close

Вход

Забыли?

вход по аккаунту

?

acs.jpcb.7b08523

код для вставкиСкачать
Subscriber access provided by the Henry Madden Library | California State University, Fresno
Article
Towards the Elucidation of the Competing Role of Evaporation
and Thermal Decomposition in Ionic Liquids: A Multitechnique
Study of the Vaporization Behaviour of BMImPF (1-Butyl-3Methylimidazolium Hexafluorophosphate) Under Effusion Conditions
6
Valeria Volpe, Bruno Brunetti, Guido Gigli, Andrea Lapi, Stefano Vecchio Ciprioti, and Andrea Ciccioli
J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b08523 • Publication Date (Web): 18 Oct 2017
Downloaded from http://pubs.acs.org on October 25, 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.
The Journal of Physical Chemistry B 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 42
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
The Journal of Physical Chemistry
Towards the Elucidation of the Competing Role of Evaporation and Thermal
Decomposition in Ionic Liquids: a Multitechnique Study of the Vaporization
Behaviour of BMImPF6 (1-butyl-3-methylimidazolium hexafluorophosphate)
under Effusion Conditions
V. Volpe†, B. Brunetti#, G. Gigli†, A. Lapi†,$, S. Vecchio Ciprioti§, A. Ciccioli†,*
†
#
Dipartimento di Chimica, Sapienza Università di Roma, P.le A. Moro 5, I-00185, Rome, Italy
Istituto per lo Studio dei Materiali Nanostrutturati, CNR, c/o Dipartimento di Chimica, Sapienza Università di Roma,
P.le A. Moro 5, I-00185, Rome, Italy
$
Istituto CNR di Metodologie Chimiche (IMC-CNR), Sezione Meccanismi di Reazione, c/o Dipartimento di Chimica,
Sapienza Università di Roma, P.le A. Moro 5, I-00185, Rome, Italy
§
Dipartimento S.B.A.I., Sapienza Università di Roma, via del Castro Laurenziano 7, I-00161, Rome, Italy
* Corresponding author. E-mail address andrea.ciccioli@uniroma1.it
Abstract
The evaporation/decomposition behaviour of the imidazolium ionic liquid 1-butyl-3-methylimidazolium
hexafluorophosphate (BMImPF6) was investigated in the overall temperature range 425 - 551 K by means of the
molecular-effusion-based techniques Knudsen Effusion Mass Loss (KEML) and Knudsen Effusion Mass Spectrometry
(KEMS), using effusion orifices of different size (from 0.2 to 3 mm in diameter). Specific effusion fluxes measured by
KEML were found to depend markedly on the orifice size, suggesting the occurrence of a kinetically delayed
evaporation/decomposition process. KEMS experiments revealed that other species are present in the vapor phase
besides the intact ion pair BMImPF6(g) produced by the simple evaporation BMImPF6(l) = BMImPF6(g), with relative
abundances depending on the orifice size – larger the orifice, larger the contribution of the BMImPF6(g) species. By
combining KEML and KEMS results, the conclusion is drawn that in the investigated temperature range, when small
effusion orifices are used a significant part of the mass loss/volatility of BMImPF6 is due to molecular products formed
by decomposition/dissociation processes rather than to evaporated intact ion pairs. Additional experiments performed
by non-isothermal Thermogravimetry-Differential Thermal Analysis (TG-DTA) further support the evidence of
simultaneous evaporation/decomposition, although the conventional decomposition temperature derived from TG
curves is much higher than temperatures covered in effusion experiments. Partial pressures of the BMImPF6(g) species
o
were derived from KEMS spectra and analyzed by the second- and third-law methods giving a value of ∆evapH298
=
K
145.3 ± 2.9 kJ mol–1 for the standard evaporation enthalpy of BMImPF6. A comparison is done with the behavior of the
1-butyl-3-methylimidazolium bis(trifluoromethyl)sulfonylimide (BMImNTf2) ionic liquid.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Introduction
The exceptionally low volatility compared to common molecular organic liquids is one of the most
distinctive property of ionic liquids (ILs), making these substances attractive as potential substitutes
of volatile solvents and in other applications.1-4 As a consequence, in recent years the determination
of vapor pressure of ILs has been the subject of considerable efforts.5-11 Besides the practical
interest in determining the volatility of these materials as a function of temperature, vaporization
studies are also aimed at measuring the evaporation enthalpy for the more fundamental purpose of
evaluating the cohesion energy in the liquid phase, so as to adjust/validate force field parameters for
molecular dynamics calculations.12-18 Nevertheless, absolute vapor pressure data for ILs are still
scarce and uncertain, mostly limited to few classes of stable compounds, such as those with the
bis(trifluoromethyl)sulfonylimide (NTf2) anion.7-9
One of the main problems with vapor pressure measurements of ILs is that, though thermally stable
compared to molecular solvents, they can start to decompose at the same temperatures where their
vapor pressure become measurable.17,19-24 This has prevented vapor pressure of several ILs from
being measured25,26 and probably limits the accuracy of a number of data presently available in the
literature. If volatile products are formed during thermal decomposition, techniques such as
thermogravimetry and transpiration, which are blind with respect to the composition of the vapor
phase, might provide unreliable results even if decomposition occurs to a small extent. Moreover,
the occurrence of decomposition processes can affect the accuracy of other high temperature
measurements, such as those of heat capacity and other thermophysical properties of the liquid.
1-butyl-3-methylimidazolium hexafluorophosphate (in the following BMImPF6) is one of the most
common aprotic ILs with an imidazolium cation. In two of the first papers on the vaporization
behavior of ILs, Kabo and co-workers25,27 reported Knudsen Effusion thermogravimetric
experiments that they argued to provide results unsuitable for deriving vapor pressure and
evaporation enthalpy, due to the occurrence of thermal decomposition phenomena. No further
ACS Paragon Plus Environment
Page 2 of 42
Page 3 of 42
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
The Journal of Physical Chemistry
vaporization study on this IL was carried out until last year, when Zaitsau et al.11 succeeded in
measuring the vapor pressure of BMImPF6 with a Quartz Crystal Microbalance apparatus (QCM) at
very low temperatures (403 – 450 K), where authors assume decomposition to be negligible.
The use of mass spectrometry-based vaporization techniques can be of great help in such cases,
because information on the composition of the vapor can be obtained.9,10,19,21,28-31 In this paper, the
vaporization of BMImPF6 was reconsidered and new measurements were performed, by using both
the Knudsen Effusion Mass Loss and the Knudsen Effusion Mass Spectrometry (KEML and
KEMS, respectively) techniques, with the goal to investigate the competition between simple
evaporation and thermal decomposition. Measurements were done with effusion cells having
different orifice size to study the dependence of specific mass loss rate and vapor phase
composition from the extent of the effusing flow and from the closeness to equilibrium conditions.
The MS analysis of the vapor allowed us to study the evaporation process as a function of
temperature regardless the possible simultaneous decomposition. Additional measurements were
carried out by non-isothermal TG-DTA, using a protocol recently proposed by Heym,32 which is
able to reveal rapidly the occurrence of thermal decomposition simultaneous to evaporation.
Finally, by exploiting the thermal functions available in the literature for BMImPF6 both in the
gaseous25 and in the liquid33-40 phase, an experimental value for the evaporation enthalpy of
BMImPF6 is proposed, based on the third-law analysis of partial pressure data derived by KEMS. A
comparison with the prototypical IL 1-butyl-3-methylimidazolium
bis(trifluoromethyl)sulfonylimide (BMImNTf2), whose vaporization behaviour was recently studied
by our group9 with the same techniques used here, is presented.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 4 of 42
Experimental
The BMImPF6 sample was purchased from Iolitec. Sample purity (99%) was checked by HNMR.
Knudsen Effusion Mass Loss (KEML)
In the classic Knudsen Effusion Mass Loss (KEML) technique the mass of the sample is
monitored and the mass loss rate ∆m/∆t measured at each given constant temperature. If the vapor is
composed only of one species, the pressure in the effusion cell (peffusion) is evaluated from the wellknown relation41
peffusion =
∆m T
1 ∆m 2πRT
2πRT
=J
=K
Wo Ao ∆t
M
M
∆t M
(1)
where T is the temperature, W o the Clausing factor (transmission probability), A o the area of the
effusion hole, M the molar mass of the vapor species, and J the specific flux. K is a constant
depending on the geometrical characteristics of the effusion hole and can be measured by
calibration with known substances. If more than one species is present in the vapor, eq 1 gives the
total pressure provided that the appropriate average value is used for M41 (see below, eq 9).
Due to the presence of the hole, peffusion obtained according to eq 1 generally does not coincide with
the true equilibrium pressure p. According to the analysis of Motzfeldt,42 the two quantities can be
related by the following expression:
 W A 
p = p effusion 1 + o o 
α A

(2)
ACS Paragon Plus Environment
Page 5 of 42
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
The Journal of Physical Chemistry
where A is the vaporizing surface and α is the so-called vaporization coefficient. If the vaporization
process is not kinetically hindered, α is unity and a small value of the A o /A ratio makes the
peffusion/p ratio unity for most purposes, regardless the orifice size. On the contrary, if a significant
kinetic delay does exist, α can be much less than 1 and the peffusion/p ratio can be sizably less than
unity.
The apparatus used for KEML measurements is a home-modified Ugine-Eyraud B60 Setaram
thermobalance described in detail in ref. 9. Briefly, an effusion cell made from alumina is inserted
into a capped copper cylinder. Cell lids were made from pyrophyllite, with effusion orifices of 0.2,
1, and 3 mm in diameter. IL samples were loaded directly into the alumina cell or into a smaller Pt
crucible placed inside it. A vertical hole is made on the upper border in the lateral thickness of the
copper container, where a Pt100 platinum resistance thermometer is inserted from above for
temperature measurement. The effusion source is surrounded by a quartz tube and externally by a
tubular resistance of graphite as heating element. To test the presence of volatile impurities, all the
samples were kept under vacuum at approximately 420 K for 24 h before measurements, with no
significant mass loss observed. During the experimental runs, temperature was changed by a steplike program and the mass loss rates were evaluated at each constant temperature. For each cell the
measurements were carried out by varying the temperature two or three times from the highest to
the lowest value. Cells were calibrated by vaporizing very pure benzoic acid (Sigma-Aldrich, for
calorimetrical determination), whose vapor pressure is well-known.43,44 The following values of the
constant K in eq 1 have been determined for the three cells, respectively: K3 mm = (3.92 ± 0.16) 106
m-1·s-1 kg1/2 ·(mol·K)-1/2, K1 mm = (1.88 ± 0.06) 107 m-1·s-1 kg1/2 ·(mol·K)-1/2and K0.2 mm = (6.84 ±
0.16) 108 m-1·s-1 kg1/2 ·(mol·K)-1/2, where the error on the instrument constant is taken as the
standard deviation (1σ) of values obtained in various calibration runs at various temperatures.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 6 of 42
Knudsen Effusion Mass Spectrometry (KEMS)
In the Knudsen Effusion Mass Spectrometry (KEMS) technique,45 the effusing vapors are
analyzed with a mass spectrometer. The apparatus used here was described in previous papers.9,46
Briefly, it is based on a single focusing 90° magnetic sector mass spectrometer. The effusing vapor
species are ionised by impact with an electron beam whose energy can be changed continuously
from appearance of each ion up to 80 eV. Ion current for a given m/e value (single ion mode) is
measured by an electron multiplier. A movable shutter is interposed between the effusion orifice
and entrance to the ionization area to subtract background contributions from the total ion current of
+
a given ion. The background-subtracted ion intensity of the isotope n of the species X, I n X , can be
converted into the partial pressure of X inside the cell by the equation45
PX =
Kinstr I n+X T
σ X γ n an
X
(3)
X
where σX is the ionization cross section, γ n X the electron multiplier gain, a n X the isotopic
abundance, and Kinstr a constant which depends on geometry of the apparatus, cell features and
operating conditions. The multiplier gain was assumed to be proportional to the square root of the
molecular mass, a dependence well documented in KEMS studies.45 The instrumental constant was
measured ex situ by repeated vaporization experiments of pure zinc (source and purity) (σZn = 5.0
10-20 m2).45 The Knudsen source of the KEMS apparatus used in this study is heated by irradiation
from a spiral-shaped tungsten resistor surrounded by several tantalum shields to keep the
temperature uniform. Alumina (orifice diameter 0.5 and 1 mm) and platinum (orifice diameter 1
mm) effusion cells were used, inserted in an outer tantalum crucible. The temperature of the cell
was measured with a W-Re/W-Re 5/26% thermocouple inserted in the bottom of the tantalum
container. The heating system of our KE source requires rather long thermal equilibration times in
ACS Paragon Plus Environment
Page 7 of 42
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
The Journal of Physical Chemistry
the low-temperature regime of interest for studies on ILs. In order to avoid prolonged heating times
(that would favour thermal decomposition), we performed a number of different runs on fresh
sample rather than one or two prolonged runs, and relatively few data were collected for each fresh
loaded sample.
The intensity vs. electron energy curves (IEC, Ionization Efficiency Curves) were registered for
the most important ion species to estimate their appearance energy. As a rule, measurements were
carried out with an electron energy corresponding to the maximum of the ionization efficiency
curve of each ion.
Thermogravimetry/Differential Thermal Analysis (TG-DTA)
Simultaneous thermogravimetry and differential thermal analysis (TG-DTA) experiments were
carried out following a protocol recently proposed by Heym to tackle the problem of simultaneous
evaporation/decomposition.32 In general, the mass loss rate due to the simultaneous occurrence of
evaporation and thermal decomposition in a liquid during a non-isothermal TG experiment under a
flowing inert gas atmosphere may depend by many factors.32,47-48 If thermal decomposition obeys a
1st order reaction, the mass loss rate is directly proportional to the sample mass, and independent of
the crucible’s shape, the type of gas and its flow rate. By contrast, the evaporation rate significantly
depends on the size of the liquid–gas-interphase as well as on the diffusion coefficient of the vapor
transported by the carrier gas. As a consequence, the type of gas used may influence the evaporation
rate. On this basis, Heym and co-workers proposed recently32 a protocol for a fast check of the
occurrence of evaporation with or without simultaneous decomposition.
TG-DTA measurements were carried out using a Stanton-Redcroft STA1500 apparatus,
operating through a Rheometric Scientific system interface controlled by the RSI Orchestrator
software. Liquid samples of comparable size (about 20-22 mg) have been placed into an open
cylindrical 70 µL platinum crucible using an insulin syringe, while an open empty identical crucible
was used as reference. Comparable amounts of this liquid sample were subjected to two identical
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 8 of 42
kind of heating experiments at a low heating rate (2 K·min−1) from room temperature to 873 K
using, alternatively, two different inert gas carriers at a flow rate of 50 ml min−1: Ar and N2. Three
replicates were carried out using each inert gas atmosphere and very pure indium reference sample
have been used for calibration of temperature, obtaining a final standard uncertainty (1σ) of ± 0.1
K.
Results and Discussion
KEML experiments: mass loss rates
In KEML experiments, the IL sample was placed in a platinum crucible, in turn placed in an
alumina Knudsen cell. Some measurements were also done putting the sample directly in the
alumina cell, but in this case creeping phenomena were observed that occasionally caused the liquid
to come out from the cell. In the following, only the results obtained with the platinum crucible will
be presented.
The mass loss rates per unit orifice area corrected by the Clausing factor (J, specific effusion flux),
i.e. the term ( A°W ° )−1 ∆ m ∆ t of eq 1 (proportional to p effusion
MT
−1
), are reported in Table 1. If no
kinetic hindrance occurs, this quantity should not depend on the effusion orifice characteristics.
However, in our case a remarkable dependence is found, which calls for an explanation. By
analysing Table 1 and Figure 1 it can be argued that the evaporation flux of BMImPF6 measured
with the 1 mm orifice is about twice that with the 3 mm orifice and more than one order of
magnitude lower than that measured with the smallest orifice of 0.2 mm. In Figure 1, the specific
flux measured9 for the prototypical IL BMImNTf2 is also reported for comparison. In the latter, no
orifice-dependence was observed. The specific flux of BMImPF6 measured with the 1 mm and 3
mm orifices is 10 to 30 times lower than BMImNTf2. However, the strong increase of the
ACS Paragon Plus Environment
Page 9 of 42
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
The Journal of Physical Chemistry
BMImPF6 flux with the 0.2 mm orifice makes the volatility of the two compounds very similar
(about 20% higher for BMImPF6) under closer-to-equilibrium conditions.
Several evidences49-53 were presented in the literature that imidazolium ILs vaporize, mainly or
exclusively, as individual neutral ion pairs (NIP), according to the following evaporation process:
BMImNTf2(l) = BMImNTf2(g)
(4)
The concentration of charged species in the vapor was recently confirmed to be very low.31
Although the direct experimental detection of intact ion pairs in the vapor has been reported only
for few particularly stable species,52-53 and actually the existence of higher aggregates was
suggested by some theoretical evaluations,50,54-55 the evaporation process 4 is commonly accepted
for ILs having weakly nucleophilic anions. In the case of BMImNTf2, KEMS results are consistent
with this view, as reported in our previous study9 and confirmed in the present paper, where further
KEMS measurements on this benchmark IL were carried out for sake of comparison (see next
section).
The behaviour here observed for BMImPF6 (Figure 1) seems not consistent with this simple picture.
Such a strong dependence of the effusion flux on the orifice size would indicate that the
vaporization process is kinetically hindered and saturation in the gas phase is not achieved because
vapor species escape from the effusion orifice with a flux higher than that leaving the evaporating
surface, in accordance with eq 2. This phenomenon is unlikely to occur for the simple evaporation
4, whereas it has been reported in a number of cases where vapor species are not present as such in
the condensed phase and are formed through a decomposition process.56-57 Therefore, the results of
KEML measurements suggest that, in the explored temperature range and under effusion
conditions, BMImPF6 undergoes thermal decomposition accompanied by the release of volatile
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 10 of 42
products. As mentioned in the introduction, this hypothesis was already put forward in previous
studies.25,27 As we will discuss in the next section, now it is also supported by mass spectrometric
results.
As shown by eq 1, a strictly linear trend of ln J vs 1/T is not expected a priori, more so because in
the presence of decomposition + evaporation processes, M in eq 1 may depend on temperature.
However, as a matter of fact, the data shown in Figure 1 display good linear trends fitted by the
following equations
ln J 3 mm / kg s −1m − 2 = (24 .504 ± 0 .283 ) −
ln J 1 mm / kg s −1m − 2 = (24 .220 ± 0 .499 ) −
ln J 0.2 mm / kg s −1m −2 = (21 .483 ± 0.275 ) −
(17003
± 134 )
T /K
(452-494 K)
(5)
(16536
(478-519 K)
(6)
(505-551 K)
(7)
± 248 )
T /K
(13905 ± 145 )
T /K
The slope of these lines is found to decrease with decreasing the orifice size, probably due to the
superimposition of different processes with different enthalpy changes.
TG-DTA experiments
Although the occurrence of decomposition in liquid BMImPF6 was already reported in the literature
on the basis of non-isothermal TG measurements, the decomposition onset temperatures derived
from TG-DTA curves are in the 622 – 709 K range,23-24 much higher than temperatures explored in
the present study by effusion techniques. While the TG-DTA decomposition temperatures are of
interest for practical applications of ILs, indicating the onset of massive decomposition, according
to the results presented in the previous and in the next section of this paper, thermal degradation
actually begins at much lower temperatures, making difficult to study the simultaneous non-
ACS Paragon Plus Environment
Page 11 of 42
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
The Journal of Physical Chemistry
decompositive evaporation by the tensimetric techniques in which the composition of the vapor
phase cannot be monitored.
As described in the experimental section, our TG-DTA experiments were done following the
method proposed by Heym,32 in which two identical non-isothermal TG/DTA experiments are
carried out using different flowing inert carrier gases (Ar and N2 in this case) at a moderate heating
rate. On comparing the corresponding DTG curves (TG first derivative), two alternative conclusions
may be drawn about the process which takes place: thermal degradation, if the two curves are
practically superimposable, simple evaporation, if they are remarkably different. These results could
be not exhaustive in the case of simultaneous and comparable evaporation/decomposition processes.
The TG, DTA and DTG curves measured for BMImPF6 at a heating rate of 2 K min−1 under Ar and
N2 atmospheres are shown in Figure 2. Apparently, a single step of mass loss is displayed and no
difference is evident between the curves recorded using the two carrier gas up to about 473 K, i.e.
within the temperature range covered in our KEML and KEMS experiments. When mass loss
becomes sizeable (around 523 K both in Ar and in N2 atmospheres) a slight difference emerges in
the three plots: the sample temperature in N2 is shifted towards higher temperatures up to 80% of
mass loss, with extrapolated onset temperatures equal to 634.8 and 642.5 K for the TG curves in
flowing Ar and N2, respectively, consistent with literature data.23,24 In the final part (from about 723
K) the two curves of each plot are practically superimposed. This behavior is consistent with the
simultaneous occurrence of evaporation and thermal decomposition, with the latter more and more
prevailing with increasing temperature. Although this method cannot provide information in the
low-temperature range owing to the too low mass loss, the DTG curves indicate a close competition
between evaporation and thermal decomposition, in accordance with the results obtained at lower
temperature by KEML and KEMS, with decomposition finally prevailing at temperature above 723
K. However, in matching TG and KEML-KEMS results it should be emphasized that the
experimental conditions are extremely different in the two cases. In particular, TG measurements
are performed in open crucible under a flowing carrier gas atmosphere, whereas KEML-KEMS
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
experiments under high vacuum in effusion regime. From the observed dependence of the specific
effusing flux on the orifice diameter, as reported in the previous section, it is reasonable to suppose
that open cell conditions may affect significantly the evaporation behaviour.
KEMS experiments: mass spectra
Unlikely the KEML technique, Knudsen Effusion Mass Spectrometry permits one to obtain direct
information on the composition of the vapor phase and to ascertain the evaporation/decomposition
processes taking place as a function of temperature. In order to get more information on the effect
of the effusion orifice size, KEMS experiments were performed with orifice of 1 mm (platinum and
alumina cells) and 0.5 mm (alumina). Typical mass spectra of the BMImPF6 vapor recorded with an
electron energy of 20 eV are shown in Figure 3a (effusion orifice diameter 1 mm) and 3b (effusion
orifice diameter 0.5 mm), where the ion currents were background-subtracted and corrected for the
different multiplier gains (see eq 3) by the factor M . For sake of comparison, the spectrum of
BMImNTf2 recorded at a similar temperature is also shown (Figure 3c). The spectra of the two ILs
look very different. In particular, while the pattern of BMImNTf2 displays an overwhelming peak at
m/e = 139, in the case of BMImPF6, the spectra are much more complex, with a number of peaks of
comparable intensity. Furthermore, the relative intensity of most peaks depends markedly on the
orifice size. With the smaller orifice (Figure 3b) the most intense peak is at m/e = 96, with the 1 mm
orifice (Figure 3a) it is at m/e = 139.
As discussed elsewhere,9 mass 139 corresponds to the BMIm+ cation. In the case of BMImNTf2 and
other ILs,31 this species is assigned to the fragmentation of the unstable radical ion BMIm+NTf2·
formed under electron impact from the neutral ion pair BMImNTf2(g) produced by evaporation,
according to the following scheme:
BMImNTf2(l) = BMIm+NTf2¯(g)
ACS Paragon Plus Environment
Page 12 of 42
Page 13 of 42
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
The Journal of Physical Chemistry
BMIm+NTf2¯(g) + 1e¯BMIm+NTf2·(g) + 2e¯
(8)
BMIm+NTf2·(g) BMIm+ + NTf2·(g)
where the first equation is the evaporation process 4. Minor peaks in the spectrum of BMImNTf2
are assigned to fragments of BMImNTf2(g), as discussed elsewhere.9,29
With regard to the BMImPF6 mass spectra, a reasonable assignment of the most intense peaks is as
follows: 82 = methylimidazole; 96 = ethylimidazole; 139 = BMIm+; 158 = BMImF+. Likewise in
the BMImNTf2 spectrum, mass 139 is assigned to the fragmentation of the intact ion pair BMImPF6
by the very same mechanism reported in eq 8. The above assignments are supported by the analysis
of the IECs (Ion Efficiency Curves, i.e. intensity vs. electron energy curves) recorded for the most
important ion species detected in the experiments. The appearance energy (AE) of each ion can be
deduced from these curves by appreciating the onset of intensity from the baseline at low energy.
To this end, the energy scale was calibrated by the AE of mass 18 (Ionization Energy (H2O)=12.6
eV58). The AEs of the most important species are reported in Table 2, together with the value for
mass 139 for BMImNTf2. The AE of mass 139 in BMImNTf2 was preliminarily reported to be 9.7 ±
0.5 eV in our previous paper9 and was re-measured in the current work, obtaining a value of 9.3 ±
0.3 eV (see Table 2), in acceptable agreement with the previous determination. It can be noted from
Table 2 that AE(139) is remarkably higher for BMImPF6 than for BMImNTf2. Since the anion PF6–
has a much lower polarizability compared to NTf2– (4.39 and 13.11 Å3, respectively, as computed
by ab initio calculations59), this result seems consistent with the well-documented inverse
correlation between ionization energy and polarizability.60 It is important to note that the AE of the
species with m/e = 96 is in excellent agreement with the ionization energy reported for the
ethylimidazole (8.58 eV61). This evidence suggests that this signal is due to the presence of neutral
ethylimidazole in the gaseous phase, probably produced by thermal decomposition of the IL,
although the formation through gas phase dissociation processes cannot be ruled out on the basis of
KEMS results. As a consequence, the mass loss rates measured in KEML experiments cannot be
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
considered as representative only of the evaporation process, although with increasing the effusion
orifice size the contribution of peaks other than BMIm+ should tend to become minor. The
formation of ethylimidazole among the thermal decomposition products of BMImPF6 at much
higher temperatures (723-823 K) was already reported on the basis of TGA-MS and pyrolysis-gas
chromatography measurements.62,63 The formation of this species was reported for a number of ILs
containing the EMIm+ cation and cyano-functionalised or halide anions, where it has been
rationalised on the basis of a SN2 mechanism.17,21,64
On the contrary, the AE measured for mass 82 is much higher than the ionization energy reported in
the literature for methylimidazole (8.66 eV65), so this ion is most probably a fragment. With regard
to mass 158 (BMImF), on the basis of the AE it is not possible to assign it to a neutral BMImF(g)
precursor or to a fragment ion (but see below).
As apparent in Figures 3a and 3b, the orifice size was found to affect the relative intensities of the
various ion species. This effect is clearly shown in Figure 4a, where the intensity ratio measured
with different cells is reported for peaks at m/e = 96 and 139. While with the 1 mm orifice the peak
139 (i.e., the gaseous ion pair) is about 3 times more intense than peak 96 (ethylimidazole), with the
0.5 mm orifice the ratio is reversed, with peak 96 being the most intense. This evidence strongly
supports the above hypothesis that the ion species with m/e = 96 originates from the primary
ionization of ethylimidazole neutral precursor rather than from the fragmentation of the
BMImPF6(g) ion pair. On the contrary, this effect is not observed for mass 158 (Figure 4b),
suggesting this ion is formed by fragmentation of the BMImPF6(g) ion pair. Note that in both cases
the intensity ratios show no significant dependence on temperature in the range investigated,
making this criterium ineffective in distinguishing parent and fragment ions.
ACS Paragon Plus Environment
Page 14 of 42
Page 15 of 42
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
The Journal of Physical Chemistry
KEMS experiments: evaluation of partial pressures
As shown in the previous section (see Figure 3a), the mass spectrum measured with the 1 mm
orifice displays the signal at m/e = 139, assigned to the BMIm+ ion, as the most intense. Since this
ion is originated by electron impact from the BMImPF6(g) species, in turn formed by the simple
evaporation 4, the larger orifice seems suitable to study the evaporation process, with the
simultaneous decomposition/dissociation processes playing a minor role. To this end, the intensity
of the main ion species was monitored as a function of temperature in the range 455-528 K by using
a 1 mm diameter platinum Knudsen cell.
In order to derive the evaporation enthalpy of BMImPF6, the partial pressure of BMImPF6(g) was
evaluated by eq 3 from the intensity of peak 139. The main issue in using eq 3 is that a cross section
value has to be estimated. To the best of our knowledge, the only approximate estimate available in
the literature for the electron impact ionization cross section of an IL was provided by us9 for the
BMIm+ ion formed from BMImNTf2(g) (σ = 35 ± 15 Å2). This value is larger than the typical
values of molecular organic species by a factor 2-3, what could be reasonable for an ion pair. In
view of the widely different electronic polarizability of the NTF2– and PF6– anions (see above),59 it
is reasonable to expect a smaller cross section for the latter.66 In the lack of more information, we
estimated the cross section of BMImPF6(g) by assuming for the PF6– and NTF2– anions a direct
proportionality between σ and electronic polarizability, well-documented for molecular species.66
The value derived for PF6– by using the ab initio polarizabilities reported above59 is equal to 11.7 ±
1.8 Å2, to which a uncertainty of ± 15% has be associated, conservatively estimated from the known
uncertainty of the polarizability vs. cross section correlation.66 The so derived partial pressures of
BMImPF6(g) are reported in Table 3 as a function of temperature and can be represented by the
equation ln p/bar = (27.939 ± 1.429) – (15732 ± 697)/T. It is worth noting that according to these
data the vapor pressure of BMImPF6 is lower than that of BMImNTf2 by a factor ranging from 10 to
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 16 of 42
20 in the explored temperature range. In the next section, KEMS pressures will be used to derive
the evaporation enthalpy of BMImPF6.
A similar analysis can be used to estimate the partial pressure of ethylimidazole, the second most
abundant vapor species according to KEMS spectra (see Figure 3). The pertinent cross section may
be estimated by the above-mentioned correlation with the electronic polarizability.66 The latter was
in turn estimated from that reported for imidazole (7.495 Å3)67, by adding the group contribution of
(CH2)2 (3.6 Å3)68. The resulting value of σ is 16.4 ± 2.5 Å2 and the partial pressures derived
thereafter are also reported in Table 3. By assuming for simplicity that no other species is present,
the mole fraction y(BMImPF6) in the vapor phase results to be 0.81 ± 0.04 (average value from
Table 3) with no evident temperature dependence. We note that the partial pressures of 1ethylimidazole estimated by KEMS are much lower than the vapor pressure of the pure compound.
By analogy with 2-methylimidazole,69 at 450 K the latter is expected to be in the order of 1 bar, so
that the activity of 1-ethylimidazole in the liquid is estimated to be 10-8 or less, which should imply
a very low concentration. As a consequence, we can be confident that the study of the simple
evaporation of BMImPF6 was performed on a liquid phase with negligible impurity.
On the basis of this roughly estimated vapor phase composition, a comparison can be attempted
between KEMS and KEML results for the 1 mm effusion orifice. If more than one species is present
in the vapor, eq 1 gives the total mass loss rate measured in KEML experiments provided that the
following expression for the mean molecular weight is used at each temperature:41


M =  ∑ yi M i 
 i

2
(9)
where yi are the mole fractions of the various vapor species in the cell and Mi the respective
molecular weight. By assuming as a first approximation that the vapor is composed only by
BMImPF6(g) and ethylimidazole, with y(BMImPF6) = 0.81 (see above), we obtain from eq 9 M =
240.8 g mol-1 (only slightly lower than the molar mass of BMImPF6, 284.2 g mol-1, as expected). In
ACS Paragon Plus Environment
Page 17 of 42
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
The Journal of Physical Chemistry
Figure 5 we report the KEML partial pressures of BMImPF6(g) calculated according to eq 1 with
the so-derived average molecular weight and the corresponding KEMS values (see Table 3). For
sake of comparison, in the same figure we reported the KEML pressures measured with the largest
orifice (3 mm), where the contribution of species other than BMImPF6(g) is argued to be very
small, and the results recently obtained by Zaitsau et al with QCM measurements11 (the only
BMImPF6 vapor pressure values currently available in the literature). Although the KEMS
pressures in Figure 5 are somewhat scattered and can be affected by uncertainties in parameters of
eq 3 such as instrumental constant and ionization cross sections, a very satisfactory agreement with
KEML pressure data is found. Admittedly, our lowest values (KEML with 3 mm orifice) are still
about 3-4 times higher than QCM results. The discrepancy could be explained by the use of open
pans in the latter technique: the decreasing trend of apparent pressures with increasing the effusion
hole size, described in the previous sections, could find its limiting value under open cell
conditions, where simple evaporation would be the only or largely dominant process. In this view,
free surface conditions seem to be the most suitable conditions under which simple evaporation may
be studied without decomposition giving a significant contribution. Nevertheless, unlikely effusion
conditions, free surface evaporation could not guarantee the achievement of a thermodynamic
equilibrium and a properly saturated vapor phase.
Thermodynamic analysis: evaporation enthalpy of BMImPF6
In Table 3 a reasonable evaluation of the partial pressures of the BMImPF6(g) species was done
from KEMS measurements using an orifice of 1 mm in diameter. In Figure 5, we showed that, in
spite of the various approximations involved in the processing of KEML data, fairly consistent
values are obtained by KEML under similar conditions. Since these pressures are derived from the
analysis of MS signals of the BMIm+ ion formed from the BMImPF6 gaseous species, they are
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 18 of 42
believed to provide a fairly good approximation to the true thermodynamic equilibrium vapor
pressures of BMImPF6(l), regardless possible simultaneous decomposition processes.
Vapor pressure data can be analysed by the so-called second- and third-law methods to derive the
corresponding evaporation enthalpy. Advantages and drawbacks of the two methods were described
elsewhere.9 Second-law analysis is based on evaluating the slope of the ln p vs. 1/T regression line
according to the Clausius-Clapeyron equation. While this procedure does not need auxiliary thermal
functions, it is known to be severely affected by the extent of dataset5 and temperature range, which
are both rather limited for the KEMS data reported in Table 3 and Figure 5. The slope of the
regression line of these data gives the enthalpy change at the average experimental temperature (488
o
K), ∆evapH 488K = 130.8 ± 5.8 kJ mol-1, where the relatively large error is due to the low number and
high scatter of data. In order to adjust this value to the reference temperature of 298 K, the heat
capacities of BMImPF6 in the liquid and in the gaseous phase are needed. Molar heat capacities of
crystal and liquid BMImPF6 were measured by adiabatic calorimetry from 5 to 550 K.33.
Measurements above room temperature were also carried out by several authors34-40 and, with the
exception of two studies,35-36 deviation among different datasets is lower than 2%, in many cases
within 1%. As far as the values at low temperature are concerned, the results by Triolo and coworkers39 agree satisfactorily with those of ref. 33. With regard to gaseous BMImPF6(g), Paulechka
and co-workers25 determined the ideal gas thermodynamic functions of BMImPF6(g) neutral ion
pair by means of ab initio calculations at the MP2 level of theory and standard statistical
o
thermodynamics. Using data from papers of refs. 33 and 25, the value ∆evapH 298K = 146.4 ± 5.8 kJ
mol-1 is obtained. Despite the above-mentioned drawbacks regarding this method applied to our
KEMS dataset, the value so obtained is in excellent agreement with the only experimental value
available in the literature, published last year by Zaitsau et al11 ( ∆evapH 298K = 146.5 ± 2.6 kJ mol-1,
o
Table 5). Such an agreement between two so different methods gives further confidence in the
approach followed in the present work to interpret KEMS results. It is worth noting that the heat
ACS Paragon Plus Environment
Page 19 of 42
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
The Journal of Physical Chemistry
(
)
o
o
content used to adjust our result to 298 K, ∆evap H 488K − H 298K = 15.6 kJ mol-1, is only marginally
different from the value one would obtain by the semiempirical estimation used by Zaitsau et al.11,
(
)
o
o
o
-1
∆evap H488
K − H 298K = 14.0 kJ mol , and so a direct comparison between the ∆ r H 298K values is
feasible.
The third-law analysis is based on the following equation:
o
o
∆r H298
K = −RT ln K p (T ) − T∆r Gef (T , Tref = 298 K )
(10)
where ∆ r Gef (T ) is the Gibbs energy function change (Gef, with reference temperature Tref equal to 298 K)
o
of reaction 4, with
Gef
°
( T , T ref = 298 K) =
G o ( T ) − H o ( 298 K )
T
=
H o ( T ) − H o ( 298 K )
T
− S o (T )
(11)
for liquid and gaseous BMImPF6. As far as the liquid phase is concerned, Gefs can be easily
calculated if the molar heat capacities in the entire temperature range from 0 K to the experimental
temperatures are known. For the gas phase, the calculation is done using statistical thermodynamics
with partition functions evaluated from spectroscopic/computational molecular parameters. The
third-law method is a good alternative to Clausius-Clapeyron fits when Gefs are known with
reasonable accuracy and it was successfully applied in our previous study on BMImNTf2.9
Among ILs, BMImPF6 represents one of the few cases in which such an analysis can provide results
of good accuracy, because a complete derivation of thermodynamic functions was presented in the
above cited papers.25,33 The third-law evaporation enthalpies calculated from the KEMS data are
shown in Table 4, with the mean value and the corresponding standard deviation. On the basis of
the above-mentioned comparison between different sets of heat capacities data available in the
literature, we estimate a further uncertainty of about ± 1.5 kJ·mol–1 due to ∆Gef inaccuracy. Finally,
an uncertainty of ± 0.5 kJ·mol–1 was considered for the aforementioned uncertainty on the cross
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 20 of 42
o
section. The resulting third-law value for the evaporation enthalpy of BMImPF6 is then ∆evapH 298K
= 145.3 ±2.9 kJ mol-1. This result is in excellent agreement with the second-law value. As a
confirmation of the good overall agreement, we note that the standard evaporation entropy change
obtained as the y-intercept of the second-law fitting line ( ∆r S488K =137 ± 12 J K-1 mol-1) is consistent
o
with the value (133 J K-1 mol-1) obtained from the absolute entropies of liquid and gaseous
BMImPF6, obtained from calorimetry33 and theoretical calculations,25 respectively.
o
In selecting a final recommended value of ∆evapH 298K , we prefer to choose that derived by the third-
law method, which is less affected by the low number of data points and by possible inaccuracies
due to temperature-dependent errors. This value (145.3 ±2.9 kJ mol-1) is in full agreement with the
QCM result (146.5 ± 2.6 kJ mol-1).11 However, it should be considered that the latter is based on a
second-law analysis, the former on a third-law treatment. Since the QCM vapor pressures11 are
lower than ours (see Figure 5), if a third-law analysis were done on QCM data, a value of 151.4
±0.2 kJ mol-1 would be obtained, not negligibly larger than ours, but not very far.
A large number of molecular dynamics (MD) studies were published on BMImPF6,11,13,14,70-75
where evaporation enthalpy was estimated with different approaches and force fields. Theoretical
results are reported in Table 5 together with the two experimental values now available. It is evident
that most simulations significantly overestimate this property, whereas the values proposed in ref.
11 and 13 are the most effective in reproducing the experimental results.
In the same table, we report also the result of two estimates based, respectively, on the Hildebrand
solubility parameter δH, i.e. the square root of the cohesive energy density76
1/ 2
∆ U 
δ H =  ev 
 V 


1/ 2
 ∆evapH − RT 

= 
V


and on the empirical correlation between evaporation enthalpy and surface tension γ:6,77
ACS Paragon Plus Environment
(12)
Page 21 of 42
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
The Journal of Physical Chemistry
γV 2 / 3 = A + B ∆ evapH
(13)
where V is the molar volume. These methods provide a very simple way to get a rough estimate of
the evaporation enthalpy. For V, the well-assessed value78 of 207.8 dm3·mol–1 was used. With
regard to eq 12, we note that ∆evapH is very sensitive to small changes in the Hildebrand parameter:
the values (measured by indirect methods) 30.279 29.880 and 28.0976 MPa0.5 at T = 298 result in
∆evapH298K of 192, 187 and 167 kJ·mol–1, respectively. As for eq 13, by using for A and B the values
derived from a set of experimental values for ILs,77 and the value γ = 43.9 mN·m–1 given in ref. 81
(average of values measured with two independent methods at 298 K), ∆evapH298K = 149.9 kJ·mol–1 is
obtained, in good agreement with the experimental value. However, other literature sources give
surface tension values exceeding 43.9 mN·m–1 by up to 4 mN·m–1,81 resulting in ∆evapH298K values
higher by up to 15 kJ·mol–1.
Conclusions
In this study we presented the results of evaporation experiments performed on the imidazolium
ionic liquid BMImPF6, aimed at studying the competition between simple evaporation and thermal
decomposition. Our multi-technique approach used two techniques based on molecular effusion:
Knudsen Effusion Mass Loss (in the temperature range 453 -551 K) and Knudsen Effusion Mass
Spectrometry (425-528 K), complemented with non-isothermal TG-DTA experiments. While
KEML measurements provided information on total mass loss as a function of temperature, the
mass spectrometric analysis of the vapor phase allowed the evaporation process to be studied even
in the presence of the simultaneous release of volatile decomposition products. On the basis of
KEMS mass spectra and KEML mass loss rates measured with effusion orifices of different
diameter, we obtained evidence for the simultaneous occurrence of simple evaporation
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
(BMImPF6(l) = BMImPF6(g)) and thermal decomposition processes. Specific mass loss rates
depended on the effusion orifice size, a result that may be explained by the occurrence of a
kinetically delayed process. Mass spectra displayed peaks assigned to the fragmentation of the
gaseous ion pair BMImPF6(g) (in particular, the peak at m/e = 139, corresponding to the BMIm+
cation), and peaks most probably due to volatile products formed by thermal
decomposition/dissociation, the most intense being at m/e = 96 (assigned to ethylimidazole). These
findings are consistent with the results obtained from TG/DTA experiments carried out using two
different carrier gas (He and N2). The KEMS technique enabled us to monitor the evaporation
process and to measure the partial pressures of the gaseous ion pair BMImPF6(g) as a function of
temperature. Pressure data were processed according to the third-law method of analysis, obtaining
o
a ∆evapH298K value of 145.3 ±2.9 kJ mol-1, in excellent agreement with the only previous experimental
value11 recently become available in the literature. The evaporation enthalpy of BMImPF6 at 298 K
is more than 20 kJ mol-1 higher compared to BMImNTf2, whose evaporation behavior is wellknown from previous studies,7-9 and vapor pressures are about 20-30 times lower in the explored
temperature range.
ACS Paragon Plus Environment
Page 22 of 42
Page 23 of 42
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
The Journal of Physical Chemistry
Acknowledgment
The authors gratefully acknowledge funding from the University of Rome La Sapienza (Progetto di
Ricerca di Università 2015 “Studio termodinamico dei processi di vaporizzazione di liquidi ionici
aprotici”).
Supporting Information Available
Ionization efficiency curves of ions with m/e = 96 and 139. This information is available free of
charge via the Internet at http://pubs.acs.org
References
(1) Freemantle, M. An Introduction to Ionic Liquids; RSC Publishing: Cambridge (UK), 2010.
(2) Amde, M.; Liu, J.-F. ; Panga, L. Environmental Application, Fate, Effects, and Concerns of
Ionic Liquids: A Review. Env. Sci. Technol. 2015, 49, 12611-12627.
(3) MacFarlane, D.R.; Tachikawa, N; Forsyth, M.; Pringle, J.M.; Howlett, P.C.; Elliott, G.D.;
Davis Jr., J.H.; Watanabe, M.; Simon, P.; Angeli et al. Energy Applications of Ionic Liquids.
Energy Environ. Sci., 2014, 7, 232-250.
(4) Shevelyova, M.P.; Paulechka, Y.U.; Kabo, G.J.; Blokhin, A.V.; Kabo, A.G.
Physicochemical Properties of Imidazolium-Based Ionic Nanofluids: Density, Heat
Capacity, and Enthalpy of Formation. J. Phys. Chem. C 2013, 117, 4782-4790.
(5) Esperança, J. M. S. S.; Canongia Lopes, J. N.; Tariq, M.; Santos, L. M. N. B. F.; Magee J.
W.; Rebelo, L. P. N. Volatility of Aprotic Ionic Liquids - A Review. J. Chem. Eng. Data
2010, 55, 3-12.
(6) Zaitsau, D. H.; Kabo, G. J.; Strechan, A. A.; Paulechka, Y. U.; Tschersich, A.; Verevkin S.
P.; Heintz, A. Experimental Vapor Pressures of 1-Alkyl-3-methylimidazolium
Bis(trifluoromethylsulfonyl)imides and a Correlation Scheme for Estimation of
Vaporization Enthalpies of Ionic Liquids. J. Phys. Chem. A 2006, 110, 7303-7306.
(7) Rocha, M. A. A.; Lima, C. F. R. A. C.; Gomes, L. R.; Schröder, B.; Coutinho, J. A. P.;
Marrucho, I. M.; Esperança, J. M. S. S.; Rebelo, L. P. N.; Shimizu, K.; Canongia Lopes J. N.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
et al. High-Accuracy Vapor Pressure Data of the Extended [CnC1Im][Ntf2] Ionic Liquid
Series: Trend Changes and Structural Shifts. J. Phys. Chem. B 2011, 115, 10919-10926.
(8) Verevkin, S. P.; Zaitsau, D. H.; Emel’yanenko, V. N.; Yermalayeu, A. V.; Schick, C.; Liu,
H.; Maginn, E. J.; Bulut, S.; Krossing I.; Kalb, R. Making Sense of Enthalpy of
Vaporization Trends for Ionic Liquids: New Experimental and Simulation Data Show A
Simple Linear Relationship and Help Reconcile Previous Data. J. Phys. Chem. B 2013, 117,
6473-6486.
(9) Brunetti, B; Ciccioli, A.; Gigli, G.; Lapi, A.; Misceo, N.; Tanzi, L.; Vecchio Ciprioti, S.
Vaporization of the Prototypical Ionic Liquid BMImNtf2 under Equilibrium Conditions: A
Multitechnique Study. Phys. Chem. Chem. Phys. 2014, 16, 15653-15661.
(10) Chilingarov, N. S.; Medvedev, A. A.; Deyko, G. S.; Kustov, L. M.; Chernikova, E. A.;
Glukhov, L. M.; Markov, V. Yu.; Ioffe, I. N.; Senyavin, V. M. et al. Mass Spectrometric
Studies of 1-Ethyl-3-Methylimidazolium and 1-Propyl-2,3-Dimethylimidazolium
Bis(Trifluoromethyl)-Sulfonylimides. Rapid Comm. Mass Spectr. 2015, 29, 1227-1232.
(11) Zaitsau, D.H.; Yermalayeu, A. V; Emel’yanenko, V.N.; Butler, S.; Schubert, T.; Verevkin,
S.P. Thermodynamics of Imidazolium-Based Ionic Liquids Containing PF6 Anions. J. Phys
Chem B 2016, 120, 7949-7957.
(12) Dommert, F.; Holm, C. Refining Classical Force Fields for Ionic Liquids: Theory and
Application to [MMIM][Cl]. Phys. Chem. Chem. Phys. 2013, 15, 2037-2049.
(13) Borodin, O. Polarizable Force Field Development and Molecular Dynamics Simulations of
Ionic Liquids. J. Phys. Chem. B 2009, 113, 11463-11478.
(14) Zhong, X.; Liu, Z; Cao, D. Improved Classical United-Atom Force Field for ImidazoliumBased Ionic Liquids: Tetrafluoroborate, Hexafluorophosphate, Methyilsulfate,
Trifluoromethylsulfonate, Acetate, Trifluoroacetate, and
Bis(Trifluoromethylsulfonyl)Amide. J Phys. Chem. B 2011, 115, 10027-10040.
(15) Canongia Lopes, J.N.; Deschamps, J.; Pádua, A.A.H. Modeling Ionic Liquids Using a
Systematic All-Atom Force Field. J. Phys. Chem. B 2004, 108, 2038-2047.
(16) Köddermann, T.; Paschek, D.; Ludwig, R. Molecular Dynamic Simulations of Ionic
Liquids: A Reliable Description of Structure, Thermodynamics and Dynamics.
ChemPhysChem 2007, 8, 2464-2470.
(17) Liu, J.; Chambreau, S.D.; Vaghjiani, G.L.; Dynamics Simulation and Statistical Modeling
of Thermal Decomposition of 1-Ethyl-3-Methylimidazolium Dicyanamide and 1-Ethyl-2,3Dimethylimidazolium Dicynamide. J. Phys. Chem. A 2014, 118, 11133-11144.
ACS Paragon Plus Environment
Page 24 of 42
Page 25 of 42
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
The Journal of Physical Chemistry
(18) Chaban, V.V.; Prezhdo, O.V. Ionic Vapor Composition in Pyridinium-Based Ionic
Liquids, J. Phys. Chem. B 2016, 120, 4661-4667.
(19) Lovelock, K. R. J.; Armstrong, J. P.; Licence, P.; Jones, R. G. Vaporisation and Thermal
Decomposition of Dialkylimidazolium Halide Ion Ionic Liquids. Phys. Chem. Chem. Phys.
2014, 16, 1339-1353.
(20) Maton, C.; De Vos N.; Stevens, C. V. Ionic Liquid Thermal Stabilities: Decomposition
Mechanisms and Analysis Tools. Chem. Soc. Rev. 2013, 42, 5963-5977.
(21) Chambreau, S. D.; Boatz, J. A.; Vaghjiani, G. L.; Koh, C.; Kostko, O.; Golan, A.; Leone,
S. R. Thermal Decomposition Mechanism of 1-Ethyl-3-Methylimidazolium Bromide Ionic
Liquid. J. Phys. Chem. B 2012, 116, 5867-5876.
(22) Chen, Y.; Cao, Y.; Shi, Y.; Xue Z.; Mu, T. Quantitative Research on The Vaporization and
Decomposition of [EMIM][Tf2N] by Thermogravimetric Analysis–Mass Spectrometry. Ind.
Eng. Chem. Res. 2012, 51, 7418-7427.
(23) Cao, Y.; Mu, T. Comprehensive Investigation on the Thermal Stability of 66 Ionic Liquids
by Thermogravimetric Analysis. Ind. Eng. Chem. Res. 2014, 53, 8651-8664.
(24) Liu, P; Wang, M.; Chen, Z.-M., Thermal Stability and Vapor-Liquid Equilibrium for
Imidazolium Ionic Liquids as Alternative Reaction Media, J Chem. Eng Data 2015, 60, 836844.
(25) Paulechka, Y.U.; Kabo, G.J.; Blokhin, A.V.; Vydrov, O.A.; Magee, J.W.; Frenkel, M.
Thermodynamic Properties of 1-Butyl-3-Methylimidazolium Hexafluorophosphate in the
Ideal Gas State. J. Chem. Eng Data 2003, 48, 457–462.
(26) Emel’yanenko, V.N.; Verevkin, S.P.; Heintz, A.; Schick, C. Ionic Liquids. Combination of
Combustion Calorimetry with High-Level Quantum Chemical Calculations for Deriving
Vaporization Enthapies. J. Phys. Chem. B 2008, 112, 8095-8098.
(27) Zaitsau, D.H.; Paulechka, Y.U.; Kabo, G.J. The Kinetics of Thermal Decomposition of 1Butyl-3-Methylimidazolium Hexafluorophosphate. J. Phys. Chem. A 2006, 110, 1160211604.
(28) Chilingarov, N.S.; Medvedev, A.A.; Deyko, G.S.; Sidorov, L.N. The Evaporation Study of
Silicon-Containing Ionic Liquid. Chem. Phys. Lett. 2016, 675, 8-10.
(29) Deyko, A.; Lovelock, K. R. J.; Licence P.; Jones, R. G. The Vapour of Imidazolium-Based
Ionic Liquids: A Mass Spectrometry Study. Phys. Chem. Chem. Phys. 2011, 13, 1684116850.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
(30) Dunaev, A.M.; Motalov, V.B.; Kudin, L.S.; Butman, M.F. Thermodynamic Properties of
the Ionic Vapor Species Over EMImNTf2 Ionic Liquid Studied by Knudsen Effusion Mass
Spectrometry. J. Mol. Liq. 2016, 223, 407-411.
(31) Dunaev, A.M.; Motalov, V.B.; Kudin, L.S.; Butman, M.F. Molecular and Ionic
Composition of Saturated Vapor Over EMImNTf2 Ionic Liquid. J. Mol. Liq. 2016, 219, 599601.
(32) Heym, F.; Korth, W.; Etzold, B. J. M.; Kern, C.; Jess, A. Determination of Vapor Pressure
and Thermal Decomposition Using Thermogravimetrical Analysis, Thermochim. Acta 2015,
622, 9-17.
(33) Kabo, G. J.; Blokhin, A. V.; Paulechka, Y. U.; Kabo, A. G.; Shymanovich, M. P.
Thermodynamic Properties of 1-Butyl-3-Methylimidazolium Hexafluorophosphate in the
Condensed State. J. Chem. Eng. Data 2004, 49, 453–461.
(34) Nieto De Castro, C.A.; Lourenço, M.J.V.; Ribeiro, A.P.C.; Langa, E.; Vieira, S.I.C.
Thermal Properties of Ionic Liquids and Ionanofuids of Imidazolium and Pyrrolidinium
Liquids. J. Chem. Eng Data 2010, 55, 653-661.
(35) Holbrey, J.D.; Reichert, W.M.; Reddy, R.G; Rogers, R.D. Heat Capacities of Ionic Liquids
and Their Applications as Thermal Fluids. ACS Symposium Series, Ionic Liquids As Green
Solvents, 2003, Chapter 11, 121-133.
(36) Fredlake, C.P.; Crosthwaite, J.M.; Hert, D.G.; Aki, S.N.V.K.; Brennecke, J.F.
Thermophysical Properties of Imidazolium-Based Ionic Liquids. J. Chem. Eng Data 2004,
49, 954-964.
(37) Troncoso, J.; Cerdeiriña, C.A.; Sanmamed, Y.A.; Romani, L.; Rebelo, L.P.N.
Thermodynamic Properties of Imidazolium-Based Ionic Liquids: Densities, Heat Capacities,
and Enthalpies of Fusion of [BMIm][PF6] and [BMIm][NTf2]. J. Chem. Eng Data 2006, 51,
1856-1859.
(38) Yu, Y.-H-; Soriano, A.N.; Li, M.-H. Heat Capacities and Electrical Conductivities of 1-NButyl-3-Methylimidazolium-Based Ionic Liquids. Thermochim. Acta 2009, 482, 42-48.
(39) Triolo, A.; Mandanici, A.; Russina, O.; Rodriguez-Mora, V.; Cutroni, M.: Hardacre, C.;
Nieuwenhuyzen, M.; Bleif, H.-J.; Keller, L.; Ramos, M.A. Thermodynamics, Structure, and
Dynamics in Room Temperature Ionic Liquids: The Case of 1-Butyl-3-Methyl Imidazolium
Hexafluorophosphate ([BMIm][PF6]. J. Phys. Chem. B 2006, 110, 21357-21364.
(40) Paulechka, E.; Liavitskaya, T.; Blokhin, A.V. Calorimetric Study of Polymorphism in 1Butyl-3-Methylimidazolium Hexafluorophosphate. J. Chem. Thermod. 2016, 102, 211-218.
ACS Paragon Plus Environment
Page 26 of 42
Page 27 of 42
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
The Journal of Physical Chemistry
(41) Freeman, R.D., Momentum Sensors, In Margrave, J.L. (Ed.) The Characterization of
High-Temperature Vapors, 1967, John Wiley & Sons, New York, pp. 157-158
(42) Motzfeldt, K. The Thermal Decomposition.of Sodium Carbonate by The Effusion Method.
J Phys. Chem. 1955, 59, 139-147.
(43) Monte, M. J. S.; Santos, L. M. N. B. F.; Fulem, M.; Fonseca J. M. S.; Sousa, C. A. D. New
Static Apparatus and Vapor Pressure of Reference Materials: Naphthalene, Benzoic Acid,
Benzophenone, and Ferrocene. J. Chem. Eng. Data 2006, 51, 757-766.
(44) Ribeiro Da Silva, M. A. V.; Monte M. J. S.; Santos, L. M. N. B. F. The Design,
Construction, and Testing of a New Knudsen Effusion Apparatus. J. Chem. Thermodyn.
2006, 38, 778–787.
(45) Drowart, J.; Chatillon, C.; Hastie J.; Bonnell, D. High-Temperature Mass Spectrometry:
Instrumental Techniques, Ionization Cross-Sections, Pressure Measurements, and
Thermodynamic Data. Pure Appl. Chem. 2005, 77, 683-737.
(46) Ciccioli, A.; Gigli, G. The Uncertain Bond Energy of the NaAu Molecule: Experimental
Redetermination and Coupled Cluster Calculations. J. Phys. Chem. A 2013, 117, 4956-4962.
(47) Heym, F.; Etzold, B. J. M.; Kern, C.; Jess, A. Analysis of Evaporation and Thermal
Decomposition of Ionic Liquids by Thermogravimetrical Analysis at Ambient Pressure and
High Vacuum. Green Chem. 2011, 13, 1453-1466.
(48) Heym, F.; Etzold, B. J. M.; Kern, C.; Jess, A. An Improved Method To Measure The Rate
of Vaporisation and Thermal Decomposition of High Boiling Organic and Ionic Liquids by
Thermogravimetrical Analysis, Phys, Chem. Chem. Phys. 2010, 12, 12089-12100.
(49) Dong, K.; Zhao, L.; Wang, Q.; Song, Y.; Zhang, S. Are Ionic Liquids Pairwise in Gas
Phase? A Cluster Approach and in Situ IR Study. Phys. Chem. Chem. Phys. 2013, 15, 60346040.
(50) Neto, B.A.D.; Meurer, E.C.; Galaverna, R.; Bythell, B.J.; Dupont, J.; Graham Cooks R.;
Eberlin, M.N. Vapors from Ionic Liquids: Reconciling Simulations with Mass Spectrometric
Data. J. Phys. Chem. Lett. 2012, 3, 3435-3441.
(51) Leal, J.P.; Esperança, J.M.S.S.; Minas Da Piedade, M.E.; Canongia Lopes, J.N.; Rebelo
L.P.N.; Seddon, K.R. The Nature of Ionic Liquids in the Gas Phase. J. Phys. Chem. A 2007,
111, 6176-6182.
(52) Gross, J.H. Molecular Ions of Ionic Liquids in the Gas Phase . J. Am. Soc. Mass Spectrom.
2008, 19, 1347-1352.
(53) Chambreau, S.D; Vaghjiani, G.L.; Koh, C.J.; Golan, A.; Leone, S.R. Ultraviolet
Photoionization Efficiency of the Vaporized Ionic Liquid 1-Butyl-3-Methylimidazolium
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Tricyanomethanide: Direct Detection of the Intact Ion Pair. J. Phys. Chem. Lett. 2012, 3,
2910-2914.
(54) Ballone, P.; Pinilla, C.; Kohanoff, J.; Del Pópolo, M.G. Neutral and Charged 1-Butyl-3Methylimidazolium Triflate Clusters: Equilibrium Ionic Vapor: What Does It Consist of? J.
Phys. Chem. B 2007, 111, 4938-4950.
(55) Chaban, V.V.; Prezhdo, O.V. Concentration in the Vapor Phase and Thermal Properties of
Nanometric Droplets. J. Phys. Chem. Lett. 2012, 3, 1657-1662
(56) Lau, K.H.; Cubicciotti, D.; Hildenbrand, D.L. Effusion Studies of the Thermal
Decomposition of Magnesium and Calcium Sulfates. J. Chem. Phys. 1977, 66, 4532-4539.
(57) Ganesan, R.; Ciccioli, A; Gigli, G; Ipser, H. Thermochemical Investigations in the TinPhosphorous System. Int. J. Mat. Res. 2011, 102, 93-103.
(58) P.J. Linstrom and W.G. Mallard, Eds., NIST Chemistry Webbook, NIST Standard
Reference Database Number 69, National Institute of Standards and Technology,
Gaithersburg MD, 20899, Http://Webbook.Nist.Gov/Chemistry/ (Retrieved October 11,
2017).
(59) Izgorodina, E.I.; Forsyth, M.; Macfarlane, D.R. On the Components of the Dielectric
Constants of Ionic Liquids: Ionic Polarization ? Phys. Chem. Chem. Phys. 2009, 11, 24522458.
(60) Blair, S.A.; Thakkar, A.J. Relating Polarizability to Volume, Ionization Energy,
Electronegativity, Hardness, Moments of Momentum, and Other Molecular Properties. J.
Chem. Phys. 2014, 141, 074306/1-5
(61) Turchaninov, V.K.; Ermikov, A.F.; Shagun, V.A.; Baikalova, L.V. Inversion of the
Highest Occupied Molecular Orbitals of Imidazole Derivatives. Bull. Acad. Sci. USSR,
Divis. Chem. Sci. 1987, 36, 2407-2410.
(62) Othani, H.; Ishimua, S; Kumai, M. Thermal Decomposition Behaviors of ImidazoliumType Ionic Liquids by Pyrolysis-Gas Chromatography. Anal. Sciences 2008, 24, 1335-1340.
(63) Hao, Y.; Peng, J.; Hu, S.; Li, J.; Zhai, M. Thermal Decomposition of Allyl-ImidazoliumBased Ionic Liquid by TGA-MS Analysis and DFT Calculations, Thermochim. Acta 2010,
501, 78-83.
(64) Chambreau, S.D; Schenk, A.C.; Sheppard, A.J.; Yandek, G.G., Vaghjiani, G. L.;
Maciejewsi, J.; Koh, C.J., Golan, A., Leone, S. R. Thermal Decomposition Mechanism of
Alkylimidazolium Ionic Liquids with Cyano-Functionalized Anions, J. Phys. Chem. A 2014,
118, 11119-11132.
ACS Paragon Plus Environment
Page 28 of 42
Page 29 of 42
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
The Journal of Physical Chemistry
(65) Ramsey, B.G. Substituent Effects on Imidazole Basicity and Photoelectron Spectroscopy
Determined Ionization Energies, J. Org. Chem. 1979, 44, 2093.
(66) Bull, J.N.; Harland, P.W.; Vallance, C. Absolute Total Electron Impact Ionization CrossSection for Many-Atom Organic and Halocarbon Species. J. Phys. Chem. A 2012, 116, 767777.
(67) Jug, K.; Chiodo, S. Electronic and Vibrational Polarizabilities and Hyperpolarizabilities of
Azoles: A Comparative Study of the Structure-Polarization Relationship. J. Phys. Chem. A
2003, 107, 4172-4183.
(68) Miller, K.J. Additivity Methods in Molecular Polarizability. J. Am. Chem. Soc. 1990, 112,
8533-8542
(69) Verevkin, S.P.; Zaitsau, D.H.; Emel’yanenko, V.N. Thermodynamics of Ionic Liquids
Precursors: 1-Methylimidazole, J. Phys Chem. B 2011, 115, 4404-4411.
(70) Morrow, T.I.; Maginn, E.J.; Molecular Dynamics Study of the Ionic Liquid 1-n-Butyl-3Methylimidazolium Hexafluorophospate. J. Phys. Chem. 2002, 49, 12807-12813.
(71) Sambasivarao, S.V.; Acevedo, O. Development of OPLS-AA Force Field Parameters for
68 Unique Ionic Liquids. J. Chem. Theory Comput. 2009, 5, 1038–1050.
(72) Liu, Z.; Wu, X.; Wang, W. A Novel United-Atom Force Field for Imidazolium-Based
Ionic Liquids. Phys Chem. Chem. Phys. 2006, 8, 1096-1104.
(73) Mondal, A.; Balasubramanian, S. Quantitative Prediction of Physical Properties of
Imidazolium Based Room Temperature Ionic Liquids Through Determination of Condensed
Phase Site Charges: A Refined Force Field. J. Phys. Chem B 2014, 118, 3409-3422.
(74) Deyko, A.; Lovelock, K.R.J.; Corfield, J.-A.; Taylor, A.W.; Gooden, P.N.; Villar-Garcia,
I.J.; Licence, P.; Jones, R.G.; Krasovskiy, V.G.; Chernikova, E.A. et al. Measuring and
Predicting ∆vapH298 Values of Ionic Liquids. Phys Chem Chem Phys 2009, 11, 8544-8555.
(75) Červinka, C.; Pádua, A.A.H.; Fulem, M. Thermodynamic Properties of Selected
Homologous Series of Ionic Liquids Calculated Using Molecular Dynamics. J. Phys. Chem.
B 2016, 120, 2362-2371.
(76) Weerachanchai, P.; Chen, Z.; Leong, S.S.J.; Change, W.W., Lee, J.-M. Hildebrand
Solubility Parameters of Ionic Liquids: Effects of Ionic Liquid Type, Temperature and DMA
Fraction in Ionic Liquid. Chem. Eng. J. 2012, 213, 356-362.
(77) Tong, J.; Yang, H.-X.; Liu, R.-J.; Li, C.; Xia, L.-X.; Yang, J.-Z. Determination of the
Enthalpy of Vaporization and Prediction of Surface Tension for Ionic Liquid 1-Alkyl-3Methylimidazolium Propionate [CnMIm][Pro] (n=4,5,6). J. Phys. Chem. B 2014, 118,
12972-12978.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
(78) Rocha, M.A.A.; Ribeiro, F.M.S.; Lobo Ferreira, A.I.M.C.; Coutinho, J.A.P.; Santos,
L.M.N.B.F. Thermophysical Properties of [CN-1C1Im][PF6]. J. Mol. Liq. 2013, 188, 196-202.
(79) Swiderski, K.; Mclean, A.; Gordon, C.M.; Vaughan, D.H. Estimates of Internal Energies
of Vaporisation of Some Room Temeprature Ionic Liquids. Chem Commun. 2004, 21782179.
(80) Lee, S.H.; Lee, S.B. The Hildebrand Solubility Parameters, Cohesive Energy Densities
and Internal Energies of 1-Alkyl-3-Methylimidazolium-Based Room Temperature Ionic
Liquids. Chem. Commun. 2005, 3469-3471.
(81) Klomfar, J.; Součková, M.; Pátek, J. Surface Tension Measurements for Four 1-Alkyl-3Methylimidazolium-Based Ionic Liquids With Hexafluorophosphate Anion. J. Chem. Eng
Data 2009, 54, 1389-1394.
ACS Paragon Plus Environment
Page 30 of 42
Page 31 of 42
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
The Journal of Physical Chemistry
Table 1. Mass loss rate per unit surface Ja measured in KEML experiments with different
effusion hole diameters (Ø).
Ø = 0.2 mm
Ø = 3 mm
T /K
103 J/kg·s–1·m–2
T/K
105J/kg·s–1·m–2
T/K
106J·/kg·s–1·m–2
551.4
23.8
501.1
15.9
492.1
46.1
547.3
19.8
496.1
10.6
486.6
29.5
542.8
15.5
491.3
7.58
481.2
19.6
538.8
13.1
486.4
5.18
475.9
13.3
534.5
10.5
516.4
39.1
471.0
9.11
530.4
8.56
512.2
30.2
466.0
5.88
526.3
6.79
508.1
22.7
461.1
4.19
522.4
5.78
503.4
17.0
456.2
2.83
518.5
4.58
499.7
13.5
494.2
48.4
514.5
3.99
495.8
10.2
488.8
33.7
510.6
3.35
491.7
8.11
483.0
22.2
506.6
2.67
487.8
6.05
478.0
15.9
549.9
23.6
484.0
4.53
472.9
10.7
545.1
18.4
519.0
50.7
468.6
8.13
541.2
15.1
514.3
37.2
463.1
4.96
536.8
12.1
509.9
28.5
459.3
3.72
532.3
9.69
505.6
21.6
452.7
2.18
528.3
7.98
501.6
16.7
524.3
6.49
497.6
12.9
520.3
5.38
493.7
10.3
516.4
3.95
489.7
7.45
512.5
3.44
485.8
5.68
508.6
2.86
482.0
4.26
504.7
2.43
478.1
3.39
<T>b/K = 527.7
a
Ø = 1 mm
<T>b/K = 498.0
<T>b/K = 473.6
J = (WoAo)–1 ∆m/∆t. This quantity is proportional to p effusion MT −1 (see text)
b
<T> is the reciprocal of the mean of 1/T values
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Page 32 of 42
Table 2. Appearance energies (AE) measured for the most abundant ion species detected during KEMS
experiments on BMImNTf2 and BMImPF6
Ionic Liquid
Mass/u
Iona
AE/eV
Neutral precursor
BMImNTf2
139
BMIm+
9.3 ± 0.3
BMImNTf2(g)
BMImPF6
82
MIm+
12.0 ± 0.5
EIm(g) (?)
96
EIm+
8.3 ± 0.3
EIm(g)
139
BMIm+
11.3 ± 0.5
BMImPF6(g)
158
BMImF+
11.1 ± 0.3
BMImPF6(g)
a
Symbols B, E, M, Im stand for butyl, ethyl, methyl, imidazole/ium
ACS Paragon Plus Environment
Page 33 of 42
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
The Journal of Physical Chemistry
Table 3. Partial pressuresa of the species BMImPF6 and ethylimidazole in the vapor of BMImPF6(l), as
derived by KEMS experiments
T/K
BMImPF6/Pa
Ethylimidazole/Pa
495.2
2.9 10-2
6.6 10-3
514.5
9.0 10-2
2.3 10-2
474.5
6.2 10-3
1.6 10-3
455.5
1.3 10-3
3.3 10-4
493.0
2.3 10-2
5.3 10-3
527.8
1.5 10-1
3.9 10-2
482.2
9.6 10-3
2.0 10-3
468.3
3.3 10-3
6.9 10-4
509.1
4.1 10-2
9.8 10-3
458.0
1.5 10-3
3.6 10-4
504.2
2.6 10-2
5.5 10-3
a
a total uncertainty of 40% on pressure values is conservatively estimated, including the uncertainty on cross sections
and multiplier gain.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
Table 4. Evaporation enthalpy of BMImPF6(l) at 298 K ( ∆ evapH 298K ) evaluated from the third-law
o
analysis of BMImPF6(g) partial pressures measured by KEMS.
T/K
- ∆ evapGef (T ) a/J
o
∆evapH 298
K /kJ
K-1 mol-1
mol-1
o
495.2
165.84
144.1
514.5
164.07
144.0
474.5
167.74
145.1
455.5
169.27
145.8
493.0
166.05
144.5
527.8
162.84
144.9
482.2
167.04
145.4
468.3
168.27
145.9
509.1
164.56
146.1
458.0
169.09
146.2
504.2
165.01
146.8
Mean:
a
145.3 ±0.9
From refs 25 and 33
ACS Paragon Plus Environment
Page 34 of 42
Page 35 of 42
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
The Journal of Physical Chemistry
Table 5. Compilation of vaporization enthalpies (∆evapH298 K) proposed for BMImPF6 by different
theoretical approaches, compared with the experimental values. MD = Molecular Dynamics
∆evapH298 K /kJ mol-1
Method
Reference
161
MD
70
150.6
MD
13
133.5
MD
71
190.4
MD
72
152.9
MD
14
138.1
MD
73
157
Semiempirical model
74
192.8
MD
75
144.5
MD
11
149.9
empirical correlation
with surface tensiona
167-192
Hildebrand solubility
parameterb
146.5 ± 2.6
Experimental (QCM)
11
145.3 ±2.9
Experimental (KEMS)
This work
a
For this estimate we used the empirical correlation given in ref 77, with surface tension values from ref 81
and molar volume from ref 78 (see text)
b
For this estimate we used the Hildenbrand parameter values taken from refs 76,79 and 80, and the molar
volume from ref 78 (see text)
ACS Paragon Plus Environment
The Journal of Physical Chemistry
-2
400 K
500 K
600 K
BMImPF6
-4
-2
PF6 in Pt 0.2 mm
NTf2 3 mm
NTf2 1 mm
BMImNTf2
-1
-6
PF6 in Pt 3 mm
PF6 in Pt 1 mm
0.2 mm
0.3 mm
ln (J/kg s m )
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
Page 36 of 42
1 mm
-8
NTf2 0.3 mm
BMImNTf2
1 mm
-10
3 mm
-12
BMImPF6
3 mm
-14
0,0016
0,0018
0,002
0,0022
0,0024
0,0026
K/T
Fig. 1 Plot of ln J vs 1/T, where J = specific effusion flux for the ionic liquids BMImPF6 (open
symbols) and BMImNTf2 (filled symbols) measured with different orifices. The uncertainty
associated to datapoints is expressed by the symbol size.
ACS Paragon Plus Environment
Page 37 of 42
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
The Journal of Physical Chemistry
(a)
(b)
(c)
Fig. 2 TG (a), DTA (b) and DTG (c) curves of BMImPF6 at 2 K min–1 heating rate registered under
flowing Ar () and flowing N2 (− −).
ACS Paragon Plus Environment
The Journal of Physical Chemistry
BMImPF6 1 mm effusion hole T = 495 K
120
+
BMIm
139
Normalized intensity
100
80
60
EIm
BMImF
96
40
158
MIm
20
82
0
80
100
120
140
160
m/e (u)
(a)
BMImPF6 0.5 mm effusion hole T = 488 K
120
96 EIm
100
Normalized intensity
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
Page 38 of 42
80
+
BMIm
60
139
MIm
40
82
BMImF
158
20
0
80
100
120
140
160
m/e (u)
ACS Paragon Plus Environment
(b)
Page 39 of 42
BMImNTf2, T = 486 K
120
+
BMIm
139
100
Normalized intensity
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
The Journal of Physical Chemistry
80
60
40
20
HMIm+
83
0
80
100
120
140
160
m/e (u)
(c)
Fig. 3. Typical mass spectra for BMImPF6, with effusion hole size 1 mm (a) and 0.5 mm (b), and BMImNTf2
(c) vapors. Spectra are normalized to the most intense peak. Note that mass 139 is largely dominant in (c),
whereas a far more complex pattern is observed in (a) and (b). Note also that mass 96 is the most intense
peak in the BMImPF6 spectrum when the smaller orifice is used. The symbols M,E,B,Im stand for methyl,
ethyl, butyl, and imidazole/imidazolium, respectively.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
3
2,5
96/139 Intensity ratio
0.5 mm diameter
2
1,5
96/139 alum 1 mm
96/139 alum 0.5 mm
96/139 Pt 1 mm
1
0,5
0
420
1 mm diameter
440
460
480
500
520
540
T/K
3
2,5
158/139 Intensity ratio
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
158/139 alum 0.5 mm
158/139 Pt 1 mm
158/139 alum 1 mm
2
1,5
1
1 mm diameter and 0.5 mm in diameter
0,5
0
420
440
460
480
500
520
540
T/K
Fig. 4. Intensity ratios of KEMS peaks at m/e = 96 and 158 to the peak at m/e = 139 with effusion hole of
different size. Note that the cell material has practically no effect.
ACS Paragon Plus Environment
Page 40 of 42
Page 41 of 42
0
BMImNTf2 (ref. 9)
-2
-4
ln (P/Pa)
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
The Journal of Physical Chemistry
-6
BMImPF6 this work
-8
-10
-12
P(BMImPF6) KEMS 1 mm
P(BMImPF6) KEML 1 mm
P(BMImPF6) KEML 3 mm
P(BMImPF6) QCM (ref. 11)
P(BMImNTf2) KEML (ref. 9)
-14
0,0018 0,0019
0,002
BMImPF6
QCM (ref. 11)
0,0021 0,0022 0,0023 0,0024 0,0025
K/T
Fig. 5. Comparison between the partial pressure of BMImPF6(g) evaluated by the KEMS spectra (open
circles) and that measured by KEML with estimated average molecular weight (crosses), 1 mm orifice in
both cases). In the plot are also reported the BMImPF6 pressures estimated by mass loss with the 3 mm
orifice (neglecting decomposition), the QCM values reported by Zaitsau et al.11 and the BMImNTf2 vapor
pressure.9 The uncertainty associated to each dataset is expressed by the symbol size.
ACS Paragon Plus Environment
The Journal of Physical Chemistry
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
TOC GRAPHIC
ACS Paragon Plus Environment
Page 42 of 42
Документ
Категория
Без категории
Просмотров
2
Размер файла
718 Кб
Теги
acs, 7b08523, jpcb
1/--страниц
Пожаловаться на содержимое документа