronoamperometric experiments were undertaken to further test the accuracy of the
measured C and D values of the various parent acids in the asꢀsynthesised PILs. Figure 3Aꢀ3C
shows chronoamperograms of asꢀsynthesised [dema][TfO], [dmea][TfO], and [tea][TfAc] (those
used to record the voltammograms in Figures 2A, 2B, and 2C). In each case, the applied
potential was stepped from a value where reduction of the excess acid did not occur, to one where
the reduction of acid occurred at the diffusionꢀcontrolled rate. The decay in current, i, after a
potential step at a UME can be described by the ShoupꢀSzabo expression:43ꢀ44
13
ACS Paragon Plus Environment

Langmuir
Page 14 of 34
1
2
3
4
(5)
i = 4nFaDC f (t)
5
6
7
where f(t) is given by:
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
1
2
f (t) = 0.7854+0.8862τ− +0.2154exp(−0.7823τ−
)
(6)
Figure 3. (AꢀC) Chronoamperograms of Arꢀsaturated PILs recorded using a 12.5ꢀꢂm radius Pt
disk UME in [dema][TfO], [dmea][TfO], and [tea][TfAc] (solid lines). In A, the applied
potential was stepped from 0.6 V to −0.5 V in [dema][TfO] at 25 °C. In B, the applied potential
was stepped from 0.2 V to −0.4 V in [dmea][TfO] at 55 °C. In C, the applied potential was
stepped from 0.2 V to −0.3 V in [tea][TfAc] at 25 °C. The dashed lines show the bestꢀfits of the
experimental data to Equation 5. (D) Mass spectrometry traces (at m/z = 2) recorded when a Pt
electrode immersed in [dema][TfO] was swept from 1.8 V to −0.8 V at 50 mV s−1.
14
ACS Paragon Plus Environment

Page 15 of 34
Langmuir
1
2
3
4
5
6
7
and τ is given by:
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4Dt
τ
=
(7)
a2
Theoretical curves were generated using Equation 5 and the values of C and D determined
previously using voltammetry of each PIL. These fits are shown in Figure 3AꢀC (dashed lines)
and the agreement between the experimental and theoretical curves is excellent, giving us
confidence in our estimates of the C and D (above) for the parent acids in the asꢀsynthesised
PILs.
A further experiment was then undertaken, in order to confirm that the anomalous waves
in the CVs in Figures 1A, 1C, and 1E were due to the reaction shown in Equation 3. A sample of
[dema][TfO], to which 0.1 M TfOH was added, was inserted into a glass vial, along with
electrodes connected to a potentiostat. The assembled cell was then inserted into a vacuum
chamber connected to a mass spectrometer. Due to the very low vapor pressure of ionic liquids,
it is possible to study ionic liquids under high vacuum conditions.45 Figure 3D shows an increase
in the m/z = 2 signal at E < 0 V when E was swept from 1.8 V to −0.8 V, confirming that H2 was
produced at E < 0 V (our calibration experiments confirmed that the gas was evolved from the Pt
working electrode). Therefore, we can conclude that not only is excess acid present in some of
the PILs, but simple voltammetry can allow its quantification.
15
ACS Paragon Plus Environment

Langmuir
Page 16 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. (A) CVs of [dmba][TfO] recorded using a Pt UME at 50 mV s−1 in the presence and
absence of added TfOH. (B) CVs of [dema][TfO] recorded using a Pt UME at 50 mV s−1 in the
presence and absence of added TfOH. (C) Graph of the steadyꢀstate current for H2 evolution in
[dmba][TfO] during voltammetry at 5 mV s−1 vs. the concentration of added TfOH. (D) Graph of
the steadyꢀstate current for H2 evolution in [dema][TfO] during voltammetry at 5 mV s−1 vs. the
concentration of added TfOH. (E) Chronoamperogram of [dmba][TfO] recorded at −0.3V after
addition of 113 mM TfOH to the PIL (solid line). The dashed line shows the best fit of the
experimental data to Equation 5. (F) Chronoamperogram of [dema][TfO] recorded at −0.4 V
16
ACS Paragon Plus Environment

Page 17 of 34
Langmuir
1
2
3
4
5
after addition of 113 mM TfOH to the PIL (solid line). The dashed line shows the best fit of the
experimental data to Equation 5.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It is notable that not all of the PILs synthesised during our study showed the anomalous
voltammetric wave a. We found that [dmba][TfO] could be synthesised using any of the methods
and did not show any excess acidꢀreduction wave. Acidꢀfree [dema][TfO], on the other hand,
could only be reliably made using Synthesis C, and the solid lines in Figure 4A and 4B show
CVs of samples of [dmba][TfO] and [dema][TfO] synthesised using Method A and C,
respectively. No reduction wave is apparent near 0 V, but after addition of 23 mM TfOH,
reduction waves with onset potentials near 0 V appeared (labelled ‘a’ on each figure). The
concentration of acid in these PILs was gradually increased, and the resulting steadyꢀstate
cathodic currents are shown in Figure 4C and D as a function of the concentration of added acid.
From the slope of the bestꢀfit lines in Figure 4C and D, DTfOH in [dmba][TfO] is 1.59 × 10−7 cm2
s−1, and DTfOH in [dema][TfO] is 2.28 × 10−7 cm2 s−1, in reasonable agreement with that
determined previously for the asꢀsynthesised PIL containing excess acid. Chronoamperometry
was also used to study these liquids and Figure 4E and 4F show the chronoamperograms
recorded in each PIL containing 113 mM added TfOH. The dashed line in each figure shows the
theoretical curve generated using Equation 5 and values of CTfOH = 113 mM (the concentration of
acid added), DTfOH in [dmba][TfO] = 1.59 × 10−7 cm2 s−1, and DTfOH in [dema][TfO] = 2.28 ×
10−7 cm2 s−1. Clearly, the theoretical curves fit the experimental curves extremely well, giving us
confidence in our estimates of C and D and demonstrating that any excess acid in the asꢀ
synthesised PILs was at concentrations below the limit of detection of our method.
17
ACS Paragon Plus Environment

Langmuir
Page 18 of 34
1
2
3
4
5
6
7
8
9
In cases where measurable concentrations of acids persisted in PILs, which was generally
when Synthesis A was used, loss of volatile base during the reaction must have occurred due to
the exothermicity of the reaction, despite that fact that all of our reaction mixtures were kept cold
(in ice baths) during very slow dropwise addition of the acids. We also tried to use dry
ice/acetone baths to keep our reaction mixtures cold (as done previously21). During Synthesis A
(5% excess base), there was insufficient base present to mitigate loss of the volatile base during
the reaction, so excess acid generally resulted. Synthesis B, on the other hand, sometimes
resulted in the presence of excess acid. The lower propensity of this method to result in excess
acid in the asꢀprepared liquids is attributed to the higher initial concentration of bases in the
reaction mixtures, which mitigated loss or decomposition of some of the volatile base from the
reaction mixture. By contrast, we found that Synthesis C (addition of 1.0 M TfOH (aq) to 5%
molar excess of neat base, with subsequent removal of base and water under vacuum) reliably
produced stoichiometric PILs. The synthesis of PILs using aqueous acid/base mixtures has
previously been reported as resulting in PILs that contain unavoidably high water contents.16
This was not, however, our experience, with all PILs synthesised using Synthesis C containing
less than 200 ppm water after drying. Therefore, based on our observations, we recommend the
use of Synthesis C to those researchers attempting to synthesise highlyꢀionic PILs.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Detection of Unreacted Base in the PILs and the Prospects for Vacuum Purification of
PILs. While the bases used to form our PILs are electrochemically inactive within the
electrochemical window of the PILs, we show here that it is possible to indirectly detect the bases
within PILs using voltammetry. Oxidation of H2 in PILs yields protons that can then either be
accepted by anions of a PIL (Equation 2) or by any neutral base (Equation 8) present.
H2 + 2B → 2HB+ + 2e−
(8)
18
ACS Paragon Plus Environment

Page 19 of 34
Langmuir
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. (A) Pt UME CVs of H2ꢀsaturated [dmea][TfO] at 50 mV s−1 (dashed line). A Pt wire
counter electrode and Pd/H reference electrode completed the cell. The red line shows the CV
recorded after addition of 0.4 M dmea. (B) UME voltammograms (swept positive at 5 mV s−1) of
H2ꢀsaturated [dmea][TfO] using the same electrodes as in part A. 0.9 M dmea was added to the
PIL and the red CV (solid line) was recorded. The green (dotꢀdashed) and blue (dotted)
responses were recorded 17 and 84 min, respectively after addition of the base. The blank
response is shown by the black dashed line. All experiments with [dmea][TfO] were conducted
at 65 °C. (C) CVs recorded using a Ptꢀwire electrode at 50 mV s−1. The initial CV was recorded
19
ACS Paragon Plus Environment

Langmuir
Page 20 of 34
1
2
3
4
5
at atmospheric pressure, then at 2.6 × 10−7 mbar (dashed), and 9.5 × 10−8 mbar (dotꢀdashed). The
blue CV was recorded after bringing the chamber to ambient pressure using N2.
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Protonation of the base should occur at less positive potentials than protonation of the anions of
the PILs, as the anions are weak conjugate bases. Figure 5A shows CVs of H2ꢀsaturated
[dmea][TfO], before and after addition of the parent base dmea to the PIL. Oxidation of H2, with
subsequent protonation of the PIL anions, is visible as the anodic peak positive of 0.4 V in Figure
5A (dashed black line). After addition of 0.4 M dmea, a new oxidation process with an onset
near 0 V appeared (red solid line), corresponding to the reaction shown in Equation 8. This
anodic process was only observed when the PILs were saturated with H2 (see Figure S2A in the
SI). Oxidation of H2 and protonation of anions was still visible as a second oxidation wave
positive of 0.5 V. Therefore, the neutral base in the PIL can be monitored by measuring the
anodic currents at potentials below about 0.4 V. Figure 5B shows a voltammogram recorded in
the same PIL in the absence (dashed black line) and presence (red line) of dmea at 5 mV s−1 and
within a smaller electrochemical window for clarity. The green (dotꢀdashed) and blue (dotted)
lines show subsequent CVs recorded 17 min and 84 min, respectively, after addition of the base.
The anodic H2ꢀoxidation currents decreased in magnitude as time progressed, showing that dmea
(boiling point 36.5 °C, and vapor pressure above pure liquid of 495 mm Hg) rapidly evaporated
from the PIL, and that its loss could be monitored electrochemically. Similar patterns were
observed when using [dema][TfO] and [dmba][TfO]. In both of these liquids, new anodic peaks
were observed in voltammograms around 0 V after the addition of the respective bases, and these
peaks decreased in size as time progressed (Figures S2B and S2C in the Supporting Information).
On the other hand, when excess base was added to [tea][TfAc], no significant difference in the
20
ACS Paragon Plus Environment

Page 21 of 34
Langmuir
1
2
3
4
5
6
7
8
9
H2ꢀoxidation voltammograms was observed (Figure 1D in the Supporting Information). The
most likely explanation for this behaviour is that [tea][TfAc] always contains some tea due to the
lower driving force for proton transfer from TfAcH to tea, so addition of base had a negligible
effect on H2 oxidation in the PIL.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5C (solid black line) shows a CV of [dema][TfO] containing 0.1 M TfOH. This
voltammogram was recorded inꢀsitu in a high vacuum chamber under atmospheric pressure and
shows a peak for reduction of TfOH (with an onset potential near 0 V) in the negativeꢀgoing
sweep, and a subsequent H2ꢀoxidation peak in the positive sweep (as expected for this PIL). The
pressure in the vacuum chamber was then lowered to 10−7 mbar, and voltammograms were
recorded over the course of several days. As time progressed, the TfOHꢀreduction peak and H2ꢀ
oxidation peak decreased in size but, even after 50 days, some acid reduction was visible negative
of 0 V during voltammetry of the PIL (blue CV). Given that TfOH could not be removed
entirely from [dema][TfO] after almost 2 months at 10−7 mbar, the removal of excess TfOH using
a simple rotary pump setup will clearly not be possible within reasonable timeframes. Therefore,
contrary to what has been suggested in the literature,28 not all neutral species can be removed
from PILs using vacuum purification. While excess amines can be removed from our PILs
relatively readily, excess acid in PILs are very persistent, most likely a result of the strong
hydrogenꢀbonding ability of the acids.
Usefulness and Limitations of Walden Plots and NMR Spectroscopy for Studying the
Ionicity of PILs. So far, we have demonstrated the usefulness of electrochemical methods to
study excess parent acids and bases in PILs. To demonstrate the usefulness and limitations of
Walden plots for determining the presence of acids in PILs (and so compare this approach with
our electrochemical method), samples of [dema][TfO] and [dmba][TfO] were synthesised using
Method C, and their conductivity and viscosity are shown as part of the Walden plot in Figure 6.
21
ACS Paragon Plus Environment

Langmuir
Page 22 of 34
1
2
3
4
5
6
7
8
9
“Poor” analogues of each PIL were then synthesised by deliberately spiking the PILs first with
0.25 M of the parent acid, then with 0.50 M of the acid, and finally with deꢀionised water. The
conductivity and viscosity of these PILs are also included in Figure 6. The data recorded for the
asꢀsynthesised [dema][TfO] and [dmba][TfO] are within about 10% of the theoretical line
(measured vertically), in agreement with data reported previously for similar PILs, and meaning
that the ionicity of these PILs is below that of the fully dissociated electrolyte.17 Addition of the
parent acid (TfOH) to each PIL has a negligible effect on the position of the PIL on the Walden
plot (data points 1ꢀ3 and 6ꢀ8 respectively), demonstrating that the presence of significant
concentrations of the parent acid affected neither the conductivity nor the viscosity of the PILs.
However, adding water to PILs shifts their positions on the Walden plot towards the topꢀright
corner, following the gradient of the ideal line. This change in position on the plot demonstrates
that adding water to the PILs decreases their viscosities, and thereby increasing their
conductivity, but without changing the mechanism of ion conduction (a change in slope from the
diagonal, ideal line can indicate the occurrence of Grotthus hopping mechanism).14 Therefore, it
is clear that the use of Waldenꢀplot analysis of asꢀsynthesised PILs would not in itself provide
any insights into presence of excess acids.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Figure 7 shows H NMR spectra of [tea][TfAc] and [dmba][TfO] before and after
addition of acid and water. The top spectrum of Figure 7A is that of the asꢀsynthesised
[tea][TfAc] and shows a resonance at 9 ppm < δ < 10 ppm, which is attributable to the
exchangeable proton bound to the nitrogen of the ammonium cation (full spectra are shown in
Figures S3ꢀS12 of the Supporting Information).20 As increasing concentrations of TfAcH were
added to the PIL, a broad, lowꢀintensity peak appeared between 12 and 13 ppm. This peak is
22
ACS Paragon Plus Environment

Page 23 of 34
Langmuir
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Walden Plot of log of the molar conductivity versus log of the fluidity of [dema][TfO]
and [dmba][TfO]. Each PIL was synthesised according to synthetic method C in the
Experimental Section and conductivity and viscosity data were recorded at 40 °C. The
concentrations of added TfOH and H2O in the PILs were (1) 0 M TfOH, 0.013 M H2O. 2) 0.25
M TfOH, 1.054 M H2O. 3) 0.50 M TfOH, 1.053 M H2O. 4) 0.50 M TfOH, 5.063 M H2O. 5) 0.50
M TfOH, 7.876 M H2O. 6) 0 M TfOH, 0.093 M H2O. 7) 0.25 M TfOH, 0.165 M H2O. 8) 0.50 M
TfOH, 0.702 M H2O. 9) 0.50 M TfOH, 3.502 M H2O. 10) 0.50 M TfOH, 4.593 M H2O.
1
Figure 7. (A) H NMR spectra of (from top to bottom) asꢀsynthesised [tea][TfAc], and
[tea][TfAc] containing 1.4 M TfAcH, 2.2 M TfAcH, 3.1 M TfAcH, and 3.1 M TfAcH/15 M H2O.
1
(B) NMR spectra of (from top to bottom) asꢀsynthesised [dmba][TfO], and [dmba][TfO]
containing 1.6 M TfOH, 4.0 M TfOH, 5.5 M TfOH, and 5.5 M TfOH/15 M H2O.
23
ACS Paragon Plus Environment

Langmuir
Page 24 of 34
1
2
3
4
5
6
7
8
9
associated with the exchangeable proton associated with TfAcH, and the broadness of the peak is
likely due to this proton exchanging between the acid and the base. Even at 3.1 M added TfAcH,
the peak between 12 and 13 ppm was very broad and low in intensity, and lower concentrations
1
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
of excess acid could not be reliably detected using H NMR spectrocopy. The usefulness of H
NMR spectroscopy in studying excess acid in this PIL is complicated further by the presence of
water. The bottom spectrum of Figure 7A is that of [tea][TfAc] containing 3.1 M TfAcH and 15
M H2O. Comparison with the spectrum of the PIL containing 3.1 M TfAcH and no added water
reveals that the peak between 12 and 13 ppm disappeared upon addition of H2O, and a large peak,
appeared at δ ~ 5ppm. In the presence of H2O, the proton resonates between the acid and the
H2O on a faster timescale than the NMR timescale, resulting in the single peak. Therefore, it is
not possible to differentiate between excess acid in the PIL and water in the PIL. Given that water
is an extremely common contaminant impurity in PILs (and RTILs in general), this is especially
problematic.
1
Figure 7B shows H NMR spectra of asꢀprepared [dmba][TfO], and [dmba][TfO]
containing various amounts of added TfOH and H2O. As was the case for the [tea][TfAc]
spectra, a single resonance due to the exchangeable proton of the [dema]+ cation is observed at 9
ppm < δ < 10 ppm. In the spectrum of the asꢀprepared [dmba][TfO], an additional peak is
observed at δ = 3.4 ppm and associated with trace water in the sample. However, when TfOH
was added to the [dmba][TfO], the peak at 3.4 ppm disappeared and new peaks appeared, which
increased in δ (up to about 10.5 ppm) as the concentration of acid increased. These peaks can be
attributed to protons exchanging between TfOH and the H2O in the samples. As more TfOH was
added, the protons sat predominantly on the acid, shifting the peak to higher δ. However, when
water was deliberately added to the sample (bottom spectrum of Figure 7B), the peak shifted to a
lower δ, as the protons sat predominantly on the water.
24
ACS Paragon Plus Environment

Page 25 of 34
Langmuir
1
2
3
4
1
In summary, it is clear that excess acids can be identified in each PIL using H NMR
5
6
1
spectroscopy, but the usefulness of H NMR spectroscopy for quantitatively studying such
7
8
9
species is complicated by the presence of the protic impurity H2O, a phenomenon that has been
recognised previously.21 This limitation is a particular disadvantage, given that water is one of
the most common contaminants in PILs (and RTILs in general).32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Effects of Unreacted Acid in PILs. Given that we have demonstrated that some
syntheses of PILs can results in the presence of excess acid, and that the acid is very stable and
detectable using voltammetry, an important question is whether excess acids in PILs have a
significant effect on the properties of PILs. A simple visual demonstration of the effects of
excess acids in PILs is shown in Figure 8. 8 mM ferrocene was added to each of the PILs used in
our study and, in those PILs containing excess acid, ferrocene was oxidised, as shown by the
change in the color of the solution from yellow to blue (most clearly evident in the [TfO]−ꢀbased
liquid and confirmed by cyclic voltammetry).
We studied whether excess acid could have an effect of the electrochemical properties of
PILs by studying the oxygen reduction reaction (ORR; O2 + 4e− + 4H+ → 2H2O) in the PILs.
The ORR is the cathode reaction in lowꢀtemperature fuel cells and has been studied in a range of
RTILs and PILs, both due to its fundamental importance as an example of an electrocatalytic
reaction, and as some researchers propose the development of PILꢀbased intermediateꢀ
temperature fuel cells.6,41,46ꢀ47 Figure 9 shows linear sweep ORR voltammograms recorded in O2ꢀ
saturated neat, acidꢀfree [dema][TfO] (blue) and acidified [dema][TfO] (red). Note that
potentials in this figure are reported vs. that of the ferrocene/ferrocinium redox couple (visible at
0 V) to avoid complications from changes in the acidity of the solution on the potential of the
25
ACS Paragon Plus Environment

Langmuir
Page 26 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. (From left to right) [dmba][TfO], [dmba][TfO] + 0.1 M TfOH, [tea][TfAc], [tea][TfAc]
+ 0.1 M TfAcH, [dema][TfO], [dema][TfO] + 0.1 M TfOH, [dmea][TfO], [dmea][TfO] + 0.1 M
TfOH. All solutions contained 8 mM ferrocene.
Figure 9. Linear sweep voltammograms recorded in [dema][TfO] at a 25ꢀꢂm diameter Pt UME.
Potentials were swept negative at 5 mV s−1. Dashed lines show the responses obtained using O2ꢀ
saturated PILs and solid lines show those recorded using Arꢀsaturated PILs. The blue lines show
the responses recorded in the absence of excess TfOH and the red lines show those recorded in
[dema][TfO] containing 107 mM TfOH.
Pd/H reference electrode (for reference, the ORR onset potential in the PIL is about 1.0 V
positive of that of the H2ꢀoxidation reaction in the PIL47). When the proton donor changed from
the [dema]+ cation (in the neat liquid) to TfOH (in the acidified liquid), the ORR onset potential
shifted positive by about 0.8 V, from about −0.2 V to about 0.6 V. Friesen and coꢀworkers have
previously shown that the ORR onset potential in an aprotic RTIL (2,3ꢀdimethylimidazolium
trifluoromethanesulfonate) shifts positive by about 0.9 V when changing the pKa of proton donors
26
ACS Paragon Plus Environment

Page 27 of 34
Langmuir
1
2
3
4
5
6
7
8
9
from 10 (that of [dema]+) to −14 (that of TfOH).48 Therefore, our data are in good agreement
with that observed previously when using aprotic RTILs, and illustrate the importance of
quantifying, and understanding the effects of, excess acids in PILs before using these media as
solvents for electrochemical measurements.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusions
We have demonstrated the usefulness of a simple ultramicroelectrodeꢀvoltammetry
method for comparing the properties of PILs synthesised using several methods. The exothermic
addition of neat acid to neat base can result in the loss or decomposition of the basic component
of the PILs and electrochemical analysis reveals that the acid TfOH can be extremely stable in
nonꢀstoichiometric PILs, even under high vacuum. In contrast, excess amines readily evaporate
from [TfO]−ꢀbased PILs. The use of dry ice/acetone baths to keep reaction mixtures cold does
not always prevent the loss of volatile bases from reaction mixtures. Therefore, based on our
observations, we recommend the synthesis of [TfO]−ꢀbased PILs using a solutionꢀbased
approach, in which at least one component of the reaction mixture is dissolved in water. The
excess water can then be readily removed under vacuum. We have also shown that the presence
of unreacted acid dramatically alters the properties of PILs, including causing a 0.8 V shift in the
onset potential for electrocatalytic oxygen reduction in the PIL [dema][TfO]. Thus, it is vital that
those researchers interested in electrocatalysis in PILs (where onset potentials are often used as a
measure of electrocatalytic performance) consider the presence of excess acids and bases in their
liquids, and we hope that the methods described here will be of aid when studying such
phenomena.
27
ACS Paragon Plus Environment

Langmuir
Page 28 of 34
1
2
3
4
Supporting Information
5
6
7
8
The Supporting Information is available free of charge on the ACS Publications website.
A schematic of the apparatus used for mass spectrometric analysis of evolved gas;
hydrogenꢀoxidation voltammograms of PILs after addition of excess amines; NMR spectra of
PILs.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgements
We thank the Engineering and Physical Sciences Research Council for funding through
Project EP/P002382/1 and the Centre for Doctoral Training in Fuel Cells and their Fuels (Project
EP/L015749/1).
References
1. Wilkes, J. S., A Short History of Ionic Liquids ꢀ from Molten Salts to Neoteric
Solvents. Green Chem. 2002, 4, 73ꢀ80.
2. Hardacre, C.; Parvulescu, V., Catalysis in Ionic Liquids: From Catalyst Synthesis to
Application. Royal Society of Chemistry: Cambridge, 2014.
3. Wasserscheid, P.; Welton, T., Ionic Liquids in Synthesis. WileyꢀVCH Verlag & Co.:
Weinheim, 2008; Vol. 2.
4. Abbott, A. P.; McKenzie, K. J., Application of Ionic Liquids to the Electrodeposition
of Metals. Phys. Chem. Chem. Phys. 2006, 8, 4265ꢀ4279.
5. Wei, D.; Ivaska, A., Applications of Ionic Liquids in Electrochemical Sensors. Anal.
Chim. Acta 2008, 607, 126ꢀ135.
28
ACS Paragon Plus Environment

Page 29 of 34
Langmuir
1
2
3
4
5
6
7
8
9
6. Lee, S.ꢀY.; Ogawa, A.; Kanno, M.; Nakamoto, H.; Yasuda, T.; Watanabe, M.,
Nonhumidified Intermediate Temperature Fuel Cells Using Protic Ionic Liquids. J. Am. Chem.
Soc. 2010, 132, 9764ꢀ9773.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7. Lewandowski, A.; SwiderskaꢀMocek, A., Ionic Liquids as Electrolytes for LiꢀIon
Batteriesꢀan Overview of Electrochemical Studies. J. Power Sources 2009, 194, 601ꢀ609.
8. Ueno, K.; Tokuda, H.; Watanabe, M., Ionicity in Ionic Liquids: Correlation with Ionic
Structure and Physicochemical Properties. Phys. Chem. Chem. Phys. 2010, 12, 1649ꢀ1658.
9. Nuthakki, B.; Greaves, T. L.; Krodkiewska, I.; Weerawardena, A.; Burgar, M. I.;
Mulder, R. J.; Drummond, C. J., Protic Ionic Liquids and Ionicity. Aust. J. Chem. 2007, 60, 21ꢀ
28.
10. Sun, X.; Liu, S.; Khan, A.; Zhao, C.; Yan, C.; Mu, T., Ionicity of AcetateꢀBased
Protic Ionic Liquids: Evidence for Both Liquid and Gaseous Phases. New J. Chem. 2014, 38,
3449ꢀ3456.
11. Greaves, T. L.; Drummond, C. J., Protic Ionic Liquids: Evolving Structure–Property
Relationships and Expanding Applications. Chem. Rev. 2015, 115, 11379ꢀ11448.
12. Greaves, T. L.; Drummond, C. J., Protic Ionic Liquids: Properties and Applications.
Chem. Rev. 2008, 108, 206ꢀ237.
13. Angell, C. A.; Byrne, N.; Belieres, J. P., Parallel Developments in Aprotic and Protic
Ionic Liquids: Physical Chemistry and Applications. Acc. Chem. Res. 2007, 40, 1228ꢀ1236.
14. Yoshizawa, M.; Xu, W.; Angell, C. A., Ionic Liquids by Proton Transfer: Vapor
Pressure, Conductivity, and the Relevance of ꢀpKa from Aqueous Solutions. J. Am. Chem. Soc.
2003, 125, 15411ꢀ15419.
29
ACS Paragon Plus Environment