- Email: [email protected]
Effect of anion on thermophysical properties of N,N-diethanolammonium based protic ionic liquids
Accepted Manuscript Effect of anion on thermophysical properties of N,Ndiethanolammonium based protic ionic liquids
Anirban Sarkar, Gyanendra Sharma, Dharmendra Singh, Ramesh L. Gardas PII: DOI: Reference:
S0167-7322(17)31718-X doi: 10.1016/j.molliq.2017.07.025 MOLLIQ 7602
To appear in:
Journal of Molecular Liquids
Received date: Revised date: Accepted date:
21 April 2017 10 June 2017 8 July 2017
Please cite this article as: Anirban Sarkar, Gyanendra Sharma, Dharmendra Singh, Ramesh L. Gardas , Effect of anion on thermophysical properties of N,N-diethanolammonium based protic ionic liquids, Journal of Molecular Liquids (2017), doi: 10.1016/ j.molliq.2017.07.025
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Effect of anion on thermophysical properties of N,N-diethanolammonium based protic ionic liquids Anirban Sarkar, Gyanendra Sharma, Dharmendra Singh and Ramesh L. Gardas* Indian Institute of Technology Madras, Chennai 600036 *Corresponding author. Tel.: +91 44 2257 4248; fax: +91 44 2257 4202 E.mail address: [email protected]; URL: http://www.iitm.ac.in/info/fac/gardas
IP
T
----------------------------------------------------------------------------------------------------------------
CR
Abstract
In this work, we investigated the influence of two different anions on the various thermophysical
US
properties of N,N-diethanolammonium based protic ionic liquids (PILs). For this purpose, two N,N-diethanolammonium based PILs have been synthesised by using trifluoroacetate and lactate
AN
as counter anions and characterized through spectral techniques. The thermophysical properties such as density and speed of sound were measured as function of temperature from (303.15 to 343.15) K at atmospheric pressure. The experimental density values used to calculate thermal
M
expansion coefficient (), isentropic compressibility (s), molar volume (Vm) and standard
ED
entropy (So). The thermal stabilities of the PILs were investigated by thermogravimetric analysis (TGA), and phase transition measurement was done by using differential scanning calorimerty
PT
(DSC). Further, to understand the interionic strength of PILs, lattice potential energy (Upot) has been calculated. The experimental densities have been compared with the estimated densities
CE
obtained from Gardas and Coutinho model and obtained results are in good agreement. Further, the density values of studied protic ionic liquids have been compared with other protic ionic
AC
liquids. Finally, the impact of structure variants in terms of the carboxylate anions has been analyzed over the thermophysical properties of N,N-diethenolammonium based PILs.
Keywords: Ionic liquid; thermal stability; isothermal expansion; isentropic compressibility; Gardas and Coutinho model.
1
ACCEPTED MANUSCRIPT 1. Introduction Ionic liquids (ILs) are molten salts with melting point below 100˚C and most of them are liquid even below room temperature. ILs have several exceptional properties such as low volatility, extremely low flammability, high specific solubility, wide liquid range, high electrical conductivity and good thermal and chemical stability [1– 6]. Combination of these properties
T
makes them an interesting class of substances for both industries as well as academia. Generally,
IP
ILs can be divided into two categories namely protic ionic liquids (PILs) and aprotic ionic liquids (AILs). Much research has been done on both protic as well as aprotic ionic liquids and
CR
owing to the above specified properties they have several potential applications. In contrast to AILS, the PILs possess additionally hydrogen bond capacity which is significant for many
US
biological processes and as a medium of proton conduction for polymer membrane fuel cells [7]. PILs can be synthesised by Brönsted acid-base neutralization reaction in that proton transfer
AN
takes place from acid to base. There is a chance of equilibrium between reactant and products due to incomplete proton transfer [8, 9], which can be avoid by selecting proper acid and base
M
pair which is having pKa difference (ΔpKa ) greater than 4 [10]. The hydroxylammonium based PILs have many advantageous applications due to their hydrogen bond donation ability [11]. In
ED
recent years, research on ammonium based PILs have been focussed for various applications, such as organic synthesis, oxygen reduction reaction, antimicrobial activity, extraction of
PT
complexes, CO2 gas absorption, separation process and polymer dissolution etc. [12-19]. However, before exploring PILs for these applications, it is essential to study the thermophysical
CE
properties of PILs. The possibility of tailoring the properties of an ionic liquid to meet the requirements of some specific application by the tuning of combinations of cation and anion.
AC
Therefore, the knowledge of accurate thermophysical properties of ILs are essential, and also for the design of processes and products involving these compounds [20–22]. For scrutinizing the intermolecular interactions between ions, there are several reports available in which effect of structure variant on thermophysical properties were studied [11, 23]. However, to the best of our information there are few reports available, which focussed the influence of structure variant on thermophysical properties of PILs. In this framework, we have synthesized two N,N-diethanolammonium (DEA) based PILs with two different counter anions (trifluoroacetic acid [Tfa] and lactic acid [Lac]) namely N,N-
2
ACCEPTED MANUSCRIPT diethanolammonium trifluoroacetate ([DEA][Tfa]) and N,N-diethanolammonium lactate ([DEA][Lac]) and characterized by spectral techniques includes 1H NMR, HRMS and IR spectral techniques. Thermal decomposition temperature (Td) and glass transition temperature (Tg) of the studied PILs were evaluated by TGA and DSC experiment respectively. Before studying the various temperature-dependent thermophysical properties such as density and speed of sound in the range (303.15 to 343.15) K the water content should be accounted. The
T
experimental density and speed of sound were used to calculate the industrially relevant
IP
parameters such as thermal expansion coefficient () and isentropic compressibility (s). Further,
CR
the experimental densities of PILs were correlated with the estimated density values obtained from Gardas and Coutinho model. In addition, the calculated and derived thermodynamic properties were scrutinized to investigate the influence of anion structure of the PILs and
AN
US
compared with other PILs [11, 24-25].
M
2. Experimental Section 2.1 Materials
ED
All reagents used in this experiment were Analytical Reagent (AR) grade and used without
presented in Table 1.
PT
further purification. The name of the used chemicals, source and their purity in mass fraction are
CE
2.2 Synthesis of ionic liquids
The PILs were synthesized by Brönsted acid-base neutralization method where proton
AC
transfer takes place from acid to base. Here, we used diethanolamine (DEA) as a base and methanol as a solvent, trifluoroacetic acid and lactic acid have been used as acid. The methanolic solution of DEA was taken in a two-necked round bottom flask which is equipped with a condenser and dropping funnel. The whole set-up was placed on an ice–bath to maintain low temperature (~5 C). The equimolar amount of acid was added drop wise to the above reaction mixture using dropping funnel. After the complete addition of acid, the reaction mixture was stirred with the help of a magnetic stirrer for next 24 hrs under nitrogen atmosphere and at room temperature. After completion, the methanol (solvent) was removed from the reaction mixture by putting it into rotary evaporator at ~50 C for 2 hrs. In order to remove moisture and the 3
ACCEPTED MANUSCRIPT resultant mixture (PIL) was dried for 36 hrs under high vacuum and kept in nitrogen atmosphere. As shown in Scheme 3.1, the syntheses of studied PILs were achieved by the equimolar mixture of acid and base, without any by-product formation.
Alkyl carboxylic acid
Protic ionic liquid
T
Diethanolammine
2.3 Characterization of PILs 1
H NMR (500 MHz, CDCl3) δ 3.248 (t, 4H); 3.877 (t, 4H).
AN
[DEA][Tfa]
CR
US
Scheme 1. Synthesis of diethanolamine based PILs
IP
R = -CF3 and -CH3CH(OH)
[DEA][Lac]
1
M
HRMS (ESI+) m/z for [M]+ calcd 106.0868 and found 106.0850. H NMR (500 MHz, CDCl3) δ 1.338 (t, 3H); 3.253 (m, 4H); 3.877 (m, 4H); 4.108
ED
(m, 1H).
CE
IR Spectroscopy
PT
HRMS (ESI+) m/z for [M]+ calcd 106.0868 and found 106.0861.
IR spectra were obtained on a JASCO FT/IR 4100 spectrometer which has maximum
AC
resolution of 0.9 cm-1 and 22000:1 signal to noise ratio by using KBr disk. In IR spectra, a broad band in the range of 3500-3000 cm-1 shows ν (O-H), ν (N-H) and also ammonium peak. Peak around 1620 cm-1 indicates the presence of carbonyl stretching ν (C=O) and bending peak of ν (N-H). (see supporting information Fig S3) 2.4 Water Content The water content measurements of the synthesized PILs have been done by using Metrohm 870KF Titrinoplus Karl-Fischer titrator. This instrument operates volumetric titration 4
ACCEPTED MANUSCRIPT principle using dual platinum electrodes that allows the detection of water from less than 10 ppm to 100%. It was calibrated with standard sample. The PILs, [DEA][Tfa] and [DEA][Lac] were dissolved in dried methanol solution and followed by titration against Karl-Fisher reagent, and observed water contents are presented in Table 2.
T
2.5 Thermogravimetric Analysis (TGA)
IP
Thermogravimetric analysis of the synthesized PILs have been carried out by using TA
CR
instrument Hi-Res TGA Q500 with weighing precision of 0.01% from room temperature to 600 C at 10 C min-1 heating rate under nitrogen atmosphere (flow rate is 40 ml·min-1 and 60 ml·minfor the balance and furnace respectively) in an open platinum pan. Two thermocouples are
US
1
positioned immediately adjacent to the sample to ensure simultaneous measurement of sample
AN
temperature and heat rate control accurately and precisely. Nickel metal was used for the calibration of Curie point temperature.
M
2.6 DSC Measurement
ED
The phase transition measurement was carried out by DSC (TA instruments DSC Q200), which have the sensitivity of 0.2 μW and temperature accuracy ±0.1 C. The scanning sequence
PT
involves the freezing the samples to -80 C temperature (up to 20 min) followed by heating the sample to 50 C with a heating rate 20 ˚C·min-1 with flow rate of 50 cm3·min-1. Using this
temperature (Tg).
CE
technique it is possible to detect fusion and crystallization events as well as glass transition
AC
2.6 Density and Speed of sound Temperature dependent density and the speed of sound were measured by Anton Paar (DSA 5000M) instrument which works on the oscillating U-tube principle. The sample was introduced into a borosilicate glass U tube which is electronically vibrated at its characteristic frequency which depends on the density of the sample. The instrument also consists of a stainless steel cell for sound velocity measurement. Both these cells are temperature-controlled by built-in Peltier thermostats (PT-100) which have an accuracy of ± 0.01 K. This system enable us to measure the density from 0 to 3000 kg·m-3 and speed of sound in the range 1000–2000 m·s-1 in a 5
ACCEPTED MANUSCRIPT wide range of pressure and temperature. Other than that sample handling is very convenient here and also the errors of sample filling can be detected automatically. As a result of these instrumental properties, we get an accurate and reproducible result of measurement. The uncertainty in the measurement of density was ± 0.1 kg·m-3 and that of speed of sound was ± 0.5
IP
T
m·s-1.
CR
3. Results and Discussion
US
The syntheses of the ILs were achieved by the reaction of diethanolammine with equimolar amount of acid and the formation of the PILs has been confirmed by 1H, and IR
AN
spectra. Though, the synthesis of studied PILs was achieved by equimolar mixture of acid and base, so there is no change of any side product formation. Further, the purity of synthesized PILs
M
was analyzed through HRMS spectrum and as shown in Figure S4 and S5, we get appropriate cationic peak of synthesized PILs in HRMS spectrum around m/z at 106.08. Before prior to the
ED
thermophysical measurements, water content of PILs were measured and found that water content is less than 3000 ppm, which has tabulated in Table 2.
PT
The density values of the PILs were recorded as a function of temperature range (303.15343.15) K are presented in Table 3 and graphically represented in Fig 1. The density of studied
CE
PILs are in range of (1416.5-1383.5) kg·m-3 and (1231.8-1206.7) kg.m-3 for [DEA][Tfa] and [DEA][Lac] respectively. The experimental density values were fitted by using second order
AC
polynomial equation (1). The fitting parameters are presented in Table 4 along with maximum absolute deviation (MAD) values and absolute relative deviation (ARD) defined by Equation (2). 𝜌 = 𝐴0 + 𝐴1 𝑇 + 𝐴2 𝑇 2 1
𝐴𝑅𝐷 = (𝑛 ∑
(1)
|𝜌𝑒𝑥𝑝 −𝜌𝑐𝑎𝑙 | 𝜌𝑒𝑥𝑝
) 100
(2)
Where A0 and A1 are the fitting parameters obtained from least square analysis, n is the number of data points, ρexp and ρcal are the experimental and calculated density values.
6
ACCEPTED MANUSCRIPT As illustrated in Fig 1, the density of the PILs decreases linearly with temperature also it has been observed that the density of [DEA][Tfa] is higher than [DEA][Lac] and it may be due the presence of the CF3 group which cause better hydrogen bonding with the cation moiety and results more close packed structure [26]. This clearly designates that a better arrangement of ions takes place in the former ionic liquid which allow a greater number of ions per unit volume is located whereas may be slightly higher molecular volume of lactate than trifluoroacetate anion
T
results its lower density. The experimental density value of [DEA][Lac] also have been
IP
compared with literature data presented by Kurnia et. al [27] and reported in Table 3. It has been
CR
found that the experimental density of [DEA][Lac] shows maximum deviation ≥ 2% with the literature, it may be due to the experimental procedure and sample purity.
US
To explore the further understanding of intermolecular interaction, thermal expansion coefficient ( was calculated using the experimental values of density (ρ) by using the Equation
AN
(3) 1 𝜕𝜌
𝛼=− ( ) 𝜌 𝜕𝑇
(3)
M
𝑃
ED
where is density; P is pressure and T is temperature. The calculated thermal expansion coefficient values are given in Table 3. The coefficient
PT
of thermal expansion of [DEA][Tfa] is higher compere to [DEA][Lac] . For both the PILs, the behaviour of thermal expansion coefficient is almost same that is there is little deviation of
CE
value with temperature. Chhotaray et. al has also reported similar kind of variation of thermal
AC
expansion coefficient for lactam class of PILs [28]. Further, the experimental density values were used for the calculation of other important thermodynamic properties including molar volume (Vm), standard entropy (So) by using the equations (4) and (5) respectively. 𝑀
𝑉𝑚 = 𝜌.𝑁
𝐴
𝑆 𝑜 = 1246.5 𝑉𝑚 + 29.5
(4) (5)
Where NA is Avogadro’s number, Vm is molar volume and ρ is density of the PILs at 303.15 K. 7
ACCEPTED MANUSCRIPT Molar volume of both the ILs is calculated at 303.15 K temperature. As shown in Table 2, [DEA][Lac] has slightly higher (0.003 nm3) molar volume than [DEA][Tfa] due to the CH3 and OH groups in lactate anion. Entropy is a measurement of randomness or disorderdness of molecules and as in general it increases with increase in molar volume that is with the increase in size of the molecule. This fact is mirrored in standard entropy values given in Table 2. In order to predict the relative stabilities of the ILs, Glasser developed a new method to
IP
T
calculate the lattice potential energy (Upot) for MX type of simple salt by using Equation (6) which only requires the chemical formula, charges of the ions and the density (or molecular
CR
volume) of the studied materials [29]. In this method there is no involvement of the structural properties, the main factor here is the columbic interaction which contributes to the lattice
US
energy. 𝜌 ⅓
(6)
AN
𝑈𝑝𝑜𝑡 = 𝛾 (𝑀) + 𝛿
Where γ and δ are the fitting coefficients for MX type (1:1) salt and their values are 19181.2
M
kJ·mol-1 and 103.8 kJ·mol-1 respectively. The lattice potential energy was calculated with the help of Equation (6) and the obtained values (at temperature 303.15 K) are listed in Table 2.
ED
Lattice potential energy actually depends on the electrostatic or columbic interaction which is inversely related to the volume of ions (which they are consisting). That is why the lattice
PT
potential energy value of [DEA][Lac] is higher than [DEA][Tfa].
CE
The experimental density data has been correlated through estimated densities obtained from the Gardas and Coutinho model [22] by Equation (7) 𝑀
AC
𝜌𝑐𝑎𝑙 = 𝑁·𝑉
𝑚 (𝑎+𝑏·𝑇+𝑐·𝑃)
(7)
Where ρcal is the calculated density in kg·m-3, M is the molar mass in kg·mol-1, N is the Avogadro’s number, Vm is the molecular volume in nm3, T is the temperature in K and P is the pressure in MPa. As it can be in Fig 2, in both PILs the calculated density (ρcal) using the Gardas and Coutinho model display a well agreement with the experimental density (ρexp) and were related by cal = (0.9906±0.0001)exp with 95% confidence level and regression factor R2 = 0.9996.
8
ACCEPTED MANUSCRIPT Expression of molecular volume in terms of the volume of its constituting ions (Equation 8) was suggested by Esperanca et al. [30]. 𝑉𝑚∗ = 𝑉𝑐∗ + 𝑉𝑎∗
(8)
Where 𝑉𝑚∗ , 𝑉𝑐∗ and 𝑉𝑎∗ are the molecular volume of PILs, molecular volume of cation and molecular volume of anion respectively. By using this approach, for a given ionic liquid it is
T
possible to determine the molar volume of cation knowing the effective size of anion and vice-
IP
versa. The obtained molar volume of anions and diethanolammonium cation are reported in
CR
Table 5.
The data obtained from ultrasonic sound velocity measurement over a wide range of
US
temperature of any substance give us information about its nature and also help to understand intermolecular interaction. Experimental speed of sound data as function of temperature at
AN
atmospheric pressure of both PILs is given in Table 3. Speed of sound also shows a linear dependency with temperature and they are shown in Fig 3. It was found that speed of sound in [DEA][Lac] is higher than that of [DEA][Tfa] which is consistent of the density values of both
M
the PILs. Another significant thermodynamic parameter, isentropic compressibility factor (s)
ED
was calculated with the help of Newton-Laplace Equation (9) 1
(9)
PT
𝑠 = 𝜌.𝑢2
Where ρ and u are the density and sound velocity respectively. As shown in Table 2, the
CE
magnitude of isentropic compressibility (s) increases with increase in temperature for the PILs. Among studied PILs, [DEA][Tfa] shows higher isentropic compressibility value and is supported
AC
by the intermolecular free length values given in Table 6. In order to comprehend the intermolecular interaction in liquid, Jacobson proposed an empirical equation [31] for calculating intermolecular free length (Lf). According to which Lf is given by Equation [10]. 𝐿𝑓 = 𝑘 · √𝛽𝑠
(10)
Where k is a temperature dependent variable called as Jacobson’s constants and the values of k at temperature T = (303.15, 313.15, 323.15 K) are 631, 642 and 652 respectively. The values of Lf 9
ACCEPTED MANUSCRIPT in Å for both the PILs are tabulated in Table 6. In accordance with Eryring’s liquid state theory, the excited acoustic wave generated in the fluid need a time interval to go through intermolecular length of one molecule to another one. Thus, the transmission time of the acoustic wave rises with increase of intermolecular free length (Lf) value or in other way with decrease in speed of sound value in that fluid [9,11,24]. In our experiment, [DEA][Tfa] shows higher Lf values with respect to [DEA][Lac]. This may account for better efficient packing in the [DEA][Tfa] compare
T
to in [DEA][Lac]. This above fact can also be validated by density and speed of sound values
IP
listed in Table 3 for studied PILs.
CR
The thermal stability of the PILs was determined by thermogravimetric analysis (TGA) and has presented in Fig 4. It indicates that our synthesized ionic liquids are less stable compared
US
to pyrrolidonium bisulphate ILs as shown by Panda et al [25]. The stability of the ionic liquids depends upon the strength of the intermolecular interaction between the cations and the anions
AN
which they are consist of and also on possibility of back proton transfer and amount of moisture in PILs.
M
The temperature correspond to 5% weight loss during the heating scan was used as decomposition temperature (Td), Td values are registered in Table 1. Fig 4 shows the TG curves
ED
of the studied PILs on heating rate 10 ˚C·min-1. The decomposition temperature (Td) of [DEA][Tfa] (Td = 443.6 K) is higher than that of [DEA][Lac] (Td = 413.8 K) which clearly
PT
indicates higher thermal stability of [DEA][Tfa] than [DEA][Lac] .
CE
DSC is a thermo-analytical technique in which the difference in amount of heat required to increase the temperature of the sample and the reference is measured as a function of
AC
temperature. DSC therefore discovers changes in heat content. DSC thermogram of the studied PILs is given in Fig 5 and the glass transition temperature (Tg) of both the studied PILs are listed in Table 2 which can be defined as the midpoint of a small change in heat flow on heating from amorphous glass to liquid state. The magnitude of glass transition temperature (Tg) epitomizes the cohesiveness of the studied IL’s and so low value of Tg indicates low cohesive energy [32]. Zhou et al. [33] suggested that glass transition temperature (Tg) is mainly ruled by interionic interaction, ionic size, flexibility and polarizibility of anion. Fig 5 shows DSC curve of studied PILs and indicates no crystallization takes place in the studied temperature range. Both the PILs are liquid at room temperature and show a very low glass transition temperature, Tg [34], 10
ACCEPTED MANUSCRIPT suggests the amorphous polymeric nature of the studied PILs. Also, it has also observed that Tg of lactate anion is higher (~ -55℃) compare to that of smaller sized trifluoro acetate anion (~65℃). This is mainly attributed due to lower electrostatic interaction in the [DEA][Lac]. Finally, the synthesized PILs have possess the combinations of properties, which makes them applicable in numerous applications such as CO2 and SO2 gas absorption and in various organic and bio-
AC
CE
PT
ED
M
AN
US
CR
IP
T
organic reactions etc. [35, 36].
11
ACCEPTED MANUSCRIPT 4. Conclusions In this work, protic ionic liquids having common diethanolammonium cation and different anions (trifluoro acetate or lactate) synthesized and characterised using 1H NMR, HRMS and FTIR spectral techniques. The important thermophysical properties such as density and speed of sound were measured in the temperature range from 303.15 to 343.15 K. The
T
density of both PILs was also predicted through Gardas and Coutinho model and it was found to
IP
be in good agreement with the experimental density values. Further, the experimental density values have been used to calculate thermal expansion coefficient, isentropic compressibility and
CR
molar volume of the studied PILs. Furthermore, thermal stabilities of the PILs were investigated and found that [DEA][Tfa] is thermally more stable compared to [DEA][Lac]; whereas, glass
AN
US
transition temperature of [DEA][Tfa] has found to be lower than [DEA][Lac].
Acknowledgement
M
Authors would like to acknowledge IIT Madras for the financial support through grant number
ED
CHY/15-16/833/RFIR/RAME. GS and DS are thankful to and University Grants Commission (UGC), India, and Council of Scientific and Industrial Research (CSIR), India respectively for
PT
the financial support in the form of Senior Research Fellowship (SRF).
CE
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at www.
AC
Elsevier.com
12
ACCEPTED MANUSCRIPT References [1]
T. Endo, H. Murata, I. Mamoru, H. Seki, K. Nishikawa, J. Phys. Chem. B 116 (2012) 3780–3788.
[2]
R. P. Swatloski, S.K. Spear, J.D. Holbrey, R.D. Rogers, J. Am. Chem. Soc. 124 (2002) 4974–4975. D. Singh, R.L. Gardas, J .Phys. Chem. B 120 (2016) 4834–4842.
[4]
K. J. Jisha, D. Singh, G. Sharma, R. L. Gardas, J. Mol. Liq. 231 (2017) 213–219.
[5]
T. Welton, Chem. Rev. 99 (1999) 2071–2083.
[6]
V. Singh, G. Sharma, R. L. Gardas, PLoS ONE 10 (2015) e0126091.
[7]
T. L. Greaves, C. J. Drummond, Chem. Rev. 108 (2008) 206-237.
[8]
T. L. Greaves, A. Weerawardena, I. Krodkiewska, C.J. Drummond, J. Phys. Chem. B 112
CR
US
(2008) 896–905.
P. K. Chhotaray, R. L. Gardas, J. Chem. Thermodyn. 72 (2014) 117–124.
AN
[9]
IP
T
[3]
[10] J. Stoimenovski, E.I. Izgorodina, D.R. MacFarlane, Phys. Chem. Chem. Phys. 12 (2010)
M
10341–10347.
[11] G. Sharma, R. L. Gardas, A. Coronas, G. Venkatarathnam, Fluid Phase Equilib. 415
ED
(2016) 1–7.
[12] V. P. Petrovic, D. Simijonovic, M. N. Zivanovic, J. V. Kosaric, Z. D. Petrovic, S.
PT
Markovic, S. D. Markovic, RSC Adv. 4 (2014) 24635–24644. [13] A. Khan, X. Lu, L. Aldous, C. Zhao, J. Phys. Chem. C 117 (2013) 18334–18342.
CE
[14] Z. D. Petrovic, L. Comic, O. Stefanovic, D. Simijonovic, V. P. Petrovic, J. Mol. Liq. 170 (2012) 61–65.
AC
[15] H. Meng, C. T. Ge, N. N. Ren, W. Y. Ma, Y. Z. Lu, C. X. Li, Ind. & Eng. Chem. 53 (2014) 355–365.
[16] J. G. Lu, Z. Y. Lu, L. Gao, S. Cao, J. T. Wang, X. Gao, Y. Q. Tang, W. Y. Tan, J. Mol. Liq. 211 (2015) 1–6. [17] S. K. Deraman, N. S. Mohamed, R.H.Y. Subban, Int. J. Elect. Sci. 8 (2013) 1459–1468. [18] F. Falcioni, H. R. Housden, Z. Ling, S. Shimizu, A. J. Walkera and N. C. Bruce, Chem. Commun. 46 (2010) 749–751. [19] J. Luo, O. Conrad, I .F. J. Vankelecom, J. Mater. Chem. 22 (2012) 20574-20579.
13
ACCEPTED MANUSCRIPT [20] H. Tokuda, K. Hayamizu, K. Ishii, M. A. B. H. Susan, M. Watanabe, J. Phys. Chem. B 108 (2004) 16593–16600. [21] T. L. Greaves, C. J. Drummond, Chem. Rev. 115 (2015) 11379–11448. [22] R. L. Gardas, J.A.P. Coutinho, Fluid Phase Equilib. 263 (2008) 26–32. [23] S. Marcin, H.C. Corey, B.W. Timothy, S. Singh, A.S. Vincek, K. Kirichenko, A. R. Katritzky, D.R.R. Robin, Chem. Eur. J. 16 (2010) 1572–1584.
T
[24] B. K. Chennuri, V. Losetty, R. L. Gardas, J. Mol. Liq. 212 (2015) 444-450.
IP
[25] S. Panda, S. Ray, V. Losettey, R.L. Gardas, J. Mol. Liq. 224 (2016) 882–892.
CR
[26] Z. Ullah, M.A. Bustam, Z. Man, N. Muhammad, A.S. Khan, RSC Adv. 5 (2015) 71449– 71461.
[27] K. A. Kurnia, C.D. Wilfred, T. Murugesan, J. Chem. Thermodynamics 41 (2009) 517–521.
US
[28] P. K. Chhotaray, S. Jella, R. L. Gardas, J. Chem. Thermodyn. 74 (2014) 255–262. [29] L. Glasser, Themochim. Acta 21 (2004) 87–93.
AN
[30] J. M. M. S. Esperanca, H.J.R. Guedes, M. Blesic, L.P.N. Rebelo, J. Chem. Eng. 51 (2006) 237–241.
M
[31] B. Jacobson, J. Chem. Phys. 20 (1952) 927–931.
[32] A. Tzani, M. Elmaloglou, C. Kyriazis, D. Aravopoulou, I. Kleidas, A. Papadopoulos, E.
ED
Ioannou, A. Kyritsis, E. Voutsas, A. Detsi. J. Mol. Liq. 224 (2016) 366–376. [33] Z. B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Eur. J. 11 (2005) 752–766.
(2012) 4276–4285.
PT
[34] S. P. Verevkin, N. Vladimir, E. Yanenko, D.H. Zaitsau, R.V. Ralys, J. Phys. Chem. B 116
CE
[35] A. Zhu, R. Liu, L. Li, L.Wang, J. Wang, Catal. Today 200 (2013) 17–23. [36] V. H. Alverez, N. Dosil, R. G. Cabalerio, S. Mattedi, M. M. Pastor, M. Iglesias, J. Chem.
AC
Eng. Data 55 (2010) 625–632.
14
ACCEPTED MANUSCRIPT Figure Captions Scheme 1. Synthesis of diethanolamine based PILs.
Figure 1.
Density of studied protic ionic liquids as function of temperature from (303.15343.15) K. ∎, [DEA][Tfa]; ●, [DEA][Lac]; the symbols represent experimental values
Correlation between experimental densities and predicted densities. , [DEA][Tfa]; ,
IP
Figure 2.
T
and solid line represent the linear fitting.
Density of studied protic ionic liquids as function of temperature from (303.15- 343.15) K. ∎, [DEA][Tfa];
US
Figure 3.
CR
[DEA][Lac].
●, [DEA][Lac]; the symbols represent experimental values of speed
AN
of sounds.
Loss of weight percent against temperature of PILs. (a) [DEA][Lac] (b) [DEA][Tfa].
Figure 5.
Differential scanning calorimeter plot, Heat flow as function of temperature (a)
ED
M
Figure 4.
AC
CE
PT
[DEA][Tfa] (b) [DEA][Lac].
15
ACCEPTED MANUSCRIPT
Table 1 Name of the chemicals, synthesized protic ionic liquids and their purity.
N,N-diethanolammine
2
Trifluoroacetic acid
3
Lactic acid
4
Methanol
Tfa
≥ 99%
Lac
≥ 98%
-
≥ 99%
[DEA][Tfa]
≥99% a
[DEA][Lac]
≥99% a
(CAS No. 76-05-1) Alfa Aesar (CAS No. 50-21-5) Sigma-Aldrich
(CAS No. 67-56-1)
NU
Synthesized
MA
N,Ndiethanolammonium
D
Synthesized
CE
PT E
From the 1H NMR spectroscopic technique
AC
a
≥ 98%
Sigma-Aldrich
diethanolammonium
Lactate
DEA
(CAS No. 111-42-2)
trifluoroacetate
6
Purity
Sigma-Aldrich
N,N5
Abbreviation
PT
1
Source
RI
Chemicals
SC
S. No.
16
ACCEPTED MANUSCRIPT Table 2 Molecular weight (M), water content, decomposition temperature (Td) at 5% weight loss, glass transition temperature (Tg), standard entropy (S°) and lattice potential energy (Upot) of the studied protic ionic liquids. Water content ppm
M g.mol-1
Td K
Tg K
a ° S -1
J.K .mol-1
a
Upot kJ.mol-1
2920
443.6
208.0
[DEA][Lac] 413.8
218.0
355.2
354.8
CE
PT E
D
MA
NU
at 303.15 K
AC
a
2880
357.4
SC
195.21
348.6
RI
219.16
PT
[DEA][Tfa]
17
ACCEPTED MANUSCRIPT Table 3 experimental densities (ρ), speed of sound (u), thermal expansion coefficient (), isentropic compressibility (s) of the studied IL’s in the temperature range from T = (303.15 to 343.15) K. T
u
·
s·102
K
kgm-3
ms-1
Pa-1
5.88
3.37
5.86
3.43
5.88
3.51
5.89
3.57
5.87
3.64
5.86
3.71
[DEA][Tfa] 1416.5
1447.5
308.15
1412.3
1434.7
313.15
1408.1
1422.6
318.15
1404.0
1410.9
323.15
1399.9
1399.7
328.15
1395.7
1388.7
333.15
1391.7
1378.1
5.85
3.78
338.15
1387.6
1367.4
5.85
3.85
343.15
1383.5
1357.0
5.95
3.92
1880.3
5.08
2.29
1842.8
5.09
2.39
1813.9
5.10
2.47
1791.1
5.10
2.54
1771.9
5.13
2.61
1755.3
5.15
2.66
1740.1
5.38
2.72
1726.1
5.14
2.77
1712.7
5.21
2.82
MA
NU
SC
RI
PT
303.15
[DEA][Lac]
1197.8a 1228.7
PT E
308.15
1225.6
313.15
328.15 333.15 338.15 343.15
1222.5
CE
323.15
1191.9a
1219.3
AC
318.15
D
1231.8
303.15
1185.8a 1216.2 1213.1 1179.6a 1209.8 1206.7 1173.5a
Standard uncertainties are u (ρ) = 0.1 kg∙m-3, u (u) = 0.5 m∙s-1, u (T) = 0.01 K. a Ref. [27]
18
ACCEPTED MANUSCRIPT Table 4 Fitting parameters, average relative deviation (ARD) and maximum average deviation (MAD) values of densities of studied PILs. AO kg·m-3
A1 kg·m-3·K-1
A2103 kg·m-3·K-2
ARD
MAD
[DEA][Tfa]
1702.30
-1.048
0.346
0.003
0.006
[DEA][Lac]
1395.16
-0.460
0.260
0.002
0.007
AC
CE
PT E
D
MA
NU
SC
RI
PT
PILs
19
ACCEPTED MANUSCRIPT Table 5 Effective molar volume of cation (𝑉𝑐∗ ), anion (𝑉𝑎∗ ), 𝑉𝑚∗ obtained using Gardas and Coutinho model (Equation 8) and estimated molar volume (Vm) obtained from Equation 4 at
𝑉𝑐∗ Å3
𝑉𝑎∗ Å3
𝑉𝑚∗ Å3
Vm Å3
% RDa
[DEA][Tfa]
115.4
141.0
256.4
256.8
0.5
[DEA][Lac]
115.4
146.4
261.8
0.2
RI
CE
PT E
D
MA
NU
SC
Relative deviation (RD); 𝑅𝐷 = (𝑉𝑚 − 𝑉𝑚∗ ) × 100⁄𝑉𝑚
263.1
AC
a
PILs
PT
303.15 K and 0.1MPa.
20
ACCEPTED MANUSCRIPT Table 6 Experimental densities (ρ), molecular radius (r) and intermolecular length (Lf) of studied PIL’s at 303.15 K and 0.1MPa. ρ kgm-3
r Å
Lf Å
[DEA][Tfa]
1416.5
2.453
0.314
[DEA][Lac]
1231.8
2.471
0.381
AC
CE
PT E
D
MA
NU
SC
RI
PT
PILs
21
Fig 1.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
22
Fig 2.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
23
Fig 3.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
24
Fig 4.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
25
Fig 5.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
26
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
Graphical abstract
27
ACCEPTED MANUSCRIPT
Research Highlights:
N,N-diethanolammonium based protic ionic liquids (PILs) have been synthesized. Several thermophysical properties of PILs were measured as function of temperature.
PT
Density of PILs was predicted using Gardas and Coutinho model.
RI
Experimental and derived properties used to analyse electrostatic and molecular
AC
CE
PT E
D
MA
NU
SC
interactions in PILs.
28
Anirban Sarkar, Gyanendra Sharma, Dharmendra Singh, Ramesh L. Gardas PII: DOI: Reference:
S0167-7322(17)31718-X doi: 10.1016/j.molliq.2017.07.025 MOLLIQ 7602
To appear in:
Journal of Molecular Liquids
Received date: Revised date: Accepted date:
21 April 2017 10 June 2017 8 July 2017
Please cite this article as: Anirban Sarkar, Gyanendra Sharma, Dharmendra Singh, Ramesh L. Gardas , Effect of anion on thermophysical properties of N,N-diethanolammonium based protic ionic liquids, Journal of Molecular Liquids (2017), doi: 10.1016/ j.molliq.2017.07.025
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Effect of anion on thermophysical properties of N,N-diethanolammonium based protic ionic liquids Anirban Sarkar, Gyanendra Sharma, Dharmendra Singh and Ramesh L. Gardas* Indian Institute of Technology Madras, Chennai 600036 *Corresponding author. Tel.: +91 44 2257 4248; fax: +91 44 2257 4202 E.mail address: [email protected]; URL: http://www.iitm.ac.in/info/fac/gardas
IP
T
----------------------------------------------------------------------------------------------------------------
CR
Abstract
In this work, we investigated the influence of two different anions on the various thermophysical
US
properties of N,N-diethanolammonium based protic ionic liquids (PILs). For this purpose, two N,N-diethanolammonium based PILs have been synthesised by using trifluoroacetate and lactate
AN
as counter anions and characterized through spectral techniques. The thermophysical properties such as density and speed of sound were measured as function of temperature from (303.15 to 343.15) K at atmospheric pressure. The experimental density values used to calculate thermal
M
expansion coefficient (), isentropic compressibility (s), molar volume (Vm) and standard
ED
entropy (So). The thermal stabilities of the PILs were investigated by thermogravimetric analysis (TGA), and phase transition measurement was done by using differential scanning calorimerty
PT
(DSC). Further, to understand the interionic strength of PILs, lattice potential energy (Upot) has been calculated. The experimental densities have been compared with the estimated densities
CE
obtained from Gardas and Coutinho model and obtained results are in good agreement. Further, the density values of studied protic ionic liquids have been compared with other protic ionic
AC
liquids. Finally, the impact of structure variants in terms of the carboxylate anions has been analyzed over the thermophysical properties of N,N-diethenolammonium based PILs.
Keywords: Ionic liquid; thermal stability; isothermal expansion; isentropic compressibility; Gardas and Coutinho model.
1
ACCEPTED MANUSCRIPT 1. Introduction Ionic liquids (ILs) are molten salts with melting point below 100˚C and most of them are liquid even below room temperature. ILs have several exceptional properties such as low volatility, extremely low flammability, high specific solubility, wide liquid range, high electrical conductivity and good thermal and chemical stability [1– 6]. Combination of these properties
T
makes them an interesting class of substances for both industries as well as academia. Generally,
IP
ILs can be divided into two categories namely protic ionic liquids (PILs) and aprotic ionic liquids (AILs). Much research has been done on both protic as well as aprotic ionic liquids and
CR
owing to the above specified properties they have several potential applications. In contrast to AILS, the PILs possess additionally hydrogen bond capacity which is significant for many
US
biological processes and as a medium of proton conduction for polymer membrane fuel cells [7]. PILs can be synthesised by Brönsted acid-base neutralization reaction in that proton transfer
AN
takes place from acid to base. There is a chance of equilibrium between reactant and products due to incomplete proton transfer [8, 9], which can be avoid by selecting proper acid and base
M
pair which is having pKa difference (ΔpKa ) greater than 4 [10]. The hydroxylammonium based PILs have many advantageous applications due to their hydrogen bond donation ability [11]. In
ED
recent years, research on ammonium based PILs have been focussed for various applications, such as organic synthesis, oxygen reduction reaction, antimicrobial activity, extraction of
PT
complexes, CO2 gas absorption, separation process and polymer dissolution etc. [12-19]. However, before exploring PILs for these applications, it is essential to study the thermophysical
CE
properties of PILs. The possibility of tailoring the properties of an ionic liquid to meet the requirements of some specific application by the tuning of combinations of cation and anion.
AC
Therefore, the knowledge of accurate thermophysical properties of ILs are essential, and also for the design of processes and products involving these compounds [20–22]. For scrutinizing the intermolecular interactions between ions, there are several reports available in which effect of structure variant on thermophysical properties were studied [11, 23]. However, to the best of our information there are few reports available, which focussed the influence of structure variant on thermophysical properties of PILs. In this framework, we have synthesized two N,N-diethanolammonium (DEA) based PILs with two different counter anions (trifluoroacetic acid [Tfa] and lactic acid [Lac]) namely N,N-
2
ACCEPTED MANUSCRIPT diethanolammonium trifluoroacetate ([DEA][Tfa]) and N,N-diethanolammonium lactate ([DEA][Lac]) and characterized by spectral techniques includes 1H NMR, HRMS and IR spectral techniques. Thermal decomposition temperature (Td) and glass transition temperature (Tg) of the studied PILs were evaluated by TGA and DSC experiment respectively. Before studying the various temperature-dependent thermophysical properties such as density and speed of sound in the range (303.15 to 343.15) K the water content should be accounted. The
T
experimental density and speed of sound were used to calculate the industrially relevant
IP
parameters such as thermal expansion coefficient () and isentropic compressibility (s). Further,
CR
the experimental densities of PILs were correlated with the estimated density values obtained from Gardas and Coutinho model. In addition, the calculated and derived thermodynamic properties were scrutinized to investigate the influence of anion structure of the PILs and
AN
US
compared with other PILs [11, 24-25].
M
2. Experimental Section 2.1 Materials
ED
All reagents used in this experiment were Analytical Reagent (AR) grade and used without
presented in Table 1.
PT
further purification. The name of the used chemicals, source and their purity in mass fraction are
CE
2.2 Synthesis of ionic liquids
The PILs were synthesized by Brönsted acid-base neutralization method where proton
AC
transfer takes place from acid to base. Here, we used diethanolamine (DEA) as a base and methanol as a solvent, trifluoroacetic acid and lactic acid have been used as acid. The methanolic solution of DEA was taken in a two-necked round bottom flask which is equipped with a condenser and dropping funnel. The whole set-up was placed on an ice–bath to maintain low temperature (~5 C). The equimolar amount of acid was added drop wise to the above reaction mixture using dropping funnel. After the complete addition of acid, the reaction mixture was stirred with the help of a magnetic stirrer for next 24 hrs under nitrogen atmosphere and at room temperature. After completion, the methanol (solvent) was removed from the reaction mixture by putting it into rotary evaporator at ~50 C for 2 hrs. In order to remove moisture and the 3
ACCEPTED MANUSCRIPT resultant mixture (PIL) was dried for 36 hrs under high vacuum and kept in nitrogen atmosphere. As shown in Scheme 3.1, the syntheses of studied PILs were achieved by the equimolar mixture of acid and base, without any by-product formation.
Alkyl carboxylic acid
Protic ionic liquid
T
Diethanolammine
2.3 Characterization of PILs 1
H NMR (500 MHz, CDCl3) δ 3.248 (t, 4H); 3.877 (t, 4H).
AN
[DEA][Tfa]
CR
US
Scheme 1. Synthesis of diethanolamine based PILs
IP
R = -CF3 and -CH3CH(OH)
[DEA][Lac]
1
M
HRMS (ESI+) m/z for [M]+ calcd 106.0868 and found 106.0850. H NMR (500 MHz, CDCl3) δ 1.338 (t, 3H); 3.253 (m, 4H); 3.877 (m, 4H); 4.108
ED
(m, 1H).
CE
IR Spectroscopy
PT
HRMS (ESI+) m/z for [M]+ calcd 106.0868 and found 106.0861.
IR spectra were obtained on a JASCO FT/IR 4100 spectrometer which has maximum
AC
resolution of 0.9 cm-1 and 22000:1 signal to noise ratio by using KBr disk. In IR spectra, a broad band in the range of 3500-3000 cm-1 shows ν (O-H), ν (N-H) and also ammonium peak. Peak around 1620 cm-1 indicates the presence of carbonyl stretching ν (C=O) and bending peak of ν (N-H). (see supporting information Fig S3) 2.4 Water Content The water content measurements of the synthesized PILs have been done by using Metrohm 870KF Titrinoplus Karl-Fischer titrator. This instrument operates volumetric titration 4
ACCEPTED MANUSCRIPT principle using dual platinum electrodes that allows the detection of water from less than 10 ppm to 100%. It was calibrated with standard sample. The PILs, [DEA][Tfa] and [DEA][Lac] were dissolved in dried methanol solution and followed by titration against Karl-Fisher reagent, and observed water contents are presented in Table 2.
T
2.5 Thermogravimetric Analysis (TGA)
IP
Thermogravimetric analysis of the synthesized PILs have been carried out by using TA
CR
instrument Hi-Res TGA Q500 with weighing precision of 0.01% from room temperature to 600 C at 10 C min-1 heating rate under nitrogen atmosphere (flow rate is 40 ml·min-1 and 60 ml·minfor the balance and furnace respectively) in an open platinum pan. Two thermocouples are
US
1
positioned immediately adjacent to the sample to ensure simultaneous measurement of sample
AN
temperature and heat rate control accurately and precisely. Nickel metal was used for the calibration of Curie point temperature.
M
2.6 DSC Measurement
ED
The phase transition measurement was carried out by DSC (TA instruments DSC Q200), which have the sensitivity of 0.2 μW and temperature accuracy ±0.1 C. The scanning sequence
PT
involves the freezing the samples to -80 C temperature (up to 20 min) followed by heating the sample to 50 C with a heating rate 20 ˚C·min-1 with flow rate of 50 cm3·min-1. Using this
temperature (Tg).
CE
technique it is possible to detect fusion and crystallization events as well as glass transition
AC
2.6 Density and Speed of sound Temperature dependent density and the speed of sound were measured by Anton Paar (DSA 5000M) instrument which works on the oscillating U-tube principle. The sample was introduced into a borosilicate glass U tube which is electronically vibrated at its characteristic frequency which depends on the density of the sample. The instrument also consists of a stainless steel cell for sound velocity measurement. Both these cells are temperature-controlled by built-in Peltier thermostats (PT-100) which have an accuracy of ± 0.01 K. This system enable us to measure the density from 0 to 3000 kg·m-3 and speed of sound in the range 1000–2000 m·s-1 in a 5
ACCEPTED MANUSCRIPT wide range of pressure and temperature. Other than that sample handling is very convenient here and also the errors of sample filling can be detected automatically. As a result of these instrumental properties, we get an accurate and reproducible result of measurement. The uncertainty in the measurement of density was ± 0.1 kg·m-3 and that of speed of sound was ± 0.5
IP
T
m·s-1.
CR
3. Results and Discussion
US
The syntheses of the ILs were achieved by the reaction of diethanolammine with equimolar amount of acid and the formation of the PILs has been confirmed by 1H, and IR
AN
spectra. Though, the synthesis of studied PILs was achieved by equimolar mixture of acid and base, so there is no change of any side product formation. Further, the purity of synthesized PILs
M
was analyzed through HRMS spectrum and as shown in Figure S4 and S5, we get appropriate cationic peak of synthesized PILs in HRMS spectrum around m/z at 106.08. Before prior to the
ED
thermophysical measurements, water content of PILs were measured and found that water content is less than 3000 ppm, which has tabulated in Table 2.
PT
The density values of the PILs were recorded as a function of temperature range (303.15343.15) K are presented in Table 3 and graphically represented in Fig 1. The density of studied
CE
PILs are in range of (1416.5-1383.5) kg·m-3 and (1231.8-1206.7) kg.m-3 for [DEA][Tfa] and [DEA][Lac] respectively. The experimental density values were fitted by using second order
AC
polynomial equation (1). The fitting parameters are presented in Table 4 along with maximum absolute deviation (MAD) values and absolute relative deviation (ARD) defined by Equation (2). 𝜌 = 𝐴0 + 𝐴1 𝑇 + 𝐴2 𝑇 2 1
𝐴𝑅𝐷 = (𝑛 ∑
(1)
|𝜌𝑒𝑥𝑝 −𝜌𝑐𝑎𝑙 | 𝜌𝑒𝑥𝑝
) 100
(2)
Where A0 and A1 are the fitting parameters obtained from least square analysis, n is the number of data points, ρexp and ρcal are the experimental and calculated density values.
6
ACCEPTED MANUSCRIPT As illustrated in Fig 1, the density of the PILs decreases linearly with temperature also it has been observed that the density of [DEA][Tfa] is higher than [DEA][Lac] and it may be due the presence of the CF3 group which cause better hydrogen bonding with the cation moiety and results more close packed structure [26]. This clearly designates that a better arrangement of ions takes place in the former ionic liquid which allow a greater number of ions per unit volume is located whereas may be slightly higher molecular volume of lactate than trifluoroacetate anion
T
results its lower density. The experimental density value of [DEA][Lac] also have been
IP
compared with literature data presented by Kurnia et. al [27] and reported in Table 3. It has been
CR
found that the experimental density of [DEA][Lac] shows maximum deviation ≥ 2% with the literature, it may be due to the experimental procedure and sample purity.
US
To explore the further understanding of intermolecular interaction, thermal expansion coefficient ( was calculated using the experimental values of density (ρ) by using the Equation
AN
(3) 1 𝜕𝜌
𝛼=− ( ) 𝜌 𝜕𝑇
(3)
M
𝑃
ED
where is density; P is pressure and T is temperature. The calculated thermal expansion coefficient values are given in Table 3. The coefficient
PT
of thermal expansion of [DEA][Tfa] is higher compere to [DEA][Lac] . For both the PILs, the behaviour of thermal expansion coefficient is almost same that is there is little deviation of
CE
value with temperature. Chhotaray et. al has also reported similar kind of variation of thermal
AC
expansion coefficient for lactam class of PILs [28]. Further, the experimental density values were used for the calculation of other important thermodynamic properties including molar volume (Vm), standard entropy (So) by using the equations (4) and (5) respectively. 𝑀
𝑉𝑚 = 𝜌.𝑁
𝐴
𝑆 𝑜 = 1246.5 𝑉𝑚 + 29.5
(4) (5)
Where NA is Avogadro’s number, Vm is molar volume and ρ is density of the PILs at 303.15 K. 7
ACCEPTED MANUSCRIPT Molar volume of both the ILs is calculated at 303.15 K temperature. As shown in Table 2, [DEA][Lac] has slightly higher (0.003 nm3) molar volume than [DEA][Tfa] due to the CH3 and OH groups in lactate anion. Entropy is a measurement of randomness or disorderdness of molecules and as in general it increases with increase in molar volume that is with the increase in size of the molecule. This fact is mirrored in standard entropy values given in Table 2. In order to predict the relative stabilities of the ILs, Glasser developed a new method to
IP
T
calculate the lattice potential energy (Upot) for MX type of simple salt by using Equation (6) which only requires the chemical formula, charges of the ions and the density (or molecular
CR
volume) of the studied materials [29]. In this method there is no involvement of the structural properties, the main factor here is the columbic interaction which contributes to the lattice
US
energy. 𝜌 ⅓
(6)
AN
𝑈𝑝𝑜𝑡 = 𝛾 (𝑀) + 𝛿
Where γ and δ are the fitting coefficients for MX type (1:1) salt and their values are 19181.2
M
kJ·mol-1 and 103.8 kJ·mol-1 respectively. The lattice potential energy was calculated with the help of Equation (6) and the obtained values (at temperature 303.15 K) are listed in Table 2.
ED
Lattice potential energy actually depends on the electrostatic or columbic interaction which is inversely related to the volume of ions (which they are consisting). That is why the lattice
PT
potential energy value of [DEA][Lac] is higher than [DEA][Tfa].
CE
The experimental density data has been correlated through estimated densities obtained from the Gardas and Coutinho model [22] by Equation (7) 𝑀
AC
𝜌𝑐𝑎𝑙 = 𝑁·𝑉
𝑚 (𝑎+𝑏·𝑇+𝑐·𝑃)
(7)
Where ρcal is the calculated density in kg·m-3, M is the molar mass in kg·mol-1, N is the Avogadro’s number, Vm is the molecular volume in nm3, T is the temperature in K and P is the pressure in MPa. As it can be in Fig 2, in both PILs the calculated density (ρcal) using the Gardas and Coutinho model display a well agreement with the experimental density (ρexp) and were related by cal = (0.9906±0.0001)exp with 95% confidence level and regression factor R2 = 0.9996.
8
ACCEPTED MANUSCRIPT Expression of molecular volume in terms of the volume of its constituting ions (Equation 8) was suggested by Esperanca et al. [30]. 𝑉𝑚∗ = 𝑉𝑐∗ + 𝑉𝑎∗
(8)
Where 𝑉𝑚∗ , 𝑉𝑐∗ and 𝑉𝑎∗ are the molecular volume of PILs, molecular volume of cation and molecular volume of anion respectively. By using this approach, for a given ionic liquid it is
T
possible to determine the molar volume of cation knowing the effective size of anion and vice-
IP
versa. The obtained molar volume of anions and diethanolammonium cation are reported in
CR
Table 5.
The data obtained from ultrasonic sound velocity measurement over a wide range of
US
temperature of any substance give us information about its nature and also help to understand intermolecular interaction. Experimental speed of sound data as function of temperature at
AN
atmospheric pressure of both PILs is given in Table 3. Speed of sound also shows a linear dependency with temperature and they are shown in Fig 3. It was found that speed of sound in [DEA][Lac] is higher than that of [DEA][Tfa] which is consistent of the density values of both
M
the PILs. Another significant thermodynamic parameter, isentropic compressibility factor (s)
ED
was calculated with the help of Newton-Laplace Equation (9) 1
(9)
PT
𝑠 = 𝜌.𝑢2
Where ρ and u are the density and sound velocity respectively. As shown in Table 2, the
CE
magnitude of isentropic compressibility (s) increases with increase in temperature for the PILs. Among studied PILs, [DEA][Tfa] shows higher isentropic compressibility value and is supported
AC
by the intermolecular free length values given in Table 6. In order to comprehend the intermolecular interaction in liquid, Jacobson proposed an empirical equation [31] for calculating intermolecular free length (Lf). According to which Lf is given by Equation [10]. 𝐿𝑓 = 𝑘 · √𝛽𝑠
(10)
Where k is a temperature dependent variable called as Jacobson’s constants and the values of k at temperature T = (303.15, 313.15, 323.15 K) are 631, 642 and 652 respectively. The values of Lf 9
ACCEPTED MANUSCRIPT in Å for both the PILs are tabulated in Table 6. In accordance with Eryring’s liquid state theory, the excited acoustic wave generated in the fluid need a time interval to go through intermolecular length of one molecule to another one. Thus, the transmission time of the acoustic wave rises with increase of intermolecular free length (Lf) value or in other way with decrease in speed of sound value in that fluid [9,11,24]. In our experiment, [DEA][Tfa] shows higher Lf values with respect to [DEA][Lac]. This may account for better efficient packing in the [DEA][Tfa] compare
T
to in [DEA][Lac]. This above fact can also be validated by density and speed of sound values
IP
listed in Table 3 for studied PILs.
CR
The thermal stability of the PILs was determined by thermogravimetric analysis (TGA) and has presented in Fig 4. It indicates that our synthesized ionic liquids are less stable compared
US
to pyrrolidonium bisulphate ILs as shown by Panda et al [25]. The stability of the ionic liquids depends upon the strength of the intermolecular interaction between the cations and the anions
AN
which they are consist of and also on possibility of back proton transfer and amount of moisture in PILs.
M
The temperature correspond to 5% weight loss during the heating scan was used as decomposition temperature (Td), Td values are registered in Table 1. Fig 4 shows the TG curves
ED
of the studied PILs on heating rate 10 ˚C·min-1. The decomposition temperature (Td) of [DEA][Tfa] (Td = 443.6 K) is higher than that of [DEA][Lac] (Td = 413.8 K) which clearly
PT
indicates higher thermal stability of [DEA][Tfa] than [DEA][Lac] .
CE
DSC is a thermo-analytical technique in which the difference in amount of heat required to increase the temperature of the sample and the reference is measured as a function of
AC
temperature. DSC therefore discovers changes in heat content. DSC thermogram of the studied PILs is given in Fig 5 and the glass transition temperature (Tg) of both the studied PILs are listed in Table 2 which can be defined as the midpoint of a small change in heat flow on heating from amorphous glass to liquid state. The magnitude of glass transition temperature (Tg) epitomizes the cohesiveness of the studied IL’s and so low value of Tg indicates low cohesive energy [32]. Zhou et al. [33] suggested that glass transition temperature (Tg) is mainly ruled by interionic interaction, ionic size, flexibility and polarizibility of anion. Fig 5 shows DSC curve of studied PILs and indicates no crystallization takes place in the studied temperature range. Both the PILs are liquid at room temperature and show a very low glass transition temperature, Tg [34], 10
ACCEPTED MANUSCRIPT suggests the amorphous polymeric nature of the studied PILs. Also, it has also observed that Tg of lactate anion is higher (~ -55℃) compare to that of smaller sized trifluoro acetate anion (~65℃). This is mainly attributed due to lower electrostatic interaction in the [DEA][Lac]. Finally, the synthesized PILs have possess the combinations of properties, which makes them applicable in numerous applications such as CO2 and SO2 gas absorption and in various organic and bio-
AC
CE
PT
ED
M
AN
US
CR
IP
T
organic reactions etc. [35, 36].
11
ACCEPTED MANUSCRIPT 4. Conclusions In this work, protic ionic liquids having common diethanolammonium cation and different anions (trifluoro acetate or lactate) synthesized and characterised using 1H NMR, HRMS and FTIR spectral techniques. The important thermophysical properties such as density and speed of sound were measured in the temperature range from 303.15 to 343.15 K. The
T
density of both PILs was also predicted through Gardas and Coutinho model and it was found to
IP
be in good agreement with the experimental density values. Further, the experimental density values have been used to calculate thermal expansion coefficient, isentropic compressibility and
CR
molar volume of the studied PILs. Furthermore, thermal stabilities of the PILs were investigated and found that [DEA][Tfa] is thermally more stable compared to [DEA][Lac]; whereas, glass
AN
US
transition temperature of [DEA][Tfa] has found to be lower than [DEA][Lac].
Acknowledgement
M
Authors would like to acknowledge IIT Madras for the financial support through grant number
ED
CHY/15-16/833/RFIR/RAME. GS and DS are thankful to and University Grants Commission (UGC), India, and Council of Scientific and Industrial Research (CSIR), India respectively for
PT
the financial support in the form of Senior Research Fellowship (SRF).
CE
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the online version, at www.
AC
Elsevier.com
12
ACCEPTED MANUSCRIPT References [1]
T. Endo, H. Murata, I. Mamoru, H. Seki, K. Nishikawa, J. Phys. Chem. B 116 (2012) 3780–3788.
[2]
R. P. Swatloski, S.K. Spear, J.D. Holbrey, R.D. Rogers, J. Am. Chem. Soc. 124 (2002) 4974–4975. D. Singh, R.L. Gardas, J .Phys. Chem. B 120 (2016) 4834–4842.
[4]
K. J. Jisha, D. Singh, G. Sharma, R. L. Gardas, J. Mol. Liq. 231 (2017) 213–219.
[5]
T. Welton, Chem. Rev. 99 (1999) 2071–2083.
[6]
V. Singh, G. Sharma, R. L. Gardas, PLoS ONE 10 (2015) e0126091.
[7]
T. L. Greaves, C. J. Drummond, Chem. Rev. 108 (2008) 206-237.
[8]
T. L. Greaves, A. Weerawardena, I. Krodkiewska, C.J. Drummond, J. Phys. Chem. B 112
CR
US
(2008) 896–905.
P. K. Chhotaray, R. L. Gardas, J. Chem. Thermodyn. 72 (2014) 117–124.
AN
[9]
IP
T
[3]
[10] J. Stoimenovski, E.I. Izgorodina, D.R. MacFarlane, Phys. Chem. Chem. Phys. 12 (2010)
M
10341–10347.
[11] G. Sharma, R. L. Gardas, A. Coronas, G. Venkatarathnam, Fluid Phase Equilib. 415
ED
(2016) 1–7.
[12] V. P. Petrovic, D. Simijonovic, M. N. Zivanovic, J. V. Kosaric, Z. D. Petrovic, S.
PT
Markovic, S. D. Markovic, RSC Adv. 4 (2014) 24635–24644. [13] A. Khan, X. Lu, L. Aldous, C. Zhao, J. Phys. Chem. C 117 (2013) 18334–18342.
CE
[14] Z. D. Petrovic, L. Comic, O. Stefanovic, D. Simijonovic, V. P. Petrovic, J. Mol. Liq. 170 (2012) 61–65.
AC
[15] H. Meng, C. T. Ge, N. N. Ren, W. Y. Ma, Y. Z. Lu, C. X. Li, Ind. & Eng. Chem. 53 (2014) 355–365.
[16] J. G. Lu, Z. Y. Lu, L. Gao, S. Cao, J. T. Wang, X. Gao, Y. Q. Tang, W. Y. Tan, J. Mol. Liq. 211 (2015) 1–6. [17] S. K. Deraman, N. S. Mohamed, R.H.Y. Subban, Int. J. Elect. Sci. 8 (2013) 1459–1468. [18] F. Falcioni, H. R. Housden, Z. Ling, S. Shimizu, A. J. Walkera and N. C. Bruce, Chem. Commun. 46 (2010) 749–751. [19] J. Luo, O. Conrad, I .F. J. Vankelecom, J. Mater. Chem. 22 (2012) 20574-20579.
13
ACCEPTED MANUSCRIPT [20] H. Tokuda, K. Hayamizu, K. Ishii, M. A. B. H. Susan, M. Watanabe, J. Phys. Chem. B 108 (2004) 16593–16600. [21] T. L. Greaves, C. J. Drummond, Chem. Rev. 115 (2015) 11379–11448. [22] R. L. Gardas, J.A.P. Coutinho, Fluid Phase Equilib. 263 (2008) 26–32. [23] S. Marcin, H.C. Corey, B.W. Timothy, S. Singh, A.S. Vincek, K. Kirichenko, A. R. Katritzky, D.R.R. Robin, Chem. Eur. J. 16 (2010) 1572–1584.
T
[24] B. K. Chennuri, V. Losetty, R. L. Gardas, J. Mol. Liq. 212 (2015) 444-450.
IP
[25] S. Panda, S. Ray, V. Losettey, R.L. Gardas, J. Mol. Liq. 224 (2016) 882–892.
CR
[26] Z. Ullah, M.A. Bustam, Z. Man, N. Muhammad, A.S. Khan, RSC Adv. 5 (2015) 71449– 71461.
[27] K. A. Kurnia, C.D. Wilfred, T. Murugesan, J. Chem. Thermodynamics 41 (2009) 517–521.
US
[28] P. K. Chhotaray, S. Jella, R. L. Gardas, J. Chem. Thermodyn. 74 (2014) 255–262. [29] L. Glasser, Themochim. Acta 21 (2004) 87–93.
AN
[30] J. M. M. S. Esperanca, H.J.R. Guedes, M. Blesic, L.P.N. Rebelo, J. Chem. Eng. 51 (2006) 237–241.
M
[31] B. Jacobson, J. Chem. Phys. 20 (1952) 927–931.
[32] A. Tzani, M. Elmaloglou, C. Kyriazis, D. Aravopoulou, I. Kleidas, A. Papadopoulos, E.
ED
Ioannou, A. Kyritsis, E. Voutsas, A. Detsi. J. Mol. Liq. 224 (2016) 366–376. [33] Z. B. Zhou, H. Matsumoto, K. Tatsumi, Chem. Eur. J. 11 (2005) 752–766.
(2012) 4276–4285.
PT
[34] S. P. Verevkin, N. Vladimir, E. Yanenko, D.H. Zaitsau, R.V. Ralys, J. Phys. Chem. B 116
CE
[35] A. Zhu, R. Liu, L. Li, L.Wang, J. Wang, Catal. Today 200 (2013) 17–23. [36] V. H. Alverez, N. Dosil, R. G. Cabalerio, S. Mattedi, M. M. Pastor, M. Iglesias, J. Chem.
AC
Eng. Data 55 (2010) 625–632.
14
ACCEPTED MANUSCRIPT Figure Captions Scheme 1. Synthesis of diethanolamine based PILs.
Figure 1.
Density of studied protic ionic liquids as function of temperature from (303.15343.15) K. ∎, [DEA][Tfa]; ●, [DEA][Lac]; the symbols represent experimental values
Correlation between experimental densities and predicted densities. , [DEA][Tfa]; ,
IP
Figure 2.
T
and solid line represent the linear fitting.
Density of studied protic ionic liquids as function of temperature from (303.15- 343.15) K. ∎, [DEA][Tfa];
US
Figure 3.
CR
[DEA][Lac].
●, [DEA][Lac]; the symbols represent experimental values of speed
AN
of sounds.
Loss of weight percent against temperature of PILs. (a) [DEA][Lac] (b) [DEA][Tfa].
Figure 5.
Differential scanning calorimeter plot, Heat flow as function of temperature (a)
ED
M
Figure 4.
AC
CE
PT
[DEA][Tfa] (b) [DEA][Lac].
15
ACCEPTED MANUSCRIPT
Table 1 Name of the chemicals, synthesized protic ionic liquids and their purity.
N,N-diethanolammine
2
Trifluoroacetic acid
3
Lactic acid
4
Methanol
Tfa
≥ 99%
Lac
≥ 98%
-
≥ 99%
[DEA][Tfa]
≥99% a
[DEA][Lac]
≥99% a
(CAS No. 76-05-1) Alfa Aesar (CAS No. 50-21-5) Sigma-Aldrich
(CAS No. 67-56-1)
NU
Synthesized
MA
N,Ndiethanolammonium
D
Synthesized
CE
PT E
From the 1H NMR spectroscopic technique
AC
a
≥ 98%
Sigma-Aldrich
diethanolammonium
Lactate
DEA
(CAS No. 111-42-2)
trifluoroacetate
6
Purity
Sigma-Aldrich
N,N5
Abbreviation
PT
1
Source
RI
Chemicals
SC
S. No.
16
ACCEPTED MANUSCRIPT Table 2 Molecular weight (M), water content, decomposition temperature (Td) at 5% weight loss, glass transition temperature (Tg), standard entropy (S°) and lattice potential energy (Upot) of the studied protic ionic liquids. Water content ppm
M g.mol-1
Td K
Tg K
a ° S -1
J.K .mol-1
a
Upot kJ.mol-1
2920
443.6
208.0
[DEA][Lac] 413.8
218.0
355.2
354.8
CE
PT E
D
MA
NU
at 303.15 K
AC
a
2880
357.4
SC
195.21
348.6
RI
219.16
PT
[DEA][Tfa]
17
ACCEPTED MANUSCRIPT Table 3 experimental densities (ρ), speed of sound (u), thermal expansion coefficient (), isentropic compressibility (s) of the studied IL’s in the temperature range from T = (303.15 to 343.15) K. T
u
·
s·102
K
kgm-3
ms-1
Pa-1
5.88
3.37
5.86
3.43
5.88
3.51
5.89
3.57
5.87
3.64
5.86
3.71
[DEA][Tfa] 1416.5
1447.5
308.15
1412.3
1434.7
313.15
1408.1
1422.6
318.15
1404.0
1410.9
323.15
1399.9
1399.7
328.15
1395.7
1388.7
333.15
1391.7
1378.1
5.85
3.78
338.15
1387.6
1367.4
5.85
3.85
343.15
1383.5
1357.0
5.95
3.92
1880.3
5.08
2.29
1842.8
5.09
2.39
1813.9
5.10
2.47
1791.1
5.10
2.54
1771.9
5.13
2.61
1755.3
5.15
2.66
1740.1
5.38
2.72
1726.1
5.14
2.77
1712.7
5.21
2.82
MA
NU
SC
RI
PT
303.15
[DEA][Lac]
1197.8a 1228.7
PT E
308.15
1225.6
313.15
328.15 333.15 338.15 343.15
1222.5
CE
323.15
1191.9a
1219.3
AC
318.15
D
1231.8
303.15
1185.8a 1216.2 1213.1 1179.6a 1209.8 1206.7 1173.5a
Standard uncertainties are u (ρ) = 0.1 kg∙m-3, u (u) = 0.5 m∙s-1, u (T) = 0.01 K. a Ref. [27]
18
ACCEPTED MANUSCRIPT Table 4 Fitting parameters, average relative deviation (ARD) and maximum average deviation (MAD) values of densities of studied PILs. AO kg·m-3
A1 kg·m-3·K-1
A2103 kg·m-3·K-2
ARD
MAD
[DEA][Tfa]
1702.30
-1.048
0.346
0.003
0.006
[DEA][Lac]
1395.16
-0.460
0.260
0.002
0.007
AC
CE
PT E
D
MA
NU
SC
RI
PT
PILs
19
ACCEPTED MANUSCRIPT Table 5 Effective molar volume of cation (𝑉𝑐∗ ), anion (𝑉𝑎∗ ), 𝑉𝑚∗ obtained using Gardas and Coutinho model (Equation 8) and estimated molar volume (Vm) obtained from Equation 4 at
𝑉𝑐∗ Å3
𝑉𝑎∗ Å3
𝑉𝑚∗ Å3
Vm Å3
% RDa
[DEA][Tfa]
115.4
141.0
256.4
256.8
0.5
[DEA][Lac]
115.4
146.4
261.8
0.2
RI
CE
PT E
D
MA
NU
SC
Relative deviation (RD); 𝑅𝐷 = (𝑉𝑚 − 𝑉𝑚∗ ) × 100⁄𝑉𝑚
263.1
AC
a
PILs
PT
303.15 K and 0.1MPa.
20
ACCEPTED MANUSCRIPT Table 6 Experimental densities (ρ), molecular radius (r) and intermolecular length (Lf) of studied PIL’s at 303.15 K and 0.1MPa. ρ kgm-3
r Å
Lf Å
[DEA][Tfa]
1416.5
2.453
0.314
[DEA][Lac]
1231.8
2.471
0.381
AC
CE
PT E
D
MA
NU
SC
RI
PT
PILs
21
Fig 1.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
22
Fig 2.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
23
Fig 3.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
24
Fig 4.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
25
Fig 5.
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
26
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
Graphical abstract
27
ACCEPTED MANUSCRIPT
Research Highlights:
N,N-diethanolammonium based protic ionic liquids (PILs) have been synthesized. Several thermophysical properties of PILs were measured as function of temperature.
PT
Density of PILs was predicted using Gardas and Coutinho model.
RI
Experimental and derived properties used to analyse electrostatic and molecular
AC
CE
PT E
D
MA
NU
SC
interactions in PILs.
28