Thermochimica Acta 664 (2018) 81–90

Contents lists available at ScienceDirect

Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca

Measurements of the density, refractive index, electrical conductivity, T

thermal conductivity and dynamic viscosity for tributylmethylphosphonium
and methylsulfate based ionic liquids
⁎
Eva Rodila, , Alberto Arce Jr.b, Alberto Arcea, Ana Sotoa
a
Department of Chemical Engineering, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Spain
b
School of Engineering, Universidade de A Coruna, 15403 Ferrol, Spain

A R T I C LE I N FO A B S T R A C T

Keywords: In this work, the effect of ion and temperature change on the thermophysical properties of several ionic liquids
Ionic liquids was investigated. Five room-temperature ionic-liquids were selected: tributylmethylphosphonium chloride
Thermal stability ([P4441]Cl), tributylmethylphosphonium bis(trifluoromethylsulfonyl) amide ([P4441][NTf2]), tributylmethyl-
Thermophysical properties phosphonium methylsulfate ([P4441][MeSO4]), 1-methylpyridinium methylsulfate ([C1py][MeSO4]), and 1-
Temperature dependence
butyl-3-methylimidazolium methylsulfate ([C4mim][MeSO4]). Densities, refractive indices, electrical con-
ductivities, thermal conductivities and dynamic viscosities were reported ranging from (283.15 to 343.15) K at
atmospheric pressure. The thermal stability of all the ionic liquids has been also studied at a heating rate of
5 K min−1 using a thermogravimetric analyzer. The effect of temperature change on density was discussed in
terms of ion volume, and the influence on electrical conductivity and dynamic viscosity was discussed in terms of
ion mobility. The effect of the variation of the ion was discussed in terms of size, steric effects and interaction
strength between cation and anion. Finally, the temperature dependence for density, refractive index and
thermal conductivity was correlated with linear functions, while a Vogel-Fulcher-Tammann type function was
used for the correlation of the temperature dependence for electrical conductivity and dynamic viscosity.

1. Introduction
A large number of researchers are devoted to demonstrating the
great potential that ionic liquids (ILs) have for practical applications
due to their nature and qualities [1]. Among these applications, their
use as heat-transfer fluids can be highlighted [2]. Recent works [3–9]
show that the behavior of the thermal properties with temperature of
ILs is similar to those of molecular liquids, including commercial heat-
transfer fluids, such as the Dowtherm A® [10]. For this reason, the
knowledge of ILs temperature dependence properties is of major im-
portance for their application in practical processes. Such information,
considering the diverse nature and large number of ILs, raises the de-
velopment of reliable predictive models for which accurate experi-
mental determinations are critical.
In the current bibliography, a significant number of studies of ionic-
liquid families can be found. Careful selection and study of complete
sets of compounds, representing different families, can provide in-
formation about the relationships between the property and the IL
structure and allow to obtain suitable correlations, state equations, or
other models for these properties. For example, Bhattacharjee et al.
[11], Deive et al. [12] and Adomavá et al. [13] study the thermo-
physical properties of phosphonium-based ILs with different cations
and anions. Ferreira et al. [5] focus their study on the transport and
thermal properties of several quaternary phosphonium ionic liquids.
Beigi et al. [14], Soriano et al. [15], Fernández et al. [16], García-Miaja
et al. [17] and Zhang et al. [18] investigate physical and electro-
chemical properties of alkyl-imidazolium based ionic liquids. The data
studied in these works are experimental measurements of density, dy-
namic viscosity, refractive index and surface tension as a function of the
temperature. Less numerous are studies of thermal conductivity, while
no systematic studies are found for electrical conductivity.
In this work, density, refractive index, electrical conductivity,
thermal conductivity and dynamic viscosity for five ionic liquids,
[P4441]Cl, [P4441][NTf2], [P4441][MeSO4], [C1py][MeSO4], and
[C4mim][MeSO4], are experimentally determined from (283.15 to
343.15) K and at atmospheric pressure. Thermal stability of the studied
ILs is analyzed by TGA. The selection of liquids in this work is done in
order to make a systematic study of the influence of anions and cations
on the properties of the liquids. These five ionic liquids can be orga-
nized into two groups, one with three ILs containing the same cation,

⁎
Corresponding author.
E-mail address: eva.rodil@usc.es (E. Rodil).

https://doi.org/10.1016/j.tca.2018.04.007
Received 22 September 2017; Received in revised form 10 April 2018; Accepted 14 April 2018
Available online 18 April 2018
0040-6031/ © 2018 Elsevier B.V. All rights reserved.
E. Rodil et al. Thermochimica Acta 664 (2018) 81–90

tributylmethylphosphonium ([P4441]+), and another with three ILs

wH2O (ppm)
containing the same anion, methylsulfate ([MeSO4]−). Comparisons
among those having the same cation or anion are established. For ILs

350
175

200
50

50
for which literature values of the measured properties are available, a

–
–
–
–
–
comparison is made to prove that in this study the measurements are
correct.

Analysis method
2. Experimental

NMR
NMR
NMR
NMR
NMR
GCa
GCa
GCa
GCa
GCa
2.1. Chemicals

–
The [C1py][MeSO4] and [P4441][NTf2] ILs were synthesized in our

Purity (wt%)
laboratory as described below. The CAS (Chemical Abstract Service)
number, source and grades of the other ILs used in this work are as

≥99.8
≥99.5

≥99.8
≥99.5
follows: [C4mim][MeSO4] (401788-98-5, Sigma-Aldrich, ≥95%),

≥99
≥99
≥99
≥99
≥99

≥99
[P4441]Cl (2304-24-7, IoLiTec, ≥95%) and [P4441] [MeSO4] (34217-64-

–
6, IoLiTec, ≥95%). In order to minimize the water-content influence in
the properties of the pure liquids, all the ILs were dried under vacuum

pump
pump
pump
pump
pump
and moderate temperature for 48 h and their water content was de-

vacuum
vacuum
vacuum
vacuum
vacuum
termined using a titration method (Metrohm 737 KF coulometer). The
water content of the ILs was, in all the cases, lower than 400 ppm. The
specifications of the chemicals source are presented in Table 1 and the

high
high
high
high
high
chemical structures of the ILs are shown in Table S1 in Supporting In-

and
and
and
and
and
formation. The structures of prepared and purchased ILs were con-

Rotatory evaporator
Rotatory evaporator
Rotatory evaporator
Rotatory evaporator
Rotatory evaporator
firmed by 1H and 13C NMR and the corresponding spectra have been

Purification method
added to Supporting Information, Figs. S1 to S5.

Distillation
2.1.1. Synthesis of the ionic liquids

none
none
none
none
none
The ionic liquid [C1py][MeSO4] (CAS 37943-43-4) was synthesized
by alkylation of pyridine (Riedel-de Häen, ≥95%) with dimethylsulfate
(Aldrich, ≥97%) as described elsewhere [19]. Pyridine was placed in a
round-bottomed flask, with a reflux condenser attached, located in an
Purity stated by suppliers (wt%)

ice bath to keep the temperature low due to the exothermic character of
the reaction. Careful dropwise addition of dimethylsulfate was carried
out under Argon atmosphere. Once finished and after melting of the ice
and natural warming of the water to room temperature, the flask was
placed in an oil bath and heated up to 373 K for 8 h. The resulting
mixture, with an ionic liquid appearance, was placed in a rotary eva-
porator for removal of unreacted materials or other volatile impurities.
≥99.8
≥99.5

≥99.8
≥99.0
> 95
> 95
≥95

≥99

This removal of volatiles was completed by heating of the mixture at ca.

–

–
–

343 K, with stirring, for 48 h under high vacuum (< 0.1 mbar).
The ionic liquid [P4441][NTf2] (CAS 324575-10-2) was synthesized
Riedel-de Häen

via methathetic reaction through [P4441] [MeSO4](IoLiTec, ≥95%),

Sigma-Aldrich
Home made
Home made

following a procedure similar to that described previously [12]. [P4441]

Solvionic
Aldrich
IoLiTec

[MeSO4] was dissolved in water and mixed with slight excess of an

Source

Iolitec

Fluka
Fluka

aqueous solution of lithium bis(trifluoromethyl-sulfonyl)imide (Sol-

–

vionic, ≥99%), at ambient temperature. Turbidity was observed in-

stantaneously due to the hydrophobibity of the produced IL, however
401788−98-5
324575−10-2

the medium was stirring for at least 4 h. Dichloromethane (Fluka,

69056−62-8

37943−43-4

90076−65-6
7732−18-5
2304−24-7

110−86-1

≥99.9%) was added, dissolving the [P4441][NTf2]. This organic phase

75−09-2
77-78−1
CAS no.

56-81-5
Specification of pure components used in this work.

was then separated and washed several times with fresh water. To se-
parate the dichloromethane, rotary evaporation was used and for the
removal of the residual volatile impurities present in the IL, it was
connected to a high vacuum line (< 0.1 mbar), while continuous stir-
Lithium bis(trifluoromethyl-sulfonyl)imide

ring and heating at 343.15 K for at least 48 h.

2.1.2. Characterization of ILs

2.1.2.1. Tributylmethylphosphonium chloride ([P4441]Cl). 1H NMR
Gas chromatography.

(CDCl3, 300 MHz) δH ppm 0.98 (t, J = 7.0 Hz, 9H, 3 × P(CH2)3CH3),
1.40–1.70 (m, unresolved, 12H, 3 × P(CH2)CH2CH2CH3), 2.12 (d,
J = 13.5 Hz, 3H, PCH3), 2.36–2.58 (m, unresolved, 6H, 3 × PCH2);
1,2,3-Propanetriol
[C4mim][MeSO4]

Dichloromethane
Dimethyl sulfate
[P4441] [MeSO4]
Chemical name

[C1py][MeSO4]

13
C NMR (CDCl3, 75.4 MHz) δC ppm 4.8 (d, JP-C = 52 Hz, PCH3), 13.4
[P4441][NTf2]

(3 × P(CH2)3 CH3), 20.0 (d, JP-C = 49 Hz, 3 × PCH2), 23.5 (d, JP-
[P4441]Cl

Pyridine

C = 5 Hz, 3 × PCH2 CH2), 23.7 (d, JP-C = 16 Hz, 3 × PCH2 CH2CH3).

Table 1

Water

2.1.2.2. Tributylmethylphosphonium bis(trifluoromethylsulfonyl)imide

82
E. Rodil et al.
([P4441][NTf2]). 1H NMR (CDCl3, 300 MHz) δH ppm 0.95 (t, J = 7.0 Hz,
9H, 3 × P(CH2)3CH3), 1.35–1.60 (m, unresolved, 12H, 3 × P
(CH2)CH2CH2CH3), 1.75 (d, J = 13.3 Hz, 3H, PCH3), 2.00–2.20 (m,
unresolved, 6H, 3 × PCH2); 13C NMR (CDCl3, 75.4 MHz) δC ppm 3.8 (d,
JP-C = 53 Hz, PCH3), 13.1 (3 × P(CH2)3 CH3), 19.8 (d, JP-C = 50 Hz,
3 × PCH2), 23.3 (d, JP-C = 5 Hz, 3 × PCH2 CH2), 23.6 (d, JP-C = 16 Hz,
3 × PCH2 CH2CH3), 119.9 (q, JC-F = 321 Hz, 2 × CF3).
2.1.2.3. Tributylmethylphosphonium methylsulfate ([P4441][MeSO4]). 1H
NMR (CDCl3, 300 MHz) δH ppm 0.82 (t, J = 7.0 Hz, 9H, 3 × P
(CH2)3CH3), 1.25–1.50 (m, unresolved, 12H, 3×P
(CH2)CH2CH2CH3), 1.78 (d, J = 13.5 Hz, 3H, PCH3), 2.05–2.25 (m,
unresolved, 6H, 3 × PCH2), 3.52 (s, 3H, OCH3); 13C NMR (CDCl3,
75.4 MHz) δC ppm 4.0 (d, JP-C = 51.5 Hz, PCH3), 13.3 (s, 3 × P(CH2)3
CH3), 19.8 (d, JP-C = 49.3 Hz, 3 × PCH2), 23.5 (d, JP-C = 4.3 Hz,
3 × PCH2 CH2), 23.7 (d, JP-C = 15.9 Hz, 3 × PCH2 CH2CH3), 54 (s,
OCH3).
2.1.2.4. Methylpyridinium methylsulfate ([C1py][MeSO4]). 1H NMR
(DMSO-d6, 300 MHz) δH ppm 3.40 (s, 3H, OCH3), 4.39 (s, 3H, NCH3),
8.11 (t, J = 7.1 Hz, 2H, C(3)H and C(5) H), 8.57 (t, J = 7.8 Hz, 1H, C(4)
H), 8.97 (d, J = 5.6 Hz, 2H, C(2)H and C(6) H); 13C NMR (DMSO-d6,
75.4 MHz) δC ppm 47.8 (NCH3), 52.8 (OCH3), 127.6 (C(4)H), 145.0
(unresolved, C(3)H and C(5)H), 145.5 (unresolved, C(2)H and C(6)H).
2.1.2.5. 1-butyl-3-methylimidazolium methylsulfate ([C4mim]
[MeSO4]). 1H NMR (CDCl3, 300 MHz) δH ppm 0.66 (t, J = 7.26 Hz,
3H, CH3CH2), 1.12 (m, J = 7.6 Hz, 2H, CH2CH3), 1.63 (p, J = 7.5 Hz,
2H, NCH2CH2), 3.43 (s, 3H, OCH3), 3.75 (s, 3H, NCH3), 4.00 (t,
J = 7.5 Hz, 2H, NCH2), 7.34-7.38 (m, unresolved, 2H, C(4)H and
C(5)H), 9.12 (s, 1H, C(2)H); 13C NMR (CDCl3, 75.4 MHz) δC ppm
13.3 (CH2CH3), 19.3 (CH2CH3), 32.0(NCH2CH2), 36.3 (NCH3), 49.7
(NCH2), 54.3 (OCH3), 122.2 (C(4)H), 123.8 (C(5)H), 137.3 (C(2)H).
2.2. Apparatus and procedure
At least two measures were made for each property and tempera-
ture, making sure that they were repetitive within the uncertainty of the
measurement, and the average value was recorded.
2.2.1. Thermal stability
Thermal stability of ILs was measured with a thermo-gravimetric
analyzer (TGA, TA Hi-Res. Instruments Q500) with a weighing preci-
sion of ± 0.01%. The samples (ca. 15–20 mg) were placed in an open
platinum pan under a flow of N2 gas. The measuring method consisted
in a ramp at a heating rate of 10 K min−1 from (283–373) K, then kept
isothermal at 373 K for 60 min to ensure that all the water has been
eliminated and followed by a ramp at a heating rate of 5 K min−1 from
(373–873) K. The evaluation of the TGA curves was made with the
Universal Analysis 2000 software, version 4.5.0.5, by TA Instruments.
2.2.2. Density
Densities were measured using an Anton Paar vibrating U-tube
densimeter (DMA-5000) with viscosity correction and with self-control
of temperature to ± 0.001 K. The equipment was calibrated using
distilled water and air. The uncertainty in density is estimated to be
0.001 g cm−3.
2.2.3. Refractive index
Refractive indices were measured with an uncertainty of 2·10−4 in
an ATAGO RX-5000 refractometer connected to a Techne RB-5a ther-
mostatic bath to keep temperature constant ( ± 0.05 K). The equipment
was calibrated using distilled water.
2.2.4. Electrical conductivity
A Crison Basis 30 electrical conductivimeter using a DC conductivity
83
Thermochimica Acta 664 (2018) 81–90
cell was used to measure the electrical conductivity with a reproduci-
bility of ± 0.1% and an error lesser than ± 5%. The samples were
measured in jacketed cell connected to a Selecta Ultraterm thermostatic
bath to control the temperature ( ± 0.05 K).
2.2.5. Thermal conductivity
The thermal conductivity was measured using KD2 Pro Thermal
Properties Analyzer (Labcell Ltd., UK). The measuring principle is based
on the transient hot wire method. The meter consists of a thermal probe
(KS-1, 1.3 mm diameter, 60 mm length), containing a heating element
and a thermoresistor, which is vertically inserted into the sample to
reduce the possibility of generating convection. The thermal probe is
electrically isolated from the media to measure. The hot wire technique
is a standard and rapid method, in which a thermal pulse is sent to a
very thin wire and response time of temperature during each thermal
pulse is monitored either in the hot wire or in an adjacent wire. A
jacketed cells made of glass connected to a thermostatic bath (Julabo
F12-EH) is utilized to keep a constant temperature. The essence of
temperature response depends on thermal properties of the material.
The sensor was checked, as suggested by the supplier, by measuring the
thermal conductivity of a verification liquid, Glycerol (Fluka, > 99%).
The results are presented in the Table S2 in Supporting Information.
The validation was performed in a range of temperature between
(283.15 and 333.15) K and was found to be in very good agreement
with literature values of 0.284 W m−1 K−1 for Glycerol over the same
temperature range [20,21]. At least five measurements were taken at
each temperature. An uncertainty of 0.001 W m−1 K−1 was estimated
from the standard deviations of experimental and validation measure-
ments. The uncertainty of the temperature is 0.05 K.
2.2.6. Dynamic viscosity
Kinematic viscosity (ν) was determined using a micro-Ubbelohde
viscometer technique. Previous to its use, the Newtonian character of
the ILs was checked by using an Anton Paar Physica MCR 301 rhe-
ometer. The capillaries used (types II and III) were manufactured, ca-
librated and certified by Schott and tested by ourselves with reference
liquids of known viscosities. Flow time measurement was performed in
a Lauda Processor Viscosity system PVS1 with a resolution of 0.01 s.
The temperature of the viscometer was kept constant using a Lauda
clear view thermostat D 20 KP with a through-flow cooler DLK 10.
Dynamic viscosity is given by the formula:
η = ν·ρ = K·t·ρ (1)
where η and ν are the dynamic and kinematic viscosity respectively, ρ is
the density of the sample, K is the capillary constant provided by the
manufacturer, and t is the flow time. The uncertainty for the dynamic
viscosity is estimated to be 0.5%.
3. Data correlation
The variation of the density (ρ), the refractive index (nD) and the
thermal conductivity (λ) with the temperature was correlated to the
equation:
Y = A0 + A1 T (2)
−1
where Y is the density, in g mL , or the refractive index, or the thermal
conductivity, in W m−1 K−1, and T is the temperature in K. A0 and A1
are fitting parameters.
In the case of the electrical conductivity and dynamic viscosity the
temperature change was correlated using an empirical Vogel-Fulcher-
Tammann type equation:
B
ln Y = A +
T−C (3)
−1
where Y is the electrical conductivity, in μS cm , or the dynamic
viscosity, in mPa s, and T is the temperature in K. A, B, and C are fitting
E. Rodil et al. Thermochimica Acta 664 (2018) 81–90

Table 2
Density, ρ, refractive index, nD, electrical conductivity, σ, thermal conductivity, λ, dynamic viscosity, η, and thermal expansion coefficient αp of the pure RTILs at the
temperature of 293.15 K and at 0.1 MPa.
Property [P4441]Cl [P4441][NTf2] [P4441][MeSO4] [C1py][MeSO4] [C4mim][MeSO4]

This work Literature This work Literature This work Literature This work Literature This work Literature

ρ (g cm−3) 0.95867 – 1.25868 1.25799 [30] 1.0604 1.0703 [11], 1.34810 1.34822 [23], 1.21196 1.21018 [25], 1.21295
1.06163 [12] 1.34612 [24] [16], 1.21562 [22], 1.2053
[14]
nD 1.5037 – 1.4374 – 1.4755 1.47742 [11], 1.5139 1.51431 [23], 1.4804 1.48076 [22], 1.4777 [14]
1.47489 [12] 1.51340 [24]
σ (μS cm−1) 25.40 – 367.5 – 74.95 – 2875 – 1041 1260 [26]
λ (W m−1 K−1) 0.159 – 0.126 – 0.160 0.156 [5] 0.205 – 0.177 –
η (mPa·s) 13511 14065.8 [13] 299.3 – 1434 2476.4 [11], 113.5 – 269.1 288.99 [22], 122.3 [14]
719.09 [12]
αp·104 (K−1) 5.64 5.75 [13] 6.59 – 5.76 – 5.05 5.02 [23] 5.36 4.81 [14]

−3
Standard uncertainties: u(P) = 5 kPa; u(ρ) = 0.001 g cm with u(T) = 0.01 K; u(nD) = 0.0002, u(σ) = 5%, uc(η) = 0.5%, u(λ) = 0.001 W m−1 K−1 with u
(T) = 0.05 K.

parameters, the last two in K.
The S.D. values were calculated by applying the following expres-
sion:
1
2
S. D = [∑ (z exp − z calc )2 / n]
where, z and n are the property values and the number of experimental
points, respectively.
4. Results and discussion
The experimental values of density, refractive index, electrical
conductivity, thermal conductivity and dynamic viscosity at 293.15 K
and atmospheric pressure for [P4441]Cl, [P4441][NTf2], [P4441][MeSO4],
[C1py][MeSO4], and [C4mim][MeSO4], are compared with literature
values, when available, in Table 2. In general a good agreement was
found but a more detailed comparison, at several temperatures, is
presented below.
values of the literature, with mean relative deviations (MRD) with va-
lues ranging from (0.01 to 0.7) %. The density measurements for
[C4mim][MeSO4] are in agreement with those reported by Beigi et al.
[14], Soriano et al. [15], Fernández et al. [16], García-Miaja et al. [17],
(4) Pereiro et al. [22], Galán Sánchez et al. [25], Mbondo Tsamba et al.
[28], and Pandit et al. [29] with MRD of 0.4%, 0.09%, 0.04%, 0.05%,
0.1%, 0.07%, 0.09%, and 0.02%, respectively. In the case of [C1py]
[MeSO4], MRD with data reported by Gómez et al. [23] is 0.01%. Two
comparisons could be established for [P4441][MeSO4], MRD with data
reported by Deive et al. [12] and Bhattacharjee et al. [11] are 0.06%
and 0.7%, respectively. Densities were reported by Adamová et al. [13]
for [P4441]Cl and by Tagiuri et al. [30] for [P4441][NTf2], MRD obtained
are 0.5% and 0.03%, respectively.
Densities were correlated with temperature using Eq. (2). The cor-
relation parameters for all the ILs are summarized in Table 4 together
with the standard deviations of the fits.
To estimate the thermal expansion coefficients, the density values
were used as a function of temperature with the following equation:

4.1. Density and thermal expansion 1 δρ A1

αp = − ⎛ ⎞ = − ⎛ ⎜
⎞ ⎟

ρ ⎝ δT ⎠P ⎝ A0 + A1 T ⎠ (5)
The density data for the ILs at temperature range from (283.15 to −1
where αp is the thermal expansion coefficient in K , ρ is the density
343.15) K are listed in Table 3 and showed in Fig. 1. The density of the
and A0 and A1 are fitting parameters of Eq. (2), and T is the tempera-
studied ILs, at a given temperature, follows the order: [C1py]
ture.
[MeSO4] > [P4441][NTf2] > [C4mim][MeSO4] > [P4441]
As it can be seen in Table 2, thermal expansion coefficients of the ILs
[MeSO4] > [P4441]Cl. Furthermore, the density of the studied ILs de-
studied in this work are smaller than those of common organic solvents
creases in an almost linear fashion with increasing temperature. As
(such as alkanes or alcohols) but larger than water, and are closer to
previously reported in the literature, the density of phosphonium-based
glycols [27]. Values obtained are in the same order of magnitude than
ILs is usually lower than that of the imidazolium and pyridinium
thermal expansion coefficients reported for RTILs of the same families
[11,14,23].
[14,23,31,32]. All values vary with temperature in an increasing trend
As depicted in Fig. 1a for a common cation, [P4441]+, there is an
although the variation is weak. The [P4441][NTf2] shows the largest
increase in the density with the increase of the molecular weight of the
values, 6.55·10−4 K−1 at 293.15 K, while the [C1py][MeSO4] shows the
anion, and so the overall size and charge dispersion. At 293.85 K of
smallest values, 5.02·10−4 K−1 at the same temperature.
temperature, with the largest anion, the [NTf2]−, density is
Only one work [14] was found presenting thermal expansion coef-
1.25868 g mL−1, while with the smallest, the Cl−, density is
ficients data for one of the ILs studied, [C4mim][MeSO4]. The com-
0.95867 g mL−1. With the [MeSO4]− the value lies in between,
parison of values leads to a MRD of 1%. This difference is probably
1.06659 g mL−1. In Fig. 1b the same anion, [MeSO4]−, is analyzed with
because they are using a second order polynomial to fit the temperature
different cations. The cation [P4441]+, contains four alkyl chains around
dependency of the density.
a phosphorus atom giving this cation a large volume which explains the
lowest density, contrary to the cases of [C4mim]+ and [C1py]+ whose
smaller volumes gives the ILs larger densities, 1.21196 and 4.2. Refractive index
1.3481 g mL−1, respectively at 293.85 K of temperature. For some of
the studied ILs the density data are already available in the literature The refractive index of a compound indicates its ability to refract
[11–17,22,23,25,28–30] and presented in Fig. 1(a and b). The experi- light as it enters the medium, therefore, the greater the refractive index
mental data of the ILs studied are in agreement with the available of a compound, the more refracted light. For applications, a substance is

84
E. Rodil et al. Thermochimica Acta 664 (2018) 81–90

Table 3
Experimental density, ρ, refractive index, nD, electrical conductivity, σ, thermal conductivity, λ, dynamic viscosity, η, and calculated thermal expansion coefficient αp,
as a function of temperature and at 0.1 MPa.
T [K] ρ (g mL−1) nD σ (μS cm−1) η (mPa s) αp·104(K−1) λ (W m−1 K−1)

[P4441]Cl
283.15 0.96423 1.5053 5.61
293.15 0.95867 1.5037 25.40 13511 5.64 0.159
303.15 0.95344 1.4999 58.95 4692 5.67 0.158
313.15 0.94808 1.4972 130.8 1914 5.70 0.157
323.15 0.94274 1.4952 290.0 859.0 5.73 0.156
333.15 0.93742 1.4923 520.0 416.8 5.77 0.155
343.15 0.93212 1.4888 941.5 5.80 0.154

[P4441][NTf2]
283.15 1.26721 1.4403 616.7 6.55 0.126
293.15 1.25868 1.4374 367.5 299.3 6.59 0.126
303.15 1.25031 1.4344 743.0 156.0 6.64 0.125
313.15 1.24207 1.4314 1232 87.7 6.68 0.122
323.15 1.23384 1.4285 1873 54.7 6.73 0.123
333.15 1.22564 1.4255 2785 38.4 6.77 0.122
343.15 1.21749 1.4228 4045 27.5 6.82 0.121

[P4441][MeSO4]
283.15 1.06659 1.4782 5.72
293.15 1.06040 1.4755 74.95 1434 5.76 0.160
303.15 1.05425 1.4728 158.2 650.6 5.79 0.159
313.15 1.04813 1.4702 278.5 339.2 5.82 0.159
323.15 1.04202 1.4673 483.0 189.4 5.86 0.158
333.15 1.03612 1.4647 881.5 113.9 5.89 0.157
343.15 1.03018 1.4619 1390 72.4 5.93 0.157

[C1py][MeSO4]
283.15 1.3550 1.5164 199.0 5.02 0.205
293.15 1.3481 1.5139 2875 113.5 5.05 0.205
303.15 1.3413 1.5113 4500 69.2 5.07 0.204
313.15 1.3345 1.5088 6835 46.0 5.10 0.204
323.15 1.3278 1.5063 9860 32.0 5.12 0.203
333.15 1.3211 1.5039 13565 23.0 5.15 0.202
343.15 1.3145 1.5013 17505 17.3 5.18 0.201

[C4mim][MeSO4]
283.15 1.21878 1.4831 531.6 5.33 0.179
293.15 1.21196 1.4804 1041 269.1 5.36 0.177
303.15 1.20533 1.4781 1770 151.9 5.39 0.176
313.15 1.19879 1.4753 2785 91.7 5.42 0.176
323.15 1.19226 1.4729 4260 59.6 5.44 0.174
333.15 1.18576 1.4704 6225 40.5 5.47 0.173
343.15 1.17931 1.4678 8845 28.9 5.50 0.172

Standard uncertainties: u(P) = 5 kPa; u(ρ) = 0.001 g cm−3 with u(T) = 0.01 K; u(nD) = 0.0002, u(σ) = 5%, uc(η) = 0.5%, u(λ) = 0.001 W m−1 K−1 with u
(T) = 0.05 K.

considered to have a high refractive index when its value is greater than
1.4, above that of the range of the most common organic materials.
Table 3 and Fig. 2 shows temperature dependence of the refractive
index for the studied ILs. Values higher than 1.4 were found in all the
cases. The refractive index linearly decreases with increasing the tem-
perature. Furthermore, the refractive index increases in the following
sequence: [P4441][NTf2] < [P4441][MeSO4] < [C4mim][MeSO4] <
[P4441]Cl < [C1py][MeSO4].
As it can be seen in Fig. 2a for a common cation, [P4441]+, changing
the anion has some influence in the refractive index. At any tempera-
ture, the refractive indices with Cl− are larger, 1.50371 at 293.85 K,
than with [NTf2]− and [MeSO4]− but between these two the former
shows the smallest values, 1.43738 at the same temperature. In Fig. 2b,
refractive indices of the three ILs with the same anion, [MeSO4]−, are
also presented. The cation [C1py]+ leads to the largest refractive in-
dices, showing a value of 1.51275 at 293.85 K of temperature.
For comparative purposes, refractive indices measured in this work
and some data found in literature are showed in Fig. 2. Experimental
data for [C4mim][MeSO4] are in good agreement with those reported
by Beigi et al. [14], Soriano et al. [15], and Pereiro et al. [22], with
MRD of 0.05%, 0.02%, and 0.01%, respectively. Data at several tem-
peratures for [P4441][MeSO4] were reported by Deive et al. [12] and
Bhattacharjee et al. [11], MRD are 0.05% and 0.05%, respectively. Only
one work was found containing data for [C1py][MeSO4], Gómez et al.
[23], and MRD is 0.01%. In the case of [P4441][NTf2] and [P4441]Cl, no
data have been found in the available literature.
Refractive indices are linearly correlated with temperature. The
parameters and mean standard deviations are listed in Table 4 for all
the ILs.
4.3. Electrical conductivity
Electrical conductivity data for the ILs studied are listed in Table 3.
The differences between the electrical conductivity can be explained by
the interaction strength between cation and anion, the ion volumes, and
steric effects affecting these volumes. Fig. 3 shows temperature de-
pendence of electrical conductivity for the ILs studied. At a given
temperature, the electrical conductivity follows the order: [C1py]
[MeSO4] > [C4mim][MeSO4] > [P4441][NTf2] > [P4441]
[MeSO4] > [P4441]Cl. The electrical conductivity of the ILs studied
increases with increasing temperature. The [P4441]Cl shows the lowest
electrical conductivity, 12 μσ cm−1 at 293.85 K, on the contrary the
[C1py][MeSO4] shows the largest values, 3230 12 μσ cm−1 at the same
temperature. As previously stated, the cation [P4441]+, contains four
alkyl chains around a phosphorus atom giving this cation a large vo-
lume. The anion Cl− is smaller compared to [P4441]+ and will tend to

85
E. Rodil et al. Thermochimica Acta 664 (2018) 81–90

Table 4
Parameter A0 and A1 and standard deviations, SD, of the fittings with Eq. (2) for
density, ρ, refractive index, nD and thermal conductivity λ.
Physical Property A0 A1 SD

[P4441]Cl
ρ (g mL−1) 1.11722 −0.00054 0.00014
nD 1.58524 −0.00028 0.00045
λ (W m−1·K−1) 0.1920 −0.00011 0.0004

[P4441][NTf2]
ρ (g mL−1) 1.50210 −0.00083 0.00013
nD 1.52227 −0.00029 0.00009
λ (W m−1 K−1) 0.1549 −0.00010 0.0005

[P4441][MeSO4]
ρ (g mL−1) 1.23924 −0.00061 0.00012
nD 1.55464 −0.00027 0.00008
λ (W m−1 K−1) 0.1795 −0.00007 0.0009

[C1py][MeSO4]
ρ (g mL−1) 1.54750 −0.00068 0.00014
nD 1.59338 −0.00027 0.00039
λ (W m−1 K−1) 0.2264 −0.00007 0.0006

[C4mim][MeSO4]
ρ (g mL−1) 1.40239 −0.00065 0.00018
nD 1.55682 −0.00026 0.00015
λ (W m−1 K−1) 0.2115 −0.00011 0.0006

4.4. Thermal conductivity

The experimental thermal conductivities of the studied ILs are listed

Fig. 1. Experimental density as a function of temperature: (–) Eq. (2).
(a) Same cation:
[P4 4 4 1]Cl: (●) this work, (△) ref. [13],
[P4 4 4 1] [MeSO4]: (▼) this work, (×) ref. [11], (▽) ref. [12],
[P4 4 4 1][NTf2]: (○) this work, (♢) ref. [30],
(b) Same anion:
[P4 4 4 1][MeSO4]: (▼) this work, (×) ref. [11], (▽) ref. [12],
[C4mim][MeSO4]: (□) this work, (●) ref. [14], (△) ref. [15], (♦) ref. [16], (−)
ref. [17], (▲) ref. [22], (+) ref. [25], (*) ref. [28], (−) ref. [29],
[C1py][MeSO4]: (♢) this work, (○) ref. [23].
stick to the phosphorus atom [13], this strong interaction leads to a low
electrical conductivity because there will be less unpaired ions re-
sponsible for the current transport. As increasing the size of the anion
the interaction between the ions is weaker increasing the electrical
conductivity at any temperature as it is showed in Fig. 3a. The inter-
action strength between the cation [C1py]+ and the anion [MeSO4]− is
weak as denotes its large electrical conductivity. This means that the
average distance between cations and anions is larger. In addition, the
[C1py]+ does not suffer any steric effect due to presence of the
[MeSO4]−, and its size is not altered. At any temperature, for the three
ILs with the same anion, [MeSO4]−, by changing cation, as the cation
volume increases the electrical conductivity decrease as it can be ob-
served in Fig. 3b. Only reports of electrical conductivity were found in
the surveyed literature for pure [C4mim][MeSO4]. Experimental elec-
trical conductivities of [C4mim][MeSO4] are compared with literature
values in Fig. 3b. At low temperatures, a good agreement with values
reported by Mbondo Tsamba et al. [28], Pandit et al. [29] and Yu et al.
[33] has been found. However, as temperature increases the deviation
is large resulting in MRD of 6, 8 and 9%, respectively.
A Vogel-Fulcher-Tammann type function (Eq. (3)) was used to
correlate electrical conductivity of the ILs with temperature. Results are
showed in Table 5.
in Table 3 and showed in Fig. 4. The ILs in this work have a similar
thermal conductivities, their values being within the range of (0.1 to
0.2) W m−1 K−1, like most molecular liquids. The thermal conductivity
presents very small temperature dependence, showing a slight linear
decrease with the increase of the temperature, a similar tendency in all
ILs and previously reported in the literature [5]. The thermal con-
ductivity of the studied ILs follows the order: [C1py]
[MeSO4] > [C4mim][MeSO4] > [P4441][MeSO4] ≈ [P4441]
Cl > [P4441][NTf2]. As showed in Fig. 4a, among the ILs sharing the
same cation, [P4441]+, the anions Cl− and [MeSO4]− have no influence
on the conductivities of the resulting IL with values of 0.159 and
0.160 W m−1 K−1, respectively at 293.15 K. However, the anion
[NTf2]− strongly decreases the thermal conductivity to
0.126 W m−1 K−1 at the same temperature. Cation change in Fig. 4b
shows a great influence on the anion [MeSO4]−. At the temperature of
293.15 K, [C1py][MeSO4] has a thermal conductivity of
0.205 W m−1 K−1, the cation [C4mim]+ lowers this property to 0.177
and the cation [P4441]+ to 0.159 W m−1 K−1.
In a survey of the literature, only one paper was found that allows
the comparison of the thermal conductivity data for [P4441][MeSO4]
[5]. The MRD between the data measured in this work and those re-
ported by Ferreira et al. [5] is 1.4%. This small difference, showed in
Fig. 4, can be due to the presence of water in the IL, which can modify
the thermal conductivity. A few authors have studied how the water
content modifies the thermal conductivity in the ILs [3], and they ob-
served an increase of the thermal conductivity with the water content,
Ferreira et al. [5] report 52 ppm and in this work the water content of
the IL used is 175 ppm.
Thermal conductivities were correlated with temperature using Eq.
(2). The correlation parameters and standard deviations are summar-
ized in Table 4.
4.5. Dynamic viscosity
As shown in Fig. S7 in supporting information the dynamic viscosity
remains constant for a shear rate range between 0,1 and 100 s−1,
stating the Newtonian behavior of the ILs studied. Viscosity data
(Table 3) for the ILs at temperature ranging from 283.15 to 343.15 K

86
E. Rodil et al. Thermochimica Acta 664 (2018) 81–90

Fig. 3. Experimental electrical conductivity as a function of temperature: (–)

Fig. 2. Experimental refractive index as a function of temperature: (–) Eq. (2). Eq. (3).
(a) Same cation: (a) Same cation:
[P4 4 4 1]Cl: (●) this work, [P4 4 4 1]Cl: (●) this work,
[P4 4 4 1] [MeSO4]: (▼) this work, (×) ref. [11], (▽) ref. [12], [P4 4 4 1][NTf2]: (○) this work,
[P4 4 4 1][NTf2]: (○) this work, [P4 4 4 1][MeSO4]: (▼) this work,
(b) Same anion: (b) Same anion:
[P4 4 4 1][MeSO4]: (▼) this work, (×) ref. [11], (▽) ref. [12] [P4 4 4 1][MeSO4]: (▼) this work,
[C4mim][MeSO4]: (□) this work, (●) ref. [14], (▲) ref. [22], (△) ref. [15], [C4mim][MeSO4]: (□) this work, (*) ref. [28], (−) ref. [29], (△) ref. [33],
[C1py][MeSO4]: (♢) this work, (○) ref. [23]. [C1py][MeSO4]: (♢) this work.

are showed in Fig. 5. One of the limitations of ILs is their high viscosity, Table 5
Parameters A, B, C and standard deviations, SD, of the fittings with Eq. (3) for
which will adversely affect most of their possible industrial applica-
electrical conductivity, σ, and dynamic viscosity, η.
tions. As is usually observed, the viscosity decreases considerably with
increasing temperature. The viscosity in the ILs studied increases in the Physical Property A B C SD
following order: [C1py][MeSO4] < [C4mim][MeSO4] < [P4441]
[P4441]Cl
[NTf2] < [P4441][MeSO4] < [P4441]Cl. The differences between the σ (μS cm−1) 16.535 −1915.2 110.62 3
dynamic viscosities for these ILs can be explained by means of the ion η (mPa s) −4.0046 1591.9 175.37 13
charge. The delocalization of the charge on the anion seems to results in [P4441][NTf2]
lower viscosities and the longer alkyl chains of the cation at higher σ (μS cm−1) 15.273 −1410.3 140.82 25
viscosities [11–13,14,16,22,23]. These tendencies are showed in η (mPa s) −4.7175 1601.0 139.45 2
Fig. 5a, where ILs with the same cation [P4441]+ and the higher anions [P4441][MeSO4]
size present lower viscosity values. The fact that the Cl− is closer to the σ (μS cm−1) 18.522 −2726.6 101.44 14
phosphorus atom also gives more free space to the alkyl chains ex- η (mPa s) −2.1375 1029.3 183.71 2
plaining why the [P4441]Cl shows the largest viscosity [C1py][MeSO4]
([P4441]Cl > [P4441][MeSO4] > [P4441][NTf2]). In the case of [C1py] σ (μS cm−1) 13.726 −621.36 185.93 83
η (mPa s) −3.5506 1334.1 132.24 1
[MeSO4] (Fig. 5b), the smaller cation volume and larger average dis-
tance between cation and anion could explain the low viscosity for this [C4mim][MeSO4]
IL. σ (μS cm−1) 17.082 −1902.7 105.14 10
η (mPa s) −3.5028 1334.9 146.60 2
For the compounds studied, few viscosity data are available in the
literature [11–14,16,22,23,28,29]. MRD of 32%, 6%, 2%, 4%, and 8%;
with maximum deviations of 59, 36, 7, 24, and 10%, were obtained
when our viscosity data for [C4mim][MeSO4] were compared with

87
E. Rodil et al.
Fig. 4. Experimental thermal conductivities as a function of temperature: (–)
Eq. (2).
(a) Same cation:
[P4 4 4 1]Cl: (●) this work, (×) ref. [5],
[P4 4 4 1][NTf2]: (○) this work,
[P4 4 4 1][MeSO4]: (▼) this work,
(b) Same anion:
[P4 4 4 1][MeSO4]: (▼) this work, (×) ref. [5],
[C4mim][MeSO4]: (□) this work,
[C1py][MeSO4]: (♢) this work.
those published by Beigi et al. [14], Fernández et al. [16], Pereiro et al.
[22], Mbondo Tsamba et al. [28], and Pandit et al. [29], respectively.
There is a high discrepancy with data published by Beigi et al. [14].
However all the other published data are in good agreement with ours
and Beigi’s data are clearly smaller, for the whole temperature range,
probably as a result of high water content of their IL (0.194 mass%). A
MRD of 2% and 5% were found comparing our data with those pub-
lished by Adomavá et al. [13] for [P4441] Cl (with a maximum deviation
of 9%) and by Gómez et al. [23] for [C1py][MeSO4] (with a maximum
deviation of 20%), respectively. It is noticeable the large deviations
found for [P4441][MeSO4]. Our deviations from data reported by Deive
et al. [12] are 20% above (with a maximum deviation of 50%) while
deviations from Bhattacharjee et al. [11] are 20% under (with a max-
imum deviation of 73%). Both authors studied samples from the same
supplier while in this work the IL has been synthetized and purified by
ourselves. Deive et al. [12] do not report water content and Bhatta-
charjee et al. [11] have reported a water content of 0.0783 in mass
fraction. It is known that viscosity data are very sensitive to the water
content and impurities. Moreover, low water contents drastically affect
the viscosity of this IL due to its high value. As far as we know, no
viscosity data are available for [P4441][NTf2].
A Vogel-Fulcher-Tammann type function (Eq. (3)) was used to
88
Thermochimica Acta 664 (2018) 81–90
Fig. 5. Experimental dynamic viscosity as a function of temperature: (–) Eq. (3).
(a) Same cation:
[P4 4 4 1]Cl: (●) this work, (△) ref. [13],
[P4 4 4 1][NTf2]: (○) this work,
[P4 4 4 1][MeSO4]: (▼) this work, (×) ref. [11], (▽) ref. [12].
(b) Same anion:
[P4 4 4 1][MeSO4]: (▼) this work, (×) ref. [11], (▽) ref. [12],
[C4mim][MeSO4]: (□) this work, (●) ref. [14], (♦) ref. [16], (▲) ref. [22], (*)
ref. [28], (−) ref. [29],
[C1py][MeSO4]: (♢) this work, (○) ref. [23].
correlate dynamic viscosity of the ILs with temperature. Results are
shown in Table 5.
Walden plots are shown in the supporting information, Fig. S6, for
all the ILs studied. Plots are below the reference line (KCl, 0.01 M) with
slopes (α) between 0.88 and 0.98. This behavior has been found for
other ILs that follow a fractional Walden Rule [34,35].
4.6. Thermal stability
For the five ILs studied in this work, typical one-step decomposition
thermograms were obtained. These thermograms show the sample
weight loss as the temperature increases at a fixed heating rate.
Thermal stability is usually analyzed in terms of the onset decomposi-
tion temperature, Td,onset, which is the intersection of the baseline
weight and the tangent of the weight vs. temperature curve as the de-
composition occurs. However, in this work, the 5% onset decomposi-
tion temperatures, Td,5%onset, taken as the onset temperature corre-
sponding to a weight loss of 5% of the original sample mass by
decomposition, are reported. From the values reported in Table 6, the
hydrocarbon chains in the cations seem to play an important role in the
thermal stability of the ILs. The ILs having the same cation show a very
similar value whereas larger variations are found in those ILs having the
E. Rodil et al.
Table 6
Thermal stability at 0.1 MPa for the ILs studied.
IL TdOnset (K) Td5%Onset (K)
[P4441]Cl 633 585
[P4441][NTf2] 623 597
[P4441][MeSO4] 609 583
[C1py][MeSO4] 576 448
[C4mim][MeSO4] 594 522
Standard uncertainties: u(P) = 5 kPa; u(T) = 1 K.
same anion following the trend: [P4441][MeSO4] > [C4mim]
[MeSO4] > [C1py][MeSO4]. The [P4441][NTf2] shows the higher
thermal stability with a Td,5%onset value of 597 K whereas the [C1py]
[MeSO4] shows the lowest value of 448 K. Tsunashima and Sugiya [36]
have reported a decomposition temperature of 652 K at a 10% weigh
loss which is above the Td,onset of 623 K and the Td,5%onset of 597 K re-
ported in the present work. Beigi et al. [14] report a Td for [C4mim]
[MeSO4] of 310 °C, value which is in good agreement with the Td,onset
reported in this work. Although the Td,5%onset could be considered as a
more conservative value than the Td,onset, also constitutes an over-
estimate of the maximum temperature that these ILs could be used in
practical applications [37].
5. Conclusions
Density, refractive index, electrical conductivity, thermal con-
ductivity and dynamic viscosity were measured in the temperature
range from (283.15 to 343.15) K and at atmospheric pressure for the
ILs: [P4441]Cl, [P4441][NTf2], [P4441][MeSO4], [C1py][MeSO4], and
[C4mim][MeSO4]. These five ILs have been found to be thermally stable
beyond 400 K. Over this temperature range, density decreased linearly
and refractive index slightly decreased, also linearly, for the five ILs.
Increasing temperature increases ion volumes resulting in a density
decrease. Increasing temperature increases ion mobility, critically af-
fecting the electrical conductivity which therefore increases with tem-
perature. This ion mobility accounts for the dramatic decrease of the
dynamic viscosity as temperature rises. The measurements for re-
fractive index and thermal conductivity over the temperature range
studied are so close for all the ILs studied that is difficult to establish
any relationship between these properties and ion volume, ion mobility,
or ion changes. Steric effects affecting ion size explain the density dif-
ferences between the ILs within each group. Distinguishing between the
ILs with the same cation, [P4441]+, and those with the same anion,
[MeSO4]−, the steric effects along with different average distances
between cation and anion account for the electrical conductivity and
dynamic viscosity differences within each group. Temperature depen-
dence of density, refractive index, and thermal conductivity were well
correlated with linear functions, with low deviations, while Vogel-
Fulcher-Tammann type functions better fit the temperature dependence
for electrical conductivity and dynamic viscosity.
Acknowledgements
The authors acknowledge the Ministry of Economy and
Competitiveness of Spain for financial support through project
CTQ2015-68496-P (including European Regional Development Fund
advanced funding).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.tca.2018.04.007.
89
Thermochimica Acta 664 (2018) 81–90
References
[1] M. Freemantle, An introduction to ionic liquids, R. Soc. Chem. Camb. (2010).
[2] M.E.V. Valkenburg, R.L. Vaughn, M. Williams, J.S. Wilkes, Thermochemistry of ionic
liquid heat-transfer fluids, Thermochim. Acta 425 (2005) 181–188.
[3] R. Ge, C. Hardacre, P. Nancarrow, D.W. Rooney, Thermal conductivities of ionic liquids
over the temperature range from 293 K to 353 K, J, Chem. Eng. Data 52 (2007)
1819–1823.
[4] A.F. Ferreira, P.N. Simoes, A.G.M. Ferreira, Quaternary phosphonium-based ionic liquids:
thermal stability and heat capacity of the liquid phase, J. Chem. Thermodyn. 45 (2012)
16–27.
[5] A.G.M. Ferreira, P.N. Simoes, A.F. Ferreira, M.A. Fonseca, M.S.A. Oliveira, A.S.M. Trino,
Transport and thermal properties of quaternary phosphonium ionic liquids and
IoNanofluids, J. Chem. Thermodyn. 64 (2013) 80–92.
[6] J.M.P. França, C.A. Nieto de Castro, V.M.B. Nunes, M.L.S. Matos Lopes, The influence of
thermophysical properties of ionic liquids in chemical process design, J. Chem. Eng. Data
54 (9) (2009) 2569–2575.
[7] A.P. Froba, M.H. Rausch, K. Krzeminski, D. Assenbaum, P. Wasserscheid, A. Leipertz,
Thermal conductivity of ionic liquids: measurement and prediction, Int. J. Thermophys.
31 (2010) 2059–2077.
[8] J.M.P. França, C.A. Nieto de Castro, V.M.B. Nunes, M.L.S. Matos Lopes, The influence of
thermophysical properties of ionic liquids in chemical process design, J. Chem. Eng. Data
54 (9) (2009) 2569–2575.
[9] V.V. Wadekar, Ionic liquids as heat transfer fluids –an assessment using industrial ex-
changer geometries, Appl. Therm. Eng. 111 (2017) 1581–1587.
[10] VDI Heat Atlas, in: VDI-GC (Ed.), 2nd ed., Springer Düsseldorf, 2010.
[11] A. Bhattacharjee, J.A. Lopes-da-Silva, M.G. Freire, J.A.P. Coutinho, P.J. Carvalho,
Thermophysical properties of phosphonium-based ionic liquids, Fluid Phase Equilibr. 400
(2015) 103–113.
[12] F.J. Deive, M.A. Rivas, A. Rodríguez, Study of thermodynamic and transport properties of
phosphonium-based ionic liquids, J. Chem. Thermodyn. 62 (2013) 98–103.
[13] G. Adamová, R.L. Gardas, M. Nieuwenhuyzen, A.V. Puga, L.P.N. Rebelo, A.J. Robertson,
K.R. Seddon, Alkyltributylphosphonium chloride ionic liquids: synthesis, physicochem-
ical properties and crystal structure, Dalton T. 41 (2012) 8316–8332.
[14] A.A.M. Beigi, M. Abdouss, M. Yousefi, S.M. Pourmortazavi, A. Vahid, Investigation on
physical and electrochemical properties of three imidazolium based ionic liquids (1-
hexyl-3-methylimidazolium tetrafluoroborate, 1-ethyl-3-methylimidazolium bis(tri-
fluoromethylsulfonyl) imide and 1-butyl-3-methylimidazolium methylsulfate), J. Mol.
Liq. 177 (2013) 361–368.
[15] N. Soriano, B.T. Doma Jr., M.-H. Li, Measurements of the density and refractive index for
1-n-butyl-3-methylimidazolium-based ionic liquids, J. Chem. Thermodyn. 41 (3) (2009)
301–307.
[16] A. Fernández, J. García, J.S. Torrecilla, M. Oliet, F. Rodríguez, Volumetric, transport and
surface properties of [bmim] [MeSO4] and [emim] [MeSO4] ionic liquids as a function of
temperature, J. Chem. Eng. Data 53 (2008) 1518–1522.
[17] G. García-Miaja, J. Troncoso, L. Romaní, Excess molar properties for binary systems of
alkylimidaxolium-based ionic liquids + nitromethane. Experimental results and ERAS-
model calculations, J. Chem. Thermodyn. 41 (2009) 334–341.
[18] Y. Zhang, T. Li, Z. Wu, P. Yu, Yunbai Luo, Synthesis and thermophysical properties of
imidazolate-based ionic liquids: influences of different cations and anions, J. Chem.
Thermodyn. 74 (2014) 209–215.
[19] S. Lago, H. Rodríguez, A. Soto, A. Arce, Alkylpyridinium alkylsulfate ionic liquids as
solvents for the deterpenation of citrus essential oil, Sep. Sci. Technol. 47 (2012)
292–299.
[20] O.K. Bates, Binary mixtures of water and glycerol – thermal conductivity of liquids, Ind.
Eng. Chem. 28 (1936) 494–498.
[21] D.K.H. Briggs, Thermal conductivity of liquids, Ind. Eng. Chem. 49 (1957) 418–421.
[22] A.B. Pereiro, P. Verdía, E. Tojo, A. Rodríguez, Physical properties of 1-Butyl-3-methyli-
midazolium methyl sulfate as a function of temperature, J. Chem. Eng. Data 52 (2007)
377–380.
[23] E. Gómez, N. Calvar, A. Domínguez, E.A. Macedo, Synthesis and temperature dependence
of physical properties of four pyridinium-based ionic liquids: influence of the size of the
cation, J. Chem. Thermodyn. 42 (2010) 1324–1329.
[24] E.J. González, A. Domínguez, E.A. Macedo, Physical and excess properties of eight binary
mixtures containing water and ionic liquids, J. Chem. Eng. Data 57 (2012) 2165–2176.
[25] L. Galán Sánchez, J. Ribé Espel, F. Onink, G. Wytze Meindersma, A.B. de Haan, Density,
viscosity, and surface tension of synthesis grade imidazolium, pyridinium, and pyrroli-
dinium based room temperature ionic liquids, J. Chem. Eng. Data 54 (2009) 2803–2812.
[26] J.-M. Andanson, M. Traikia, P. Husson, Ionic association and interactions in aqueous
methylsulfate alkyl-imidazolium-based ionic liquids, J. Chem. Thermodyn. 77 (2014)
214–221.
[27] J.A. Riddick, B.W. Bunger, T. Sakano, Organic Solvents, 4th ed., John Willey, New York,
1986.
[28] B.E. Mbondo Tsamba, S. Sarraute, M. Traikia, P. Husson, Transport properties and ionic
association in pure imidazolium-based ionic liquids as a function of temperature, J.
Chem. Eng. Data 59 (2014) 1747–1754.
[29] S.A. Pandit, M.A. Rather, S.A. Bhat, G.M. Rather, M.A. Bhat, Influence of the anion on the
equilibrium and transport properties of 1-Butyl-3-methylimidazolium based room tem-
perature ionic liquids, J. Solut. Chem. 45 (2016) 1641–1658.
[30] A. Tagiuri, K.Z. Sumon, A. Henni, Solubility of carbon dioxide in three [Tf2N] ionic li-
quids, Fluid Phase Equillibr. 380 (2014) 39–47.
[31] Z. Gu, J.F. Brennecke, Volume expansivities and isothermal compressibilities of imida-
zolium and pyridinium-based ionic liquids, J. Chem. Eng. Data 47 (2002) 339–345.
[32] A.K. Ziyada, C.D. Wilfred, Effect of temperature and anion on densities, viscosities, and
refractive indices of 1-Octyl-3-propanenitrile imidazolium-Based ionic liquids, J. Chem.
Eng. Data 59 (2014) 1385–1390.
[33] Y.-H. Yu, A.N. Soriano, M.-H. Li, Heat capacities and electrical conductivities of 1-n-butyl-
E. Rodil et al.
3-methylimidazolium-based ionic liquids, Thermochim. Acta 482 (2009) 42–48.
[34] W. Xu, E.I. Cooper, C.A. Angell, Ionic liquids: ion mobilities glass temperatures, and
fragilities, J. Phys. Chem. B 107 (2003) 6170–6178.
[35] C. Schreiner, S. Zugmann, R. Hartl, H.J. Gores, Fractional walden rule for ionic liquids:
examples from recent measurements and a critique of the so-Called ideal KCl line for the
walden plot, J. Chem. Eng. Data 55 (2010) 1784–1788.
90
Thermochimica Acta 664 (2018) 81–90
[36] K. Tsunashima, M. Sugiya, Physical and electrochemical properties of room temperature
ionic liquids based on quaternary phosphonium cations, Electrochemistry 75 (2007)
734–736.
[37] M.T. Clough, K. Geyer, P.A. Hunt, J. Mertes, T. Welton, Thermal decomposition of car-
boxylate ionic liquids: trends and mechanisms, Phys. Chem. Chem. Phys. 15 (2013)
20480–20495.