Glycerol as a source of designer solvents: Physicochemical properties of low melting mixtures containing glycerol ethers and ammonium salts

Article abstract of DOI:10.1039/c7cp04987k

In this work we report the preparation of mixtures of several alkyl glyceryl ethers, as hydrogen bond donor compounds, with two ammonium salts, choline chloride and N,N,N-triethyl-2,3-dihydroxypropan-1-aminium chloride. The stability of the mixtures at different molar ratios and temperatures has been evaluated in order to determine the formation of low melting mixtures. Liquid and stable mixtures have been characterized and their physico-chemical properties such as density, viscosity, refractive index, conductivity and surface tension have been measured in the temperature range of 293.15 K to 343.15 K. Comparison of the mixtures prepared herein with the ones containing glycerol and choline chloride evidences the possibility of tuning the physico-chemical properties by changing the substitution pattern in the hydrogen bond donor compound or in the ammonium salt, thus broadening the scope of application of these mixtures.

Full text of DOI:10.1039/c7cp04987k

PCCP
View Article Online
View Journal
PAPER
Glycerol as a source of designer solvents:
physicochemical properties of low melting
mixtures containing glycerol ethers and
ammonium salts†
Cite this:DOI: 10.1039/c7cp04987k
ab
c
b
Alejandro Leal-Duaso,
Elisabet Pires
Pascual Perez,
Jose A. Mayoral,
´
´
b
a
and Jose I. Garcıa
*
´
´
In this work we report the preparation of mixtures of several alkyl glyceryl ethers, as hydrogen bond
donor compounds, with two ammonium salts, choline chloride and N,N,N-triethyl-2,3-dihydroxypropan-1-
aminium chloride. The stability of the mixtures at different molar ratios and temperatures has been evaluated
in order to determine the formation of low melting mixtures. Liquid and stable mixtures have been
characterized and their physico-chemical properties such as density, viscosity, refractive index, conductivity
and surface tension have been measured in the temperature range of 293.15 K to 343.15 K. Comparison of
the mixtures prepared herein with the ones containing glycerol and choline chloride evidences the
possibility of tuning the physico-chemical properties by changing the substitution pattern in the hydrogen
bond donor compound or in the ammonium salt, thus broadening the scope of application of these
mixtures.
Received 24th July 2017,
Accepted 27th September 2017
DOI: 10.1039/c7cp04987k
rsc.li/pccp
Since the year 1998, great attention has been devoted to these
Introduction
kinds of solvents, as illustrated by the increasing number of
3
The vast majority of chemical industry processes, with perhaps publications appearing in this area. Nevertheless, some draw-
the noticeable exception of the petrochemical industry, rely on backs have also been reported about some IL, for instance their
the use of conventional organic solvents. Although there are high price, their sometimes difficult synthesis/purification, or
4
,5
undisputable advantages in the use of these kinds of solvents, their significant (eco)toxicity in many cases. Very recently, Low
not least their easy availability and low price, there are also Melting Mixtures (LMM), very often named as Deep Eutectic
important drawbacks such as their (eco)toxicity, volatility or Solvents (DES) in the literature, have been suggested as greener
6
–9
flammability. For this reason, developing cost-effective, harmless analogues of ionic liquids. This claim is based on their easy
and environmentally benign alternative solvent systems is a preparation, usually just requiring the mixing of two easily
research area of great interest. The desirable features of the so- available components, their renewable origin in most cases,
1
0
called green solvents are, among others: renewable origin, easy and the low toxicity of their components.
availability and biodegradability, and low (eco)toxicity, volatility, Most of these so-called DES are formed by a hydrogen bond
flammability and price. About twenty years ago ionic liquids (IL) donor compound (HBD) and an ammonium salt. The formation
were proposed as a green alternative to conventional hazardous of hydrogen bonds between the HBD component and the anion of
volatile organic solvents due to their low flammability, essentially the salt disrupts the crystallinity of the latter and is responsible for
1
,2
null volatility, and very special physicochemical properties.
the decrease in the melting point of the mixture relative to their
individual components. The stronger the interaction between
both components, the larger will be the decrease with regard to
the ideal behaviour of the mixture.
Unfortunately, the terminology used in the literature is not
always clear when talking of DES, and it is really complex to
find an accurate denomination for these kinds of mixtures. In
many publications the mixtures described are referred to as DES,
but a true eutectic point corresponds to the single composition
of the mixture showing the minimum melting point, and this
circumstance is seldom checked. Other publications circumvent
a
Instituto de S ´ı ntesis Qu ´ı mica y Cat a´ lisis Homog ´e nea (ISQCH),
CSIC-University of Zaragoza, Calle Pedro Cerbuna, 12, E-50009 Zaragoza, Spain.
E-mail: ji.garcia@csic.es
b
c
Department of Organic Chemistry, Faculty of Science, University of Zaragoza,
Calle Pedro Cerbuna, 12, E-50009 Zaragoza, Spain
Department of Physical Chemistry, Faculty of Science, University of Zaragoza,
Calle Pedro Cerbuna, 12, E-50009 Zaragoza, Spain
†
Electronic supplementary information (ESI) available: Complete product charac-
terization, and tables with the measured physicochemical properties. See DOI:
0.1039/c7cp04987k
1
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
11
this situation by using the fuzzier term LMM, which does not Experimental
imply that the composition corresponds to an eutectic mixture
but still assumes a definite melting point for the mixture. However,
Materials and methods
Glycidol, 3-chloropropane-1,2-diol and triethylamine were pur-
chased from Sigma-Aldrich. MeOH, EtOH, PrOH, potassium
or sodium hydroxide were purchased from Scharlab. BuOH,
in many cases, and especially in the case of glycerol and its
derivatives, the mixtures lead to glasses at low temperature,
so it should be more appropriate to call them Low Freezing
Point Mixtures (LFPM), a term almost unused in the literature,
however.
i
2,2,2-trifluoroethanol, phenol, 2-methoxyphenol and choline
chloride were purchased from Alfa Aesar. ChCl was recrystallized
from absolute ethanol, filtered and dried at 323.15 K under
vacuum prior to use. The alcohols were dried and distilled over
calcium hydride previous to use.
Description of mixture preparation, properties and applica-
1
2
tions has recently been reviewed by Smith et al. Some other
specialized publications have also appeared dealing with the use of
1
13
19
7
,13,14
15,16
H-, C- and F-NMR spectra (DMSO-d , d ppm, J Hz) were
6
DES and LMM in organic synthesis,
enzymatic catalysis,
in nanotechnology,
17,18
19
obtained using a Bruker AV-400 instrument with TMS as
standard. HRMS spectra were obtained using a Bruker
MicroTof-Q spectrometer with electrospray ionization. Solvent
boiling points were determined using differential scanning
calorimetric (DSC) analysis in a DSC-Q20, TA Instruments,
calibrated with indium, using micropore aluminium pans, in
a range of 298.15–673.15 K, at atmospheric pressure. Gas
chromatography was carried out in a Hewlett Packard 7890
Series II chromatograph using a column of phenyl silicone
and as polymerization media.
Concerning LMM composition, choline chloride is almost
universally used as the ammonium salt due to its easy acces-
sibility (it is produced on the MTon scale as a chicken feed
additive) and low price (ca. 2 h per kilogram), but the use of
other compounds, such as tetraalkylammonium salts, has also
2
0–22
been described.
It is worth mentioning that the renewable
origin of choline chloride is often claimed as a ‘‘green creden-
tial’’. However, at present most of this chemical is industrially
produced from ethylene oxide, and hence from fossil sources.
Polyols, urea and its derivatives, carboxylic acids or amides
have been described as the HBD component for the formation
of DES or LMM. Glycerol is among the most used compounds
in this category. Some authors have studied the physical
properties of glycerol-containing LMM with different ammo-
5
.5% (Zebron ZB-5HT Inferno 30 m ꢀ 0.25 mm ꢀ 0.25 mm) and
helium as carrier gas, equipped with a flame ionization
detector (FID).
Synthesis of glycerol monoethers, [R.0.0]
The appropriate amount of alcohol (1500% mol with respect to
glycidol) and potassium hydroxide (20% mol with respect to
glycidol) were placed into a round-bottomed flask. The reaction
was stirred and heated at 338 K under argon until total
dissolution of the catalyst. Then, glycidol (1 mol) was added
2
3–25
nium salts.
Furthermore, Abbott et al. have described the
use of glycerol eutectics as sustainable solvent systems in
8
glycerol esterification processes.
On the other hand, glycerol derivatives have attracted great
26–28 dropwise. The reaction was monitored at different times by
attention as renewable green solvents in different applications.
In particular, glycerol ethers have been shown to be promising
candidates for solvent substitution, given that their physicochemical
properties can be modulated through the number, size and nature
of their alkyl groups, leading to a wide range of polarities, among
extracting samples that were neutralized with HCl 0.3 M prior
to injection in GC. After the total consumption of glycidol, the
reaction was quenched with HCl 0.3 M and filtered off. After the
reaction work-up, the corresponding glycerol monoether was
purified by vacuum distillation, or by column chromatography
in the case of aromatic monoethers.
2
9–32
other properties.
Some preliminary toxicity studies indicate
that most of them display low acute toxicity for widely used
3
3,34
biomarkers.
Synthesis of N,N,N-triethyl-2,3-dihydroxypropan-1-aminium
chloride, [N.0.0]Cl
In this work, we present a systematic study of the physico-
chemical properties of mixtures composed of alkyl glyceryl
3
0,35
ethers, prepared from glycidol and epichlorohydrin,
as 42.4 mL of 3-chloropropane-1,2-diol (0.51 mol) were added to a
HBD components and two different ammonium salts: choline solution of triethylamine (105 mL, 0.76 mol) in methanol
chloride and N,N,N-triethyl-2,3-dihydroxypropan-1-aminium (80 mL) in a round bottomed flask. The reaction was stirred
chloride ([N.0.0]Cl). In the latter case both components can and heated at 338 K under an argon atmosphere for 12 hours.
be obtained using glycerol as a platform molecule. It is worth Then, triethylamine was eliminated by vacuum distillation and
mentioning that during the revision of this paper one of the the resulting oil was washed with ether and acetone till the
reviewers pointed out that glycidol is industrially obtained by appearance of a white precipitate. This solid was filtered off and
epoxidation of allyl alcohol, i.e., from a fossil source. However, dried under vacuum.
glycidol can be also obtained from glycerol, following different
3
6–40
Mixture preparation
procedures.
glycidol production could easily switch to glycerol-based proce- The mixtures were formed by stirring the two components, the
dures, as has already happened in the case of epichlorohydrin. glycerol ether and the ammonium salt (ChCl or [N.0.0]Cl),
Therefore, in a low-carbon economy scenario,
The final aim of the work is the fine tuning of the physico- under argon at 343 K for 10 hours approximately. In the correct
chemical properties of the mixture by changing the length and nature molar proportion, a homogeneous colourless liquid was
of the alkyl chain of the glycerol ether and the ammonium salt.
obtained, its composition being checked by NMR.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
Physicochemical property measurements
The viscometers were placed in a thermostated bath using a
suitable support. The relative uncertainty in kinematic viscosity
data is around 0.6%. Dynamic viscosity was calculated according
the equation:
Density and sound speed. Both properties were simulta-
neously determined by using an Anton Paar DSA 5000 M which
is equipped with density and sound speed cells. Both chambers
are temperature controlled by an internal Peltier thermostat
at ꢁ0.001 K. After the appropriate calibration, the sample is
Z = nr
introduced into a U-shaped glass tube and the density is where Z is the dynamic viscosity, r the density and n the
obtained from its frequency of vibration with an uncertainty kinematic viscosity as defined above.
ꢂ3
ꢂ3
of ꢁ5 ꢀ 10 kg m . Correction due to kinematic viscosity of
Conductivity. The ionic conductivity measurements were
the liquid is made automatically. The sound speed of the carried out with a Crison GLP31 conductimeter. The apparatus
sample is calculated by the determination of sound waves was calibrated by using two KCl reference solutions of 1413 mS cm
ꢂ1
ꢂ1
ꢂ1
through the cell with an uncertainty of ꢁ0.1 m s
.
and 12.88 mS cm . The liquid and the conductivity cell were set
Static dielectric permittivity and dipole moment. Static in a glass container equipped with a thermostated jacket. The
dielectric permittivity, or dielectric constant, data were calcu- experimental values have a relative uncertainty better than 0.5%.
lated from capacitance values measured with a WTW (DM01)
In the process of measurement of the above-mentioned
dipolmeter (Kahlsico) at 2 MHz. The apparatus works using the properties: dielectric constant, refractive index, conductivity
superposition method by which the oscillations of two high and kinematic viscosity, temperature were controlled by means
frequency oscillators are brought to superposition in a mixing of a Haake F6 thermostat with stability better than 0.01 K.
section. Two cylindrical condenser type sample holding cells
Surface tension. A drop volume tensiometer Lauda TVT-2,
were used in order to cover the entire range of experimental which is able to measure the volume of a drop with high precision,
data. Cell DFL2 covered the dielectric constant range from 1 to 7 was used to obtain the surface tension of sample liquids at a fixed
and cell MFL2/MS from 6 to 23. Both cells were provided with a temperature with a relative uncertainty not worse than 0.5%.
metal jacket and held at constant temperature by a water A Lauda E-200 thermostat allowed us to maintain the sample
circulating external bath and they were previously calibrated temperature within ꢁ0.01 K.
with high purity and low water content reference liquids. The
For all liquids and properties measured, the temperature
relative uncertainty in dielectric constant values is estimated to was registered by means of a digital thermometer ASL with a Pt
be better than 0.05%. The dielectric constant of liquid [3F.0.0] sensor with a resolution of 0.001 degrees and an uncertainty of
was only measured at 298.15 K by the time domain reflectometry ꢁ0.01 K.
(TDR) technique using the first reflection method because its
high dielectric constant was out of the range of the MFL2/MS cell.
From the dielectric constant, refractive index and density Results and discussion
2
1/2
experimental data, an average of dipole moment hm i
for
4
1
In the last few years we have focussed our attention on the
every pure liquid was calculated through the Onsager’s equation:
synthesis, characterization, use and toxicity study of alkyl
À
ÁÀ
Á
2
2
D
30,33,34,42–44
ꢀ
ꢁ
9kT M e ꢂ n
2e þ n
glyceryl ethers.
These promising solvents have some
2
D
m
¼ 4
ꢃ
2
pNA r
2
advantages like being relatively inert from a chemical point of
view (different from other glycerol derivatives, such as esters,
e ꢃ ðnD þ 2Þ
where k = Boltzmann’s constant, T = temperature in K, M = molar ketals or carbonate), non-volatile and less viscous than glycerol.
mass, N
A
= Avogadro’s number, r = density, n
D
= refractive index, Moreover, their physicochemical and toxicological properties
e = dielectric permittivity.
can be tuned by simply changing the number, size and nature
Refractive index. An Abbe refractometer ZUZI 315 with of the alkyl chains. As evidence of the stability of these glycerol
s
temperature controlled and a sodium lamp as light source was derivatives, semi quantitative analysis of peroxides (Quantofix )
used for refractive index measurements, with an uncertainty of in samples of solvents stored in our laboratory showed less than
ꢂ1
ꢂ1
ꢁ
0.0001.
1 mg L of H O in two years old ether samples and less than
2
2
Viscosities. A set of three Ubbelohde viscometers with 3 mg L in solvents stored for 12 years.
suspending ball level (Schott) and relatively long flow time were
The evidence that glycerol can be used as the HBD compo-
used in order to cover the entire range of experimental values. In nent in LMM23–25 prompted us to study the possibility of using
each experiment the flow time was measured at least three times the alkyl glyceryl ethers as components of LMM together with
with differences not greater than ꢁ0.3 s. The viscometer models different ammonium salts. For this study we have chosen
were: 501 11/Ia, 501 23/IIc and 501 30/III, whose constants k were mono- and dialkyl glyceryl ethers with alkyl chains of increasing
2
ꢂ2
0
.04957, 0.3157 and 1.000 mm s , respectively. The kinematic length, alkyl chains containing fluorine, and simple aryl substi-
viscosities were calculated from the equation:
v = k(t ꢂ DHC
where n is the kinematic viscosity, k the viscometer constant,
tuents (Fig. 1), in order to study the influence of these structural
factors both on the mixture formation and on the physico-
chemical properties of the stable mixtures.
)
As ammonium salts we chose choline chloride (ChCl) (Fig. 1),
t the average flow time and DHC the kinetic energy correction. in order to compare the properties of our mixtures to the ones
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
homogeneous colourless liquid mixtures at working temperature
at all the proportions tested with both ammonium salts. However,
after cooling to room temperature the homogeneity in the liquid
mixtures was kept only in those mixtures having a 2 : 1 HBD :
ammonium salt molar ratio. In the case of [N.0.0]Cl, some stable
liquid mixtures were also obtained at a 2.5 : 1 molar ratio. As a
consequence of these results we decided to continue the study
only with those components and molar ratios leading to homo-
geneous and stable liquid phases.
Using the abovementioned seven [R.0.0] ethers and the two
ammonium salts, up to 14 new LMM could be formed in the
molar proportion 2 : 1. However, the range of stability of these
mixtures was not the same. Some mixtures, including [2.0.0],
Fig. 1 Alkyl glyceryl ethers and ammonium salts used in this study.
[
4.0.0], [3i.0.0] and [G.0.0] displayed a ‘‘critical stability
previously described. We also included in the study an ammo-
nium salt that can be prepared using glycerol as a platform
molecule, namely the N,N,N-triethyl-2,3-dihydroxypropan-1-
aminium chloride (henceforth [N.0.0]Cl), as both components
of the mixture may come from the same renewable source.
Unfortunately, the toxicity of [N00]Cl has not yet been evaluated
experimentally. However, its similarity to glycerol and choline
chloride, both almost harmless, prompt us to reasonably
anticipate a very low toxicity for this compound.
temperature’’ below which some solid phase appears. This
temperature was characteristic of the HBD component used,
and increases with the R chain length. In the case of ChCl
mixtures, their values were 306, 325, 333, 347 and 313 K for
[1.0.0], [2.0.0], [3i.0.0], [4.0.0], and [G.0.0], respectively. On the
other hand, [N.0.0]Cl mixtures are more stable, and the ‘‘critical
temperatures’’ were lower, mixtures with [1.0.0], [2.0.0], [3F.0.0]
and [Ph.0.0] being totally stable in the whole range of tempera-
ture (from 298 to 343 K).
The homogeneous mixtures were cooled below 248 K, and
the formation of different glass phases was observed. In no case
liquid-crystal transitions were observed. For this reason, only in
the case of mixing the solid [Ph.0.0] (m.p. = 327–330 K) or
In our case, this compound was easily obtained from the
reaction of 3-chloropropane-1,2-diol with triethylamine (Scheme 1).
Preparation and stability of the mixtures
First, a complete study of the formation of the mixtures and
their stability was carried out. HBD and ammonium salt
components were mixed at 343 K at different molar ratios:
[G.0.0] (m.p. = 350–354 K) with either of the two salts, ChCl
(
m.p. = 575 K) or [N.0.0]Cl (m.p. = 373–423 K) the homogeneous
liquid mixtures obtained could be properly named as DES from
a physicochemical point of view. For the rest of the mixtures,
where the HBD component is already liquid at room temperature,
the name of LMM would be more appropriate.
2.5 : 1, 2 : 1, 1.5 : 1 and 1 : 1. The main purpose was obtaining
homogeneous liquid mixtures at moderate or low temperatures,
capable of being used as solvents. Not all the combinations tested
yielded homogeneous mixtures at the working temperature.
Diethers [1.0.1], [2.0.2], [4.0.4], [3i.0.3i], [3F.0.3F], [Ph.0.Ph], and
one triether ([1.1.1]) when mixed with both salts, led to hetero-
geneous solid–liquid systems depending on the temperature and
the nature of the ammonium salt. Thus, in the case of ChCl,
solid–liquid phases were obtained in the overall range of tempera-
ture considered (from 258 to 343 K). In the case of [N.0.0]Cl,
although homogeneous liquid mixtures were formed at 343 K,
solid phases segregated from all mixtures as soon as the tempera-
ture started to decrease. These results point to the need of having
at least two hydroxyl groups in the HBD molecule and a molar
ratio ca. 2 : 1 of this component in the mixture in order to allow
strong interactions with the chloride anion of the ammonium salt,
leading to stable low melting mixtures.
Physical property measurements
The majority of properties characterizing solvents were measured
at a temperature range of 298.15–343.15 K, except in those cases
in which the above-mentioned critical temperature falls within
this interval, and the experimental data obtained were fitted to
model equations.
Density. Density (r) is a very important property of solvents,
essential for many chemical and engineering industrial appli-
cations. Fig. 2 shows the density measurements in all studied
solvents in the temperature range of 298.15–343.15 K. For
comparative purposes, the densities of pure glycerol mono-
ethers are also included, together with those of their mixtures
with ChCl and [N.0.0]Cl. As usual in organic solvents, densities
decrease with increasing temperature, due to the thermal
expansion. The influence of the R alkyl chain of the glycerol
ether on this property can also be ascertained from Fig. 2.
As the R chain length increases, the corresponding density
value decreases, most probably because the lengthening of the
alkyl chains results in an increase of the volume occupied
by molecules in the liquid. The branched alkyl ether [3i.0.0]
displays densities similar to those of [4.0.0]. As expected, the
On the other hand, monoethers of glycerol ([1.0.0], [2.0.0],
[4.0.0], [3i.0.0], [3F.0.0], [Ph.0.0] and [G.0.0]) were able to form
Scheme 1 Synthetic procedure for the preparation of [N.0.0]Cl.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
Fig. 2 Densities of glycerol monoethers [R.0.0] and their mixtures with ChCl and [N.0.0]Cl.
a
fluorinated ether [3F.0.0] displays higher densities, due to the Table 1 Density–temperature model parameter
higher atomic mass of fluorine atoms.
2
a
b
r
Interestingly, the densities of the mixtures are not quite
different from those of the pure [R.0.0] ethers, with the excep-
tion of [3F.0.0], for which the densities of the mixtures are
noticeably lower. Another observation from the plots in Fig. 2 is [3i.0.0]
that the nature of the ammonium salt does not affect appreci-
ably the densities of the corresponding mixtures. The density
[
[
[
1.0.0]
2.0.0]
4.0.0]
ꢂ0.000808
ꢂ0.000834
ꢂ0.000789
ꢂ0.000834
ꢂ0.001155
ꢂ0.000620
ꢂ0.000525
ꢂ0.000815
ꢂ0.000697
ꢂ0.000700
ꢂ0.000612
ꢂ0.000638
ꢂ0.000645
ꢂ0.000653
ꢂ0.000842
ꢂ0.000674
ꢂ0.000696
1.351813
1.305653
1.236374
1.265385
1.702551
1.306844
1.241157
1.528180
1.371036
1.403065
1.307098
1.284867
1.245904
1.263256
1.514573
1.364079
1.404102
1.0000
0.9997
1.0000
1.0000
1.0000
1.0000
0.9990
1.0000
0.9991
0.9997
0.9998
0.9999
1.0000
1.0000
0.9994
0.9998
0.9990
[
[
[
3F.0.0]
1.0.0]:ChCl
2.0.0]:ChCl
values of these mixtures range(at 298.15 K) in between those of [3F.0.0]:ChCl
conventional organic solvents (0.75–1.00 g cm ) and those
ꢂ3
[Ph.0.0]:ChCl
[
[
G.0.0]:ChCl
1.0.0]:[N.0.0]Cl
ꢂ3
of the majority of ILs (1.21–1.45 g cm ), 2 : 1 ChCl DES
ꢂ3
ꢂ3
12
(
e.g. 1.24 g cm for ureline and 1.18 g cm for glyceline)
[2.0.0]:[N.0.0]Cl
[4.0.0]:[N.0.0]Cl
ꢂ3 30
or pure glycerol (1.261 g cm ).
[
[
3i.0.0]:[N.0.0]Cl
3F.0.0]:[N.0.0]Cl
The experimental data of the densities obtained at different
temperatures were fitted to linear eqn (1) (r 4 0.999):
2
[Ph.0.0]:[N.0.0]Cl
G.0.0]:[N.0.0]Cl
[
r = aT + b
(1)
a
All the experimental density values are available in the ESI.
ꢂ3
where r is the density in g cm , T is the temperature in Kelvin,
and a and b are constants (related to the molar composition
in the case of mixtures). Full density–temperature model para-
meters are gathered in Table 1.
An interesting property related to density is the sound propa-
45
gation speed of the solvent (u), very much used in refinery.
Propagation speed trends versus temperature are very similar to
density trends, with the exception of [3F.0.0], whose low propaga-
tion speed can be explained according to the inverse dependence
of this property with density (Fig. 3).
Experimental values are near those of conventional solvents
ꢂ1
like water (1493 m s at r.t.). However, this property noticeably
Fig. 3 Sound propagation speeds of glycerol monoethers [R.0.0].
increases in the mixtures in the presence of the salt. Higher sound
propagation speeds in mixtures are due to the multiple and
stronger interactions formed between the two components in molecule, which is closely related to other properties like viscosity
the liquid medium. These experimental data were also fitted to a or conductivity. The molar volume of the solvents was calculated
model equation, as a linear function of temperature (see Table 2). according to semi empirical eqn (3), where M is the molar mass
and r the solvent density.
u = aT + b
(2)
ꢂ1
ꢂ1
Mðg mol
rðg cm
Þ
where u is the sound propagation speed in m s , T is
the temperature in Kelvin, and a and b are the adjustable
parameters.
Molar volume. Molar volume (V ), for a liquid, is the volume
m
filled by 1 mol of this substance at a given temperature.
This parameter is directly correlated to the free volume of a
V ¼
(3)
m
ꢂ3
Þ
In a mixture, M is the relative molar mass (and x the mole
fraction):
M = xHBDꢃMHBD + xHBAꢃMHBA
(4)
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
a
Table 2 Sound propagation speed–temperature model parameter
2
a
b
r
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
1.0.0]
2.0.0]
4.0.0]
3i.0.0]
ꢂ2.860
ꢂ3.111
ꢂ2.940
ꢂ3.052
ꢂ2.773
ꢂ2.997
ꢂ2.696
ꢂ2.654
ꢂ3.442
ꢂ3.263
ꢂ2.910
ꢂ3.000
ꢂ2.973
ꢂ3.084
ꢂ2.978
ꢂ4.382
2401.9
2377.3
2274.1
2289.5
2033.3
2684.2
2481.2
2274.5
2849.2
2818.4
2642.7
2572.8
2464.7
2518.3
2381.5
3151.2
1.0000
0.9998
0.9999
1.0000
1.0000
0.9997
1.0000
1.0000
0.9991
0.9996
0.9996
0.9994
0.9994
0.9991
0.9996
0.9900
3F.0.0]
1.0.0]:ChCl
2.0.0]:ChCl
3F.0.0]:ChCl
Ph.0.0]:ChCl
G.0.0]:ChCl
1.0.0]:[N.0.0]Cl
2.0.0]:[N.0.0]Cl
4.0.0]:[N.0.0]Cl
3i.0.0]:[N.0.0]Cl
3F.0.0]:[N.0.0]Cl
Ph.0.0]:[N.0.0]Cl
Fig. 5 Dielectric permittivities of glycerol monoether solvents [R.0.0].
inverse variation of the permittivity with temperature (eqn (5))
is the following:
e = aꢃT + b
(5)
a
All the experimental sound propagation speed values are available in
the ESI.
where e is the dielectric permittivity of the solvent, T is the
temperature in Kelvin, and a and b are adjustable parameters
(Table 3).
Molar volume is inversely proportional to density, so it increases
Using the dielectric permittivity and other properties like
as temperature does. Molar volumes of [R.0.0] and [R.0.0] LMM densities and refractive indexes it is possible to calculate
3
ꢂ1
(
Fig. 4) are higher than that of glycerol (73.03 cm mol at the dipole moment (m) of a molecule using the Onsager’s
298.15 K). In addition, molar volumes of the mixtures with equation (see Experimental section). As can be seen in Table 4,
[
N.0.0]Cl are greater than those of the mixtures with ChCl, due calculated dipole moments of the glycerol ethers show the
ꢂ1
to the different MHBA (211.73 vs. 139.62 g mol ).
same trend already observed for dielectric permittivities.
Solvent polarity. Different physical and chemical parameters However, dipole moments of glycerol ethers are higher than
can be used to estimate the polarity of the solvents. In this work those of conventional solvents, including glycerol and water,
we have decided to determine dielectric permittivity (e) and indicating their high capacity to solvate ionic and dipolar
dipole moment (m) as indicators of the polarity of the solvents. solutes.
Unfortunately, these parameters can only be measured for
Refractive index. Another physical property of a solvent, and
nonionic substances, which excludes the study of the mixtures related to its purity, is the refractive index (RI or n ). In general,
D
(
Fig. 5).
Permittivity values indicate that the shorter the alkyl chain RI values were determined in the temperature range of 298.15–
of the glycerol ether, the higher the polarity. Again, there is an 348.15 K.
RI is a precise measure of the polarizability of the molecules.
exception in the case of the fluorinated solvent. The dielectric
In the case of [R.0.0] ethers, nD values are closely similar in
permittivity of this solvent is so high (24.5) that it could only all cases, with the exception of [3F.0.0]. In fact, as the alkyl
be measured at 298.15 K using a different method (see details chain length increases, nD slightly increases too, as expected,
in the Experimental section). In general, the e values measured given the higher global polarizability. The low RI of the fluori-
are similar to those of other conventional alcohols (for nated ether is explained by the lower polarizability of this
instance ethanol 24.3 or isopropanol 17.9 at 298.15 K), but molecule due to electronegative fluorine atoms.
much lower than that of glycerol (46.5 at 298.15 K) or even
Concerning the mixtures, as is shown in Fig. 6, they always
ethylene glycol (37.7 at 298.15 K). The equation describing the have higher RI than the corresponding [R.0.0] monoethers,
Fig. 4 Molar volume values of glycerol monoethers [R.0.0] and their mixtures with ChCl and [N.0.0]Cl.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
a
Table 3 Dielectric permittivity–temperature model parameter
n
K
data to the linear model shown in eqn (6) (with T in Kelvin, a in
and b unitless):
D
ꢂ1
2
a
b
r
[
[
[
[
1.0.0]
2.0.0]
4.0.0]
3i.0.0]
ꢂ0.0247
ꢂ0.0177
ꢂ0.0294
ꢂ0.0144
30.19
23.15
21.21
21.06
0.9999
0.9988
0.9992
0.9988
n
D
= aT + b
(6)
m
The molar refractivity (R ) of the studied solvents can be
All the experimental dielectric permittivity values are available in the calculated using the Lorentz equation, from the molar volume
a
ESI.
D
and the n values:
2
nD ꢂ 1
Rm ¼
ꢂ1
ꢃ Vm
(7)
Table 4 Calculated dipole moment m (in Debye) of selected glycerol
ether and other conventional solvents (using the Onsager’s equation) at
2
nD þ 2
2
98.15 K
3
where R
and V
with radiation of optical frequency, R
m
is in cm mol , n
is the molar volume in cm mol . As n
D
is the measured refractive index
3
ꢂ1
Glycerol
.56
[1.0.0]
3.49
[2.0.0]
3.35
[4.0.0]
3.13
[3i.0.0]
3.51
[3F.0.0]
4.37
m
D
is measured
is related to the mean
m
2
electronic polarizability and it may be considered as a measure
of the capacity of the molecular orbitals to be deformed by an
electric field. All molar refractivity values of the glycerol solvents
are available in the ESI.†
Hexane
.00
Et
2
O
DCM
1.60
MeOH
1.70
THF
1.75
Water
1.85
0
1.38
Viscosity. Viscosity is a very important solvent property as it
due to the salt presence. Refractive indices are similar for the determines some of its possible uses. In this work we have
mixtures regardless of the nature of the ammonium salt (ChCl measured kinematic viscosity (n, in cSt) in the temperature
or [N.0.0]Cl), although they are slightly higher in the case of range of 298.15–343.15 K. The dynamic viscosity (Z, in cP) can
[
N.0.0]Cl. In all the cases, the n
D
values are analogous to those be derived from kinematic viscosity and density values.
Aliphatic glycerol monoethers have viscosities much lower
of conventional solvents (1.32–1.50) and glycerol (1.47). It is
worth mentioning that the mixtures of both aromatic ethers (35–45 cP at r.t.) than glycerol (1200 cP at r.t.), probably because
have noticeably high RI (nD 4 1.53), much closer to those of substitution of a hydroxyl group by an alkoxide group has
aromatic solvents like toluene and xylene, or sugar-based a remarkable effect on the strength of the intermolecular
4
6
DES.
Interestingly, in the mixtures the variation of n
interactions.
D
with the
The high viscosity of glycerol is indeed a problem in many
length of R follows the opposite trend of that observed for pure applications, due to the higher operational and energy costs.
glycerol ethers. This can be explained because of the existence When comparing the viscosity values of glycerol ethers with
of two opposed physical effects. On one hand, as mentioned different R chain lengths, we observe that the longer the alkyl
above, the polarizability of the [R.0.0] component increases chain, the higher the viscosity. The presence of a ramification
with the R length, leading to an increase of RI. On the other in the substituent leads to an increase in the viscosity of the
hand, a longer R chain results in a lower density of the mixture, monoether ([3i.0.0] vs. [4.0.0], Fig. 7).
so light propagation speed increases and RI decreases accord-
ingly. It is clear that this second mechanism overrules the first double that of its non-fluorinated homologue [2.0.0]. This may
one in the case of the mixtures.
be due to the higher density of [3F.0.0] and hence its lower free
A singular case is that of [3F.0.0], whose viscosity is almost
1
2
Dependence of the refractive index on temperature is quite volume. According to hole theory, as this solvent has low free
similar in all glycerol monoethers, as can be seen in Table 5, volume, the flow of a molecule of [3F.0.0] through a hole inside
2
which gathers the excellent fittings (r 4 0.999) of the experimental the liquid is more difficult, which results in a higher viscosity.
Fig. 6 Refractive indices of the studied solvents: glycerol monoethers [R.0.0] and their mixtures with ChCl and [N.0.0]Cl.
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
a
Table 5 Refractive index–temperature model parameter
for malonic acid:ChCl). Finally, mixtures containing aromatic
ethers present extremely high viscosities, which can be explained
by the stronger aromatic rings pꢂp and p–OH intermolecular
interactions.
2
a
b
r
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
1.0.0]
2.0.0]
4.0.0]
3i.0.0]
ꢂ0.000346
ꢂ0.000402
ꢂ0.000345
ꢂ0.000354
ꢂ0.000317
ꢂ0.000325
ꢂ0.000299
ꢂ0.000155
ꢂ0.000291
ꢂ0.000262
ꢂ0.000315
ꢂ0.000308
ꢂ0.000267
ꢂ0.000269
ꢂ0.000266
ꢂ0.000263
ꢂ0.000259
ꢂ0.000288
ꢂ0.000270
1.5441
1.5596
1.5439
1.5422
1.4813
1.5672
1.5566
1.5055
1.5439
1.5056
1.6275
1.6297
1.5570
1.5536
1.5489
1.5477
1.5173
1.6168
1.6161
0.9998
0.9990
0.9999
Viscosity is a very temperature-dependent property. When the
0.9999 temperature increases molecular mobility increases too, resulting
3F.0.0]
0.9987
0.9993
0.9992
in an exponential decrease of viscosity. The experimental data
1.0.0]:ChCl
2.0.0]:ChCl
4.0.0]:ChCl
3i.0.0]:ChCl
3F.0.0]:ChCl
Ph.0.0]:ChCl
G.0.0]:ChCl
1.0.0]:[N.0.0]Cl
2.0.0]:[N.0.0]Cl
4.0.0]:[N.0.0]Cl
3i.0.0]:[N.0.0]Cl
3F.0.0]:[N.0.0]Cl
Ph.0.0]:[N.0.0]Cl
G.0.0]:[N.0.0]Cl
47
have been fitted to an Arrhenius-type eqn (8), which describes
0.9990 adequately the viscosity behaviour with temperature:
0.9992
0.9995
0.9997
0.9992
0.9997
EZ
ln Z ¼
þ b
(8)
RT
0.9999 where Z is the dynamic viscosity in cP, T is the temperature in
ꢂ1
ꢂ1
0.9996
0.9999
0.9999
Kelvin, R is the gas constant (8.31445 J K mol ), E
of activation of viscous flow and b is a constant. The values
in Table 6 show the different trends mentioned above,
Z
is the energy
0.9996 of E
0.9991
Z
ꢂ1
ꢂ1
being o40 kJ mol for the pure monoethers and 36–87 kJ mol
ꢂ1
a
for the different mixtures (36–51 kJ mol for aliphatic ethers and
All the experimental refractive index values are available in the ESI.
ꢂ1
6
3–87 kJ mol for aromatic ones). As can be observed in Table 6,
[
Z
R.0.0]/[N.0.0]Cl mixtures have E values higher than that of
The viscosities of the mixtures of glycerol ethers with ChCl
or [N.0.0]Cl increase moderately with regard to that of pure
R.0.0] monoethers. Abbott et al. stated that whereas the
viscosity of the mixture glycerol:ChCl decreases as the salt
concentration increases, the opposite is true in the case of
diols. Thus, the situation of glycerol monoethers seems to be
analogous to that of diols. Apart from this, [N.0.0]Cl mixtures
are three times more viscous than their homologues containing
ChCl. This seems to be due to the additional interactions
ꢂ1
8
glycerol:ChCl (45.1 kJ mol ).
We can also point out that our activation energy average for
[
ꢂ1
the pure monoethers (E
corresponding average found in eight diols (E = 37.3 kJ mol ).
Z
Regarding the mixtures, there seems to be a weak correlation
between the dynamic viscosity values from the different kind of
mixtures, at 298.15 K for example, and the activation energies,
as can be seen from Tables S29 and S31 (ESI†) (for Z values)
and Table 6 (for E values). A similar behaviour was observed by
Z
Maroncelli et al. for a series of ionic liquids.
Ionic conductivity. The ionic conductivity (k) is a physical
property typically determined for DES. This property is very
important for these kinds of solvents because of their interesting
electrochemical applications. Temperature is again an essential
parameter influencing the conductivity. Conductivity values
measured in the temperature range 298.15–343.15 K have been
fitted to an Arrhenius-type equation similar to that employed
with viscosity:
Z
= 39.3 kJ mol ) is very close to the
ꢂ1 48
(
particularly hydrogen bonds) established by the former, which
49
have a longer carbon chain and an additional hydroxyl group.
Regardless of the ammonium salts used, the viscosity order
observed is: [1.0.0] o [2.0.0] o [4.0.0] o [3i.0.0] o [3F.0.0] |
[
Ph.0.0] { [G.0.0]. The explanation for this order has been
12
mentioned above. Mixtures of aliphatic ethers with [N.0.0]Cl
show viscosities close to DES ones (4400 cP at 298.15 K) and
they are quite similar to those shown in mixtures between
50
8
glycerol and ChCl (in the same 2 : 1 ratio). Mixtures of aliphatic
ethers with ChCl display a viscosity (E150 cP at 298.15 K)
higher than that of conventional solvents, but lower than that
of other reported DES (750 cP at 298 K for urea:ChCl or 1124 cP
Ek
ln k ¼
þ b
(9)
RT
Fig. 7 Dynamic viscosities of studied solvents: glycerol monoethers [R.0.0] and their mixtures with ChCl and [N.0.0]Cl.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
a
Table 6 Viscosity–temperature model parameter
and b is a constant (Table 7). In this case E
similar to the values reported for ionic liquids,
values are quite
k
5
1–53
ꢂ1
2
but slightly
E
Z
/kJ mol
b
r
higher than those measured with the usual DES, including
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
[
1.0.0]
2.0.0]
4.0.0]
3i.0.0]
39.043
37.946
37.418
40.344
41.838
36.940
34.137
36.164
37.063
63.892
66.480
50.209
49.884
49.012
51.469
51.118
85.149
87.177
ꢂ12.087
ꢂ11.667
ꢂ11.370
ꢂ12.491
ꢂ12.707
ꢂ10.056
ꢂ8.923
0.9994
0.9987
0.9990
ꢂ1
8
glycerol:ChCl (ꢂ27.9 kJ mol ). Conductivity has an inverse
variation with viscosity, as is shown in the inverse relation
0.9991 existing between both activation energies, E
k
and E . The higher
Z
3F.0.0]
0.9972
0.9984
0.9996
the solvent viscosity, the more difficult the flow of ions, resulting
in a decrease of ionic conductivity. A strong interaction between
0.9974 the HBD component of the mixture and the chloride anion
increases the effective size of the latter, making it difficult for
ion mobility and thus decreasing ionic conductivity.
0.9994
1.0.0]:ChCl
2.0.0]:ChCl
3i.0.0]:ChCl
3F.0.0]:ChCl
Ph.0.0]:ChCl
G.0.0]:ChCl
1.0.0]:[N.0.0]Cl
2.0.0]:[N.0.0]Cl
4.0.0]:[N.0.0]Cl
3i.0.0]:[N.0.0]Cl
3F.0.0]:[N.0.0]Cl
Ph.0.0]:[N.0.0]Cl
G.0.0]:[N.0.0]Cl
ꢂ9.529
ꢂ9.912
0.9987
0.9977
ꢂ18.334
ꢂ18.695
ꢂ14.179
ꢂ14.053
ꢂ13.707
ꢂ14.484
ꢂ14.345
ꢂ25.267
ꢂ25.368
0.9980
0.9983
0.9990
0.9985
In Fig. 8 we can see this inverse behaviour exhibited by
conductivity with regard to viscosity: conductivity in the mixtures
containing ChCl (at 298.15 K) is three times higher than in the
0.9982 ones of [N.0.0]Cl. Regardless of the ammonium salts in the
0.9976
0.9992
mixture, the conductivity order is: [1.0.0] 4 [3F.0.0] 4 [2.0.0] 4
[3i.0.0] 4 [4.0.0] 4 [Ph.0.0] 4 [G.0.0]. The longer the R chain in
a
All the experimental viscosity values are available in the ESI.
the ether, the lower the conductivity. In the very viscous mixtures
containing aromatic ethers the conductivity is quite low, with
values comparable to those of viscous ILs.
ꢂ1
where k is the ionic conductivity in mS cm , T is in Kelvin,
Surface tension. The last measured property is the surface
tension (g), which provides interesting information (e.g. for
emulsions and surfactant applications) about the cohesive
forces operating in the solvent. The intermolecular attractive
forces in the liquid reduces the surface area of the liquid–air
interface, increasing the necessary energy for enlarging the
R is the gas constant, E
k
the energy for conduction activation
a
Table 7 Conductivity–temperature model parameter
ꢂ1
2
E
k
/kJ mol
b
r
2
5
[
[
[
[
[
[
[
[
[
[
[
[
[
[
1.0.0]:ChCl
2.0.0]:ChCl
4.0.0]:ChCl
3i.0.0]:ChCl
3F.0.0]:ChCl
Ph.0.0]:ChCl
G.0.0]:ChCl
1.0.0]:[N.0.0]Cl
2.0.0]:[N.0.0]Cl
4.0.0]:[N.0.0]Cl
3i.0.0]:[N.0.0]Cl
3F.0.0]:[N.0.0]Cl
Ph.0.0]:[N.0.0]Cl
G.0.0]:[N.0.0]Cl
ꢂ38.216
ꢂ31.869
ꢂ30.845
ꢂ35.198
ꢂ34.960
ꢂ59.605
ꢂ67.371
ꢂ48.916
ꢂ46.401
ꢂ41.688
ꢂ45.148
ꢂ48.787
ꢂ77.392
ꢂ81.381
22.688
20.093
19.282
21.046
21.282
28.912
30.896
25.660
24.326
22.137
23.451
25.303
34.306
34.827
0.9925 surface. As shown in Fig. 9, the surface tension values of
0.9993
0.9927
0.9986
ꢂ1
[
R.0.0] are in between 25 and 40 mN m , quite similar to those
of n-alkanols (e.g. 25.1 for BuOH), THF (28.2), halogenated
0.9895 solvents (29.5 for DCM) or aromatic solvents (29.6 for benzene).
0.9972
0.9980
0.9987
In general, the longer the R alkyl chain, the lower the surface
tension. The most probable mechanism for this structure–
0.9995 property relationship is related to the ability of the glycerol
ethers to form strong intermolecular hydrogen bonds. Pure
glycerol has a high surface tension (63.4), which diminishes
0.9928 when one of the hydroxyl groups is blocked. Besides this,
0.9984
0.9988
0.9991
0.9974
longer alkyl substituents probably disturb the intermolecular
hydrogen bonding network, leading to lower surface tensions.
a
All the experimental conductivity values are available in the ESI.
Fig. 8 Ionic conductivities of mixtures of glycerol monoethers [R.0.0] and ChCl or [N.0.0]Cl.
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Fig. 9 Surface tension of solvents: glycerol monoethers [R.0.0] and their mixtures with ChCl and [N.0.0]Cl.
a
Table 8 Surface tension–temperature model parameter
N,N,N-Triethyl-2,3-dihydroxy-propan-1-aminium chloride has
been used for the first time to prepare these kinds of solvents.
Its use leads to the so-called DES coming formally from the
same renewable source, glycerol, thus enlarging the possibilities
of this platform molecule as a source of solvents with very
2
a
b
r
[
[
[
[
[
[
[
[
[
[
[
1.0.0]
2.0.0]
4.0.0]
3i.0.0]
ꢂ0.0766
ꢂ0.0673
ꢂ0.0571
ꢂ0.0690
ꢂ0.0608
ꢂ0.0361
ꢂ0.0414
ꢂ0.0632
ꢂ0.1151
ꢂ0.0878
ꢂ0.2013
61.54
52.80
45.94
48.45
45.69
52.47
43.60
63.12
79.06
59.36
0.9999
1.0000
0.9995
0.9992 different features.
3F.0.0]
0.9999
0.9994
0.9988
Liquid and stable mixtures have been characterized and a
series of critical physicochemical properties, such as density,
0.9991 viscosity, refractive index, conductivity and surface tension,
1.0.0]:ChCl
3F.0.0]:ChCl
Ph.0.0]:ChCl
1.0.0]:[N.0.0]Cl
3F.0.0]:[N.0.0]Cl
Ph.0.0]:[N.0.0]Cl
0.9998
0.9994
0.9982
have been measured in most cases within the temperature
range of 298 to 343 K, adequate for their application as a green
solvent in soft conditions. These measurements evidence the
possibility of fine tuning the physicochemical properties of the
solvent by changing the substitution pattern in the HBD
109.43
a
All the experimental surface tension values are available in the ESI.
In the mixtures with ammonium salts, an increase in surface compound (glycerol ether) or in the ammonium salt, thus
tension is observed with regard to pure glycerol ethers. The broadening the scope of application of these mixtures, and
values are higher for [N.0.0]Cl, due to stronger intermolecular allowing their a` la carte preparation. Comparison of the mixtures
interactions. The experimental values for the mixtures range prepared herein with the ones containing HBD compounds and
ꢂ1
between 30 and 50 mN m , similar to those of the majority of choline chloride, and with commonly used ionic liquids reveals
4
6
IL reported in the literature. Again, maximum values are that these new mixtures cover a similar and wide range of
found for aromatic ether mixtures. Even so, reported surface physicochemical properties with a much more homogeneous
tension values are lower than that of water (72.8), pure glycerol origin and a consistent preparation procedure, and therefore
8
or glycerol:ChCl DES (E52).
they can be used in the substitution of many previously described
Surface tension diminishes linearly with temperature (fitting to IL and DES. Some practical applications of these solvents, as
eqn (10)), due to the decrease of the liquid cohesive forces.
reaction media for catalytic processes or as extraction solvents, are
currently being explored by our group.
g = aT + b (10)
ꢂ1
where g is the surface tension in mN m , T is temperature in
Kelvin, and a and b are fitting constants (Table 8).
Conflicts of interest
The authors declare no conflict of interest.
Conclusions
Mixtures of several monoalkyl glyceryl ethers with two ammo-
nium salts bearing hydroxyl groups, choline chloride and
Acknowledgements
N,N,N-triethyl-2,3-dihydroxypropan-1-aminium chloride, have Financial support from the MINECO (project CTQ2014-52367-R),
been prepared. These mixtures display a wide temperature Gobierno de Arag o´ n and the European Regional Development
range of the liquid state and can be considered as low melting Funds (Group E11) is acknowledged. A. L.-D. thanks the MECD
or low freezing point mixtures. In the case of the mixtures for a FPU grant (FPU14/04338). The authors thank Dr Carlos
containing [Ph.0.0] or [G.0.0], both solids at room temperature, Lafuente for access to a surface tensiometer and helpful
the mixtures are also probably deep eutectic solvents DES. advice.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
References
28 Y. Gu and F. J ´e r oˆ me, Green Chem., 2010, 12, 1127–1138.
2
3
3
3
3
3
3
3
3
3
3
4
9 J. I. Garc ´ı a, H. Garc ´ı a-Mar ´ı n, J. A. Mayoral and P. P ´e rez,
1
2
T. Welton, Chem. Rev., 1999, 99, 2071–2084.
J. P. P. Hallett and T. Welton, Chem. Rev., 2011, 111,
Green Chem., 2013, 15, 2283.
0 J. I. Garcia, H. Garcia-Marin, J. A. Mayoral and P. Perez,
Green Chem., 2010, 12, 426–434.
1 L. Moity, M. Durand, A. Benazzouz, C. Pierlot, V. Molinier
and J.-M. Aubry, Green Chem., 2012, 14, 1132–1145.
2 L. Moity, Y. Shi, V. Molinier, W. Dayoub, M. Lemaire and
J.-M. M. Aubry, J. Phys. Chem. B, 2013, 117, 9262–9272.
3 J. I. I. Garc ´ı a, E. Pires, L. Aldea, L. Lomba, E. Perales and
B. Giner, Green Chem., 2015, 17, 4326–4333.
4 E. Perales, J. I. Garc ´ı a, E. Pires, L. Aldea, L. Lomba and
B. Giner, Chemosphere, 2017, 183, 277–285.
5 A. Leal-Duaso, M. Caballero, A. Urriolabeitia, J. A. Mayoral,
J. I. Garc ´ı a and E. Pires, Green Chem., 2017, 19, 4176–4185.
6 W. F. Richey, Kirk-Othmer Encyclopedia of Chemical Technology,
3
508–3576.
M. Deetlefs, M. Fanselow and K. R. Seddon, RSC Adv., 2016,
, 4280–4288.
3
6
4
5
6
T. Welton, Green Chem., 2011, 13, 225.
M. Dilip, Nanomater. Energy, 2012, 1, 193–206.
A. Paiva, R. Craveiro, I. Aroso, M. Martins, R. L. Reis and
A. R. C. Duarte, ACS Sustainable Chem. Eng., 2014, 2, 1063–1071.
C. Ruß and B. K o¨ nig, Green Chem., 2012, 14, 2969–2982.
A. Abbott, R. Harris, K. Ryder, C. D’Agostino, L. Gladden and
M. Mantle, Green Chem., 2011, 13, 82–90.
7
8
9
A. P. Abbott, D. Boothby, G. Capper, D. L. Davies and R. K.
Rasheed, J. Am. Chem. Soc., 2004, 126, 9142–9147.
1
1
1
1
1
1
1
1
1
1
2
0 B. Kudłak, K. Owczarek and J. Namie ´s nik, Environ. Sci.
2000.
Pollut. Res., 2015, 22, 11975–11992.
1 M. Francisco, A. van den Bruinhorst and M. C. Kroon,
Angew. Chem., Int. Ed., 2013, 52, 3074–3085.
7 M. Ricciardi, F. Passarini, I. Vassura, A. Proto, C. Capacchione,
R. Cucciniello and D. Cespi, ChemSusChem, 2017, 10, 2291–2300.
8 Y. K. Endah, M. S. Kim, J. Choi, J. Jae, S. D. Lee and H. Lee,
Catal. Today, 2017, 293–294, 136–141.
9 S. E. Kondawar, C. R. Patil and C. V. Rode, ACS Sustainable
Chem. Eng., 2017, 5, 1763–1774.
0 J. R. Ochoa, O. G o´ mez de Miranda, N. Blanco, B. Maestro
and S. Prieto, WO pat., WO2017017307 (A1), 2017.
1 L. Onsager, J. Am. Chem. Soc., 1936, 58, 1486–1493.
2 L. Aldea, J. M. Fraile, H. Garcia-Marin, J. I. Garcia, C. I.
Herrerias, J. A. Mayoral and I. Perez, Green Chem., 2010, 12,
2 E. L. Smith, A. P. Abbott and K. S. Ryder, Chem. Rev., 2014,
1
14, 11060–11082.
3 P. Liu, J.-W. Hao, L.-P. Mo and Z.-H. Zhang, RSC Adv., 2015,
, 48675–48704.
4 S. Khandelwal, Y. K. Tailor and M. Kumar, J. Mol. Liq., 2016,
15, 345–386.
5
2
4
4
5 A. Abo-Hamad, M. Hayyan, M. A. AlSaadi and M. A. Hashim,
Chem. Eng. J., 2015, 273, 551–567.
6 D. V. V. Wagle, H. Zhao and G. A. A. Baker, Acc. Chem. Res.,
435–440.
2
014, 47, 2299–2308.
7 H. Zhao, G. A. Baker and S. Holmes, Org. Biomol. Chem.,
011, 9, 1908–1916.
8 H. Zhao, G. A. Baker and S. Holmes, J. Mol. Catal. B: Enzym.,
011, 72, 163–167.
4
4
4
4
3 E. Perales, C. Belen Garcia, L. Lomba, L. Aldea, J. I. Garcia
and B. Giner, Ecotoxicol. Environ. Saf., 2016, 132, 429–434.
4 L. Aldea, J. I. Garc ´ı a and J. A. Mayoral, Dalton Trans., 2012,
2
4
1, 8285–8289.
5 B. Fazelabdolabadi and A. Bahramian, Ind. Eng. Chem. Res.,
010, 49, 3469–3473.
2
9 F. del Monte, D. Carriazo, M. C. Serrano, M. C. Guti ´e rrez
and M. L. Ferrer, ChemSusChem, 2014, 7, 999–1009.
0 D. J. G. P. van Osch, L. F. Zubeir, A. van den Bruinhorst,
M. A. A. Rocha and M. C. Kroon, Green Chem., 2015, 17,
2
6 A. Hayyan, F. S. Mjalli, I. M. AlNashef, T. Al-Wahaibi, Y. M.
Al-Wahaibi and M. A. Hashim, Thermochim. Acta, 2012, 541,
7
0–75.
7 R. F u¨ rth, Math. Proc. Cambridge Philos. Soc., 1941, 37,
52–275.
4518–4521.
4
2
2
1 F. S. Mjalli, J. Naser, B. Jibril, V. Alizadeh and Z. Gano,
J. Chem. Eng. Data, 2014, 59, 2242–2251.
2 Q. Zhang, K. De Oliveira Vigier, S. Royer and F. J ´e r oˆ me,
Chem. Soc. Rev., 2012, 41, 7108–7146.
2
4
4
8 Y. Marcus, The properties of solvents, Wiley, 1998.
9 H. Jin, B. O’Hare, J. Dong, S. Arzhantsev, G. A. Baker,
J. F. Wishart, A. J. Benesi and M. Maroncelli, J. Phys. Chem.
B, 2011, 115, 1333.
2
2
3 R. B. Leron and M.-H. Li, J. Chem. Thermodyn., 2013, 57, 131–136.
4 A. Yadav, S. Trivedi, R. Rai and S. Pandey, Fluid Phase
Equilib., 2014, 367, 135–142.
5
5
5
5
0 J. O. Bockris and G. W. Hooper, Discuss. Faraday Soc., 1961,
32, 218–236.
2
2
2
5 M. K. AlOmar, M. Hayyan, M. A. Alsaadi, S. Akib, A. Hayyan
and M. A. Hashim, J. Mol. Liq., 2016, 215, 98–103.
6 J. I. Garc ´ı a, H. Garc ´ı a-Mar ´ı n and E. Pires, Green Chem., 2014,
1 I. Osada, H. De Vries, B. Scrosati and S. Passerini, Angew.
Chem., Int. Ed., 2016, 55, 500–513.
2 A. P. Abbott, G. Capper and S. Gray, ChemPhysChem, 2006, 7,
803–806.
16, 1007.
7 A. E. E. D ´ı az-Alvarez, J. Francos, B. Lastra-Barreira, P. Crochet
´
3 A. P. Abbott, ChemPhysChem, 2004, 5, 1242–1246.
and V. Cadierno, Chem. Commun., 2011, 47, 6208–6227.
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.