Density and refractive index measurement of transition-temperature

Accepted Manuscript Density and refractive index measurement of transition-temperature mixture (deep eutectic analogues) based on potassium carbonate with dual hydrogen bond donors Hosein Ghaedi, Muhammad Ayoub, Suriati Sufian, Azmi Mohd Shariff, Bhajan Lal, Cecilia Devi Wilfred PII: DOI: Reference:

S0021-9614(17)30403-2 https://doi.org/10.1016/j.jct.2017.11.008 YJCHT 5259

To appear in:

J. Chem. Thermodynamics

Received Date: Revised Date: Accepted Date:

9 June 2017 11 November 2017 14 November 2017

Please cite this article as: H. Ghaedi, M. Ayoub, S. Sufian, A.M. Shariff, B. Lal, C.D. Wilfred, Density and refractive index measurement of transition-temperature mixture (deep eutectic analogues) based on potassium carbonate with dual hydrogen bond donors, J. Chem. Thermodynamics (2017), doi: https://doi.org/10.1016/j.jct.2017.11.008

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.

Density and refractive index measurement of transition-temperature mixture (deep eutectic analogues) based on potassium carbonate with dual hydrogen bond donors

Hosein Ghaedi a, Muhammad Ayoub a,*, Suriati Sufian a, Azmi Mohd Shariff a, Bhajan Lal a, Cecilia Devi Wilfred b a

b

Department of Chemical Engineering, Universiti Teknologi PETRONAS,

Department of Fundamental and Applied Sciences, Universiti Teknologi PETRONAS, 32610 - Bandar Seri Iskandar, Perak, MALAYSIA *

Corresponding author: Email: [email protected]; Telephone/fax: +6053687623.

ABSTRACT The transition-temperature mixture (TTM) is a new type of solvent and is known as a tuneable solvent similar to deep eutectic solvent (DES). The new type of solvent called ternary transition-temperature mixture (TTTM) was prepared to use for CO2 capture purposes. In this work, TTMs and TTTMs were prepared with potassium carbonate (PC) as a hydrogen bond acceptor (HBA) and three hydrogen bond donors (HBDs) such as glycerol (GL), ethylene glycol (EG) and 2-amino-2-methyl-1-3-propanediol (AMPD) known as a hindered amine (HA). Binary TTMs were PC-GL with mole ratios 1:10 and 1:16 and PC-EG with the same mole ratios. TTTMs were prepared by adding AMPD in binary TTMs such as PC-GL-AMPD 1:16:1 and PC-EG-AMPD 1:10:1. The experimental density and refractive index of all mixtures were measured at temperatures from 293.15 K to 343.15 K with an interval of 5 K. The effect of temperature, mole ratio, molar massand alkyl chain length on the properties was investigated. The molar volumes and isobaric thermal expansion were calculated using experimental density data. The experimental refractive index values were used to derive the specific refraction, molar refraction, free molar volume, electronic polarization and polarizability at several temperatures. Keywords: TTM; TTTM; salt; HBD; density; refractive index. 1. Introduction

Deep eutectic solvents (DESs) are derived from two or more salts with hydrogen bond acceptors

(HBAs)

such

allyltriphenylphosphonium

as

chloride

allyltriphenylphosphonium (ATPPC),

bromide

methyltriphenylphosphonium

(ATPPB), bromide

(MTPPB), methyltriphenylphosphonium chloride (MTPPC), benzyltriphenylphosphonium bromide (BTPPB), benzyltriphenylphosphonium chloride (BMTPPC), choline chloride (ChCl) and potassium carbonate (PC) and hydrogen bond donors (HBDs); e.g. amides, amines, alcohols, and carboxylic acids. Similar to DESs, transition-temperature mixtures

(TTMs) may be formed by mixing HBA(s) and HBD(s), therefore TTMs can be regarded as deep eutectic analogues. One may question the difference between these types of solvents. The reason they have been termed DES’s arises because when the two constituents are mixed together in the correct ratio a eutectic point can be seen by measuring their freezing points. However, this eutectic point cannot be evident in TTM. Indeed, this type of solvent only exhibits the glass transition temperature at the freezing point. This is the main difference between TTMs and DESs. It should be mentioned that several previous research groups reported some solvents as DESs without proving a eutectic point. In the following, some of these studies will be highlighted. Morrison et al. [1] could not find the eutectic point of malonic acid–choline chloride system using differential scanning calorimetry (DSC) from −80 ◦C to 100 oC. They mentioned that no thermal events were observed except for the shift in baseline at above −50 o

C. Gutierrez et al. [2] prepared resorcinol (R), 3-hydroxypyridine (H) and choline chloride

(ChCl)-(RHChCl) at mole ratio of (2:2:1). Their results show that only the glass transition temperature about -61 oC was displayed in the DSC trace of this solvent. Guo et al. [3] prepared the phenol- choline chloride mixtures. They mentioned that it is difficult to determine freezing temperatures from DSC traces. However, according to DSC curves, there were no freezing points for this mixtures. Chemat et al. [4] prepared four ternary mixtures of choline chloride, urea and L- arginine. Their results revealed that mixtures exhibited different behaviour only for glass transition but not for melting point. Mjalli et al. [5] prepared potassium carbonate-glycerol mixtures and qualified that there is an absence clear of freezing points in DSC curves. Therefore, all solvents reported by researches noted above cannot be considered as DESs and they should be categorized in TTMs.

One of the interesting studies by Francisco et al. [6] who prepared more than 20 solvents. Their results disclosed that no melting points were found for the liquids formed, which is the reason why these liquids are named low-transition-temperature mixtures (LTTMs). Therefore, it is important to distinguish between these two different types of solvents. DES shows a freezing point while TTM displays only a glass transition temperature in DSC curve. In order to improve the features of DESs for desired applications, researchers are trying to prepare the ternary deep eutectic solvents (TDES). The DESs are known as tuneable solvents [7]. Maugeri and de Marı´a [8] reported the synthesis of TDES including ChCl, bio-based HBDs and glycerol (GL) as second HBD and measured their viscosity. Dai et al. [9] prepared TDES with the name of the natural deep eutectic solvents (NADES) by mixing the several abundant constituents (primary metabolites) for example sugars, sugar alcohols, amino acids, organic acids, and choline derivatives. Wang et al. [10] formed TDES with the components of ChCl, urea and nickel chloride hexahydrate (NiCl2.6H2O) and measured the physicochemical properties. Liu et al. [11] synthesized several TDESs by mixing imidazolium halides, zinc halides and amides and reported the physicochemical properties. Sze et al. [12] synthesized seven TDES for CO2 capture by mixing the ChCl, GL and different super bases. Khezeli and Daneshfar [13] synthesized two types of ChCl- based TDEs having ChCl, phenol, ethylene glycol and anhydrous iron (III) chloride (FeCl3) and named them magnetic deep eutectic solvents (MDESs). Sarkar and Sampath [14] reported a TDES based on acetamide, urea and gallium nitrate for the electrode position of gallium nitride/gallium indium nitride. A family of TDES formed by natural sugars, urea and different salts (NH4Cl or NaCl) has been reported by Imperato [15] and Ilgen and Konig [16]. On the other hands, it is possible and interesting to prepare ternary transition-temperature solvents (TTTMs) as prepared by Chemat et al. [4] because similar to DES, the TTM can be

considered as a tuneable solvent. Therefore, in the present work, four binary TTMs and two new TTTMs were prepared by mixing salt such as potassium carbonate (PC) with some HBDs for instance glycerol (GL), ethylene glycol (EG), and 2-amino-2methyl-1-3propanediol (AMPD) as a hindered amine. The binary TTMs were PC- GL and PC- EG with mole ratios of 1:10 and 1:16 salt to HBDs. The TTTMs were PC-GL-AMPD with mole ratio of 1:16:1 and PC-EG-AMPD with mole ratio 1:10:1. The experimental density and refractive index of HBDs, TTMs and TTTMs were measured at several temperatures from 293.15 K to 343.15 K with an interval of 5 K. These experimental values were used to calculate the molar volume, isobaric thermal expansion, specific refraction, molar refraction and electronic polarization of TTMs and TTTMs.

2. Material and methods 2.1 Chemicals Potassium carbonate (PC) with >0.999 mass fraction purity and 2-amino-2methyl-1-3propanediol (AMPD) with >0.99 mass fraction purity were supplied by Sigma-Aldrich; glycerol (GL) with >0.998 mass fraction purity and ethylene glycol (EG) >0.995 mass fraction purity were purchased from R&M Chemicals supplier. Table 1 contains the specification of these chemicals and Table 2 represents the abbreviation of these chemicals, TTMs and TTTMs along with mole ratio, symbol and molar mass. In order to prevent moisture and any contamination, all materials were kept in a controlled environment. Figure 1 shows the molecular structure of the PC and the HBDs. In this study, TTM1 and TTM2 represent PC-EG 1:10 and PC-EG 1:16, respectively; TTM3, TTM4 show PC-GL 1:10 and PC-GL 1:16, respectively; TTTM1 and TTTM2 represent PC-EG-AMPD 1:10:1 and PC-GL-AMPD 1:16:1, respectively. [Insert Table 1]

Table 1 Specification of chemical samples used in this study. [Insert Table 2] Table 2 The abbreviations, molar mass and mole ratios for TTMs, TTTMs and their single components in this work.a [Insert Figure 1] Figure 1. A schematic of chemical structure of the salt and HBDs used in this work.

2.2 Preparation of solvents The procedure for preparation of TTMs is the same as DESs as mentioned in our previous works [17-25]. Binary TTMs were prepared with the ratios of 1:10 for PC to EG and 1:16 for PC to GL. In order to form ternary TTTMs, a type of hindered amine (HA) called 2-amino-2methyl-1-3-propanediol (AMPD) was mixed with the binary mixtures with the mole ratios of 1:10:1 for PC-EG-AMPD and 1:16:1 for PC-GL-AMPD. The weight measurements of pure components of all TTMs and TTTMs were performed in a digital balance (Sartorius, model BSA 224S-CW) with an accuracy of ±0.1 mg. The salts were mixed with related HBDs using magnetic stirrer at 350 rpm, atmospheric pressure and the temperature from 343.15 K to 373.15 K for (1-2) h in a fume hood until a homogeneous and uniform liquid without any precipitate was formed. For preparation of TTTMs, after 15-30 minutes mixing the binary mixtures, AMPD as a ternary part was immediately added to binary mixture while stirring. The heating of ternary mixture continued more than 1h. The formed TTMs and TTTMs were kept in the tight bottles to prevent any contamination with atmospheric water vapour. The TTMs and TTTMs were used without any further purification. Since TTMs and TTTMs are known as the hygroscopic solvents, their water contents were measured using a Mettler Toledo Karl-Fischer (C30) after preparation. It was

found that the water content of TTMs and TTTMs is less than 0.0071 mass fraction as listed in Table 2. The TTMs and TTTMs were used without any further purification.

2.3 Differential scanning calorimetry (DSC) measurement The DSC was carried out on a Mettler-Toledo DSC 1. Data were evaluated using the Mettler-Toledo STARe Software version 9.30. To ensure the integrity of data gathered on the instrument, periodic calibrations should be performed. Common calibration substances for temperature DSC are high purity metals due to their well-defined melting point and Indium (having a melting temperature of 156.6 oC) is often used as a reference. Therefore, The DSC calibration was performed using Indium through its melting point. The extrapolated onset temperatures at different heating rate are determined and programmed into the instrument. At higher temperatures, materials such as aluminium (used in this study), nickel and gold, and can be applied, as well as eutectic systems including Fe-C and Ni-C. In each of the experiment, the aluminium 40 µL crucible pan and a purge flow of 50 mL·min-1 of dry nitrogen were used. The weight measurement of samples (9 to 10) mg was determined on a digital balance (Sartorius, model BSA 224S-CW) with an accuracy of ±0.1 mg. The sample was first cooled from 25 °C to -150 °C and heated from -150 °C to 25 °C at a rate 5 °C·min-1 followed by cooling to -150 °C, and finally heated back to 25 °C with the same heating rate, similar to the established procedure. The uncertainty in temperature was found to be ±0.2 °C based on the materials used for calibration, the specimens, the type of crucible, heating rate, extrapolated onset of the DSC temperature curve. 2.4 Density measurement A digital density meter (Anton Par, model, DMA-4500M) was used to measure the density of HBDs, TTMs and TTTMs with an accuracy of ±5×10 -5 g·cm-3, and a temperature

control accuracy of ±0.01 K. The density meter works on the principle of oscillating U-tube method. In order to gain accurate results and remove any unwanted particles, the U-shaped tube of density meter was cleaned by acetone and standard water of Millipore quality. An air blower was used to dry the moisture content inside the tube and avoid any influence on the results. Density of all TTMs and TTTMs was measured in three runs at atmospheric pressure and temperatures rising from 293.15 K to 343.15 K and the average value was reported for the further study. 2.5 Refractive index measurement Refractive index of HBDs, TTMs and TTTMs was measured by means of a digital Abbemat automatic refractometer (Anton Par, model WR), with an accuracy of ±4×10-5 nD, and a temperature control accuracy of ±0.03 K. In order to obtain accurate and reliable data, standard water of Millipore quality was utilized to calibrate the refractometer. Before pouring the sample into the sample mould, the prism face was carefully cleaned with ethanol and dried to prevent any disturbance in the results because of possible minute sediments on the prism face. Experimental refractive index of TTMs and TTTMs was measured three times at atmospheric pressure and temperature ranging from 293.15 K to 343.15 K with a regular interval of 5 K at wavelength of 589.3 nm. All refractive index values are reported as average experimental values. 3. Results and discussion 3.1 Glass transition temperature Displayed in Figure 2 are the DSC curves for solvents at the heating rate 5 oC·min-1. The glass transition temperature (Tg) of samples is listed in Table 2. The results reveal that thermal transition behaviour is identified only by the formation of an amorphous glass on cooling and reformation of the liquid on heating. These solvents have no melting or freezing points, but only glass transition temperatures. Mostly the melting temperature occurs at only

crystalline region while the glass transition temperature occurs at only amorphous region in the sample [4]. From Figure 2 and Table 2, it is clear that by increasing the amount of GL in mixtures, the onset temperatures of TTM3 and TTM4 are about −83.3 oC to −90.5 oC, respectively. This result is in good agreement with value obtained by Mjalli et al. [5] who reported the transition temperature of PC-GL solvents. Their results showed that PC-GL solvents at mole ratios of 1:4, 1:5 and 1:6 have transition temperature of (−43.7, −53.41 and −57.76) oC, respectively. Therefore, it is clear that as the mole ratio increases from 1:4 to 1:16, the glass transition temperature decreases. As seen from Figure 2, by adding AMPD into binary mixture, the glass transition temperature is about −82.00 oC for TTTM2. As mention earlier, Mjalli et al. [5] reported the PC-EG solvents as DESs. They reported the freezing points of PC-EG at mole ratios of 1:6, 1:7 and 1:8 about (−73.3, −86.56 and −96.57) oC. In all probability, these points are not freezing points and indeed are glass transition temperatures. Moreover, they did not mention the exact eutectic point for this mixture. In the case of PC-EG solvents in this work, DSC curves display only glass transition temperatures in cooling-heating loop. From Figure 2, the glass transition temperatures of PCEG 1:10 (TTM1) and 1:16 (TTM2) are (−116.47 and −121.50) oC, respectively. The glass transition temperature for TTTM1 is about −112.48 oC. Therefore, these kinds of solvents are TTMs and cannot be considered as DESs. [Insert Figure 2] Figure 2. DSC curves of TTMs and TTTMs in this work.

3.2 Density property Density is an important thermos-physical property owing to its effect on the design and operation of processes [5, 26, 27]. Hence, it is essential to identify the behaviour of density with respect to temperature. In this work, the experimental density of the TTMs, TTTMs and

their HBDs were measured over the temperature range of 293.15 K to 343.15 K at ambient pressure. Table 3 represents a comparison between the density of HBDs such as GL and EG in this work and the literature data [28-31]. Low values of average absolute deviation (%AAD) values indicate that there is consistency between experimental and literature data, and the values are reproducible. The experimental density values of pure GL, pure EG, TTMs and TTTMs are tabulated in Table 4. Illustrated in Figure 3 are these values as a function of temperature. [Insert Table 3] Table 3 Comparison of the experimental and the literature data of the density and refractive index of EG and GL.a [Insert Figure 3] Figure 3. Experimental density, ρ, of the pure GL, pure EG, TTMs and TTTMs in this work, as a function of temperature: (■), pure GL; (□), pure EG; (○), TTM1; (∆), TTM2; (◊), TTTM1; (●), TTM3; (▲), TTM4; (♦), TTTM2 in this work ; and (*), PC-GL 1:6; (), PC-EG 1:6 reported by Mjalli et al. [5]. [Insert Table 4] Table 4 Experimental density, ρ, of pure GL, pure EG, TTMs and TTTMs. Measurements were conducted within the temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a

According to Table 4 and Figure 3, density values of HBDs (GL and EG) are less than those of corresponding TTMs at all temperatures. As can be observed from Figure 3 as the temperature increases the density of HBDs, TTMs and TTTMs decreases linearly. This could be attributed to the wider spaces between the mixture molecules and the increased molecular

mobility at higher temperatures which increase the molar volume expansion and decrease the molecular interactions [4, 23, 27, 32]. As shown in Figure 3, an increase in the mole fraction of GL and EG in the mixture, results in a decrease in density of TTMs. For example, the density of PC-GL 1:10 (TTM3) is 1.36728 g·cm-3 while that of PC-GL 1:16 (TTM4) is 1.30890 g·cm-3 at 298.15 K. Similarly, the density values of PC-EG 1:10 (TTM1) and of PC-EG 1:16 (TTM2) are (1.25606 and 1.20957) g·cm-3, respectively. As mentioned earlier, Mjalli et al. [5] prepared potassium carbonate plus GL and EG with the lower mole ratios than ours. These observations were in a well agreement with their results provided that PC-EG solvent is considered as TTM. There are two possible reasons for this behaviour. Firstly, density of PC is about 2.285 g·cm-3 and is greater than that of GL (1.2590 g·cm-3) and EG (1.1098 g·cm-3) at 298.15 K. Therefore, by increasing the amount of GL and EG in the TTMs, density of TTMs decreases towards density of HBDs because of the lower density of HBDs. Figure 4 shows the composition behaviours of density for PC-EG and PC-GL. To show a clear comparison, the density values of PC-EG at mole ratios of 1:6, 1:7, 1:8 and PC-GL at mole ratios of 1:4, 1:5, 1:6 were collected from Ref. [5]. It can be seen from Figure 4 that the composition behaviour of density for PC-EG and PC-GL is different. As the amount of PC in PC-EG mixture increases, the density of this mixture increases linearly, as illustrated in Figure 4 (a), while this trend is not linear in the case of the PC-GL mixture especially at lower amount of PC in this mixture (see Figure 4 (b)). It is may be reasoned that the hydrogen bond interactions in PC-GL are stronger than those of PC-EG and have a greater effect on the density. The strengths are weaker in PC-GL 1:16 (with lower amount of PC component in mixture). [Insert Figure 4]

Figure 4. Composition behaviour of density for (a), PC-EG mixture and (b), PC-GL mixture. The red lines represent the data taken from Ref. [5].

In order to show the effect of mole ratio on the density, density values of PC-GL 1:6 and PC-EG 1:6 in Ref. [5] were collected, as shown in Figure 3. It is clear from Figure 3 that by increasing the amount of HBDs in mixtures, the density of TTMs decreases towards the density of HBDs. By adding AMPD to the binary mixture of PC-EG 1:10, the density of TTTM1 shows a decreasing trend in the value. For example, density of PC-EG 1:10 is 1.22610 g·cm-3 at 343.15 K, while that of PC-EG-AMPD 1:10:1 (TTTM1) is 1.20744 g·cm-3 at the same temperature. This trend can be seen in the case of PC-GL-AMPD 1:16:1 (TTTM2). The density of TTTM2 is greater than that of PC-GL 1:16 (TTM4) at all temperatures except for 293.15 K and 298.15 K.

In our previous work, the effect of molar mass was studied on density of DESs [23]. It is interesting to mention that as the molar mass of TTMs decreases the density decreases provided that TTMs have the same components but different mole ratio (see Table 2 for molar mass and Table 4 and Figure 3 for density of TTMs). It should be noted that the prepared samples have water content less than 0.0071 mass fraction as presented in Table 2. In our previous works [17, 18], it was reported that the water has an effect on the density of DESs. Therefore, this amount of water content can affect the density of mixtures causing a slight reduction in the density values. A linear relationship was used to fit the experimental density values, as the following:

ρ = a + b.(T )

(1)

where ρ is the density in g·cm-3, T is the temperature in Kelvin, a and b are constants that depend on the type of TTMs or TTTMs. A method of least-squares was applied using the

Levenberg-Marquardt algorithm to derive constant parameters (a and b) of Eq. (1). The values of a and b for all TTMs and TTTMs together with %AAD, RMSE and R2 values are shown in Table 5. We compared graphically the predicted density by Eq. (1) with the experimental density for all TTMs and TTTMs, as portrayed in Figure S.1 in supporting information. From Figure S.1, it is clear that the predicted density values for all TTMs and TTTMs are in good agreement with experimental density with R2 values of >0.999, which indicate that the density–temperature relationships are linear. [Insert Table 5] Table 5 The constant parameters of a, b used in Eq. (1), %AAD, RMSE and R2 values.

The molar volume, V, of TTMs and TTTMs was calculated using the experimental density values at the temperatures studied here, according to the following equation.

V=M ρ

(2)

In the Eq. (2), ρ and M are the density and molar mass of TTMs and TTTMs. The formula for calculating the molar mass of TTMs is available in our previous work [22, 23] and the molar mass of TTTMs is calculated according to Eq. (3):

M=

X HBA M HBA + X HBD,1M HBD,1 + X HBD,2 M HBD,2 X HBA + X HBD,1 + X HBD,2

(3)

where M is the molar mass of TTMs and TTTMs in g·mol−1, XHBA and MHBA are the mole ratio and molar mass of the salt as HBA in g·mol−1 , respectively; XHBD and MHBD are the mole ratio and molar mass of the HBD in g·mol−1, respectively. The numerical values of molar volume are reported in the supporting information contained in Table S.1. The temperature dependence of molar volume of TTMs and TTTMs is shown in Figure 5. Different from the

behaviour of the density, the molar volume of TTMs and TTTMs increases with rising temperature. The effect of mole ratio on the molar volume was investigated and it was found that by increasing the amount of HBDs in TTM, the molar volume of TTMs increases. However, this change was minor in the case of PC-EG TTMs, as shown in Figure 5 and Table S.1. [Insert Figure 5] Figure 5. Molar volume, V, of TTMs and TTTMs as a function of temperature. The symbols

are the same as symbols depicted in Figure 3.

3.3 Isobaric thermal expansion The liquid contained in the equipment will expand if there is an increase in temperature. Therefore it is essential to design a safety system that will relieve (or emit) the thermally expanding liquid and hinder the pressure arise from this expansion. If the pressure increase is too great, the process equipment or apparatus would be damaged. Thus, knowledge of the thermal expansion coefficient, αp / K-1, is important in process engineering [4, 33]. This property which provides the information regarding expansion with temperature shows the fractional change in density when temperature increases at constant pressure. The temperature dependence of the isobaric thermal expansion coefficients, αp / K-1, were derived at different temperatures from the parameters obtained by fitting of the experimental density values using Equation (3) and is calculated according to the following equation:

1  ∂V  1  ∂ρ  b  =−   =− V  ∂T P a + b.T ρ  ∂T P

αp = 

(4)

where ρ is the density of TTMs and TTTMs in g·cm-3, T is the temperature in Kelvin, P is the ambient pressure, and a and b represent the parameters gained from the fitting of density data

using Eq. (1). Figure 6 depicts the temperature dependence of αp for all TTMs and TTTMs. It can be clearly seen that temperature does not exert a significant effect on αp of TTMs and TTTMs. Moreover, there is a great effect on αp of TTMs due to changing the mole ratio in TTMs. As can be seen from Figure 6, as the mole ratio of PC-EG and PC-GL TTMs increased from 1:10 to 1:16, the amount of αp of TTMs increased especially in the case of PC-GL. It might be reasoned that a decrease of hydrogen bond interactions in TTMs leads to a decrease in density, greater free volume of solvent (see Table 8 for free volume of TTMs) and consequently raising the isobaric thermal expansion coefficients [4]. On the other hand, by adding AMPD in PC-GL 1:16 (TTM4), the αp values of PC-GL-AMPD (TTTM2) 1:16:1 decreased at all temperatures, while the αp values of PC-EG-AMPD 1:10:1 (TTTM1) increased when the AMPD was added in PC-EG 1:10 (TTM1). [Insert Figure 6] Figure 6. Isobaric thermal expansion, αp, of TTMs and TTTMs at several temperatures of.

The symbols are the as symbols depicted in Figure 3.

3.4 Referactive index Refractive index is a fundamental physical property of TTMs and TTTMs. The experimental refractive index of TTMs, TTTMs and their HBDs were measured as a function of temperature and repeated three times. The refractive index of pure GL and pure EG are compared with the literature data [34-38], as presented in Table 3. Low values of %AAD indicate the confirmation of the reproducibility of the experiments. The experimental refractive index values of pure GL, pure EG, TTMs and TTTMs are listed in Table 6 and graphically shown in Figure 7. From the data represented in Table 6 and displayed in Figure 7, it is evident that pure GL has the higher refractive index than pure EG at all temperature studied here.

As mentioned before, the prepared samples have water content less than 0.0071 mass fraction as shown in Table 2. It has been well established that water has an effect on the refractive index of DESs [32, 39]. Therefore, this amount of water in mixtures can decrease the refractive index values. [Insert Table 6] Table 6

Experimental refractive index, nD, of pure EG, pure GL, TTMs and TTTMs at several temperatures. Measurements were conducted within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a [Insert Figure 7] Figure 7. Refractive index, nD, as a function of temperature for pure GL, pure EG, TTMs and

TTTMs in this work and PC-GL 1:6 and PC-EG 1:6 reported by Mjalli et al. [5]. The symbols are the same as symbols depicted in Figure 3.

There are several factors which have an effect on the refractive index of TTMs and TTTMs such as type of salt to HBD, mole ratio, temperature and molar mass. The results reveal that the values of nD decrease with an increase in temperature, since temperature has an effect on the density of the TTMs and TTTMs, consequently affecting the refractive indices of TTMs and TTTMs. In the case of TTMs, the refractive index values of PC-GL TTMs were higher than those of PC-EG TTMs at all temperatures. Indeed, components with a higher density value had the higher refractive index value whether single components or TTMs. Figure 8 depicts the composition behaviour of the refractive index for PC-EG and PCGL. To compare the behaviour, the refractive values of PC-EG at mole ratios of 1:6, 1:7, 1:8 and PC-GL at mole ratios of 1:4, 1:5, 1:6 were collected from Ref. [5]. As can be seen, the composition behaviour of refractive index for both mixture is similar to their composition

behaviour in the case of density. From Figures 4 (a) and 8 (a), the composition behaviour of density and refractive index for PC-EG mixture in this work is linear. Moreover, in the case of the PC-GL mixture, the composition behaviour of density and refractive index is similar, as shown in Figures 4 (b) and 8 (b). It should be noted that as the amount of PC in mixtures increases the refractive index values of mixtures also increases (in this work); however this was not observed in the case of PC-EG mixture as reported in Ref. [5] (see Figure 8 (a), red lines). Although Mjalli et al. [5] explicitly confirmed this behaviour, the data reported in figure are different in their study. [Insert Figure 8] Figure 8. Composition behaviour of refractive index for (a), PC-EG mixture and (b), PC-GL

mixture. The red lines represent the data taken from Ref. [5]. In order to investigate the effect of mole ratio on refractive index, the refractive index values of PC-GL 1:6 and PC-EG 1:6 were taken from Ref. [5], as illustrated in Figure 7. From Figure 7, it is evident that with an increase in mole fraction of HDSs in the TTMs, the refractive indices decrease. Besides, as the molar mass of TTMs decreases, the refractive index of TTMs decreases at all temperatures. This behaviour is similar to that of density and molar volume. Moreover, it was observed that the composition behaviour of refractive index for PC-EG and PC-GL is different. This may be due to hydrogen bond interactions in PC-GL that are stronger than those of PC-EG. Thus, PC-GL is denser and more viscous than PC-EG. It is more likely that the hydrogen bonding interactions are stronger in PC-GL 1:10 rather than in PC-EG 1:10 causing different composition behaviour in the refractive index in these mixtures. Similar to density, by adding AMPD to the binary mixture of PC-EG 1:10 (TTM1), the refractive index of PC-EG-AMPD 1:10:1 (TTTM1) shows a decreasing trend in its value. For example, the refractive index of PC-EG 1:10 (TTM1) is 1.43682 at 338.15 K, while that of

PC-EG-AMPD 1:10:1 (TTTM1) was 1.44131 at the same temperature. This trend can be seen in the case of PC-GL-AMPD 1:16:1 (TTTM2) which shows the higher values of refractive index than PC-GL 1:16 (TTM4) at all temperatures. In our previous work [22], a new empirical equation was proposed to correlate physicochemical properties of DESs. This equation was applied to correlate refractive index of TTMs and TTTMs, as in the following:

ln(

nD A A A ) = A0 + 1 + 22 + ... nn = ρM T T T

n

Ai

∑T

i

(5)

i =0

Also, the experimental refractive index values were fitted linearly to the following equation:

nD = a + b.(T )

(6)

where nD is the refractive index; T is the temperature in Kelvin; Ai, a and b are the fitting parameters; ρ is the density in g·cm-1; M is molar mass in g·mol-1. In order to derive constant parameters of Eqs. (5) and (6), the method of least-squares was applied using the LevenbergMarquardt algorithm. Shown in Table 8 are the values of the constant parameters along with %AAD, RMSE and R2 values. It is worth mentioning that Eq. (5) is a power infinite series and one may be concerned about the number of terms should be considered to provide satisfactory accuracy. Here, it was found that Eq. (5) is more accurate until three constant parameters, as shown in Table 8. As can be seen from Table 8, both equations correlated the experimental refractive indices with R2 value of >0.999. However, %AAD values of the new empirical equation are lower than those yielded by Eq. (6). Therefore, Eq. (5) is powerful and reliable to correlate the experimental refractive index values.

The predicted values of refractive index by Eq. (5) were compared with experimental results. The results of this comparison are shown in Figure S.2 in the supporting information for all TTMs and TTTMs studied in this work. From Figure S.2, it is obvious that the predicted refractive indices by Eq. (5) are in good agreement with the experimental refractive index values with R2 values of >0.999. [Insert Table 8] Table 8

Estimated constant parameters of Eqs. (5) and (6), %AAD, RMSE and R2.

The molar refraction, Rm /(cm3.mol-1), is a measure of the polarizability of a substance. The knowledge of specific refraction, RD /(g-1·cm3), molar refraction and electronic polarization, E, and polarizability constant, δ, of solvents play a vital role in many fields such as chemical, medical, biotechnical and engineering. Indeed, all of these properties are associated with refractive index. In the present work, all of the aforementioned properties of TTMs and TTTMs at several temperatures were calculated as following equations [40]:

E = nD2

(7)

nD2 − 1 1 . nD2 + 2 ρ

(8)

nD2 − 1 Rm = 2 .V nD + 2

(9)

4 π . N A .δ 3

(10)

RD =

Rm =

where ρ, V and nD are density, molar volume and refractive index of TTMs and TTTMs, respectively. δ is polarizability constant of liquid TTMs and TTTMs; NA is the Avogadro constant.

In order to study the forces between molecules and their behaviour in solutions, extraction of E data is important [41]. Since E is power state of refractive index, it is expected that with an increase in temperature, the electronic polarization of TTMs and TTTMs would decrease. Similarly, this phenomenon is observed when the mole ratio of TTMs increases, as shown in Figure 9. The E values for PC-GL 1:10 (TTM3) and PC-EG 1:10 (TTM1) are 2.2023 and 2.0931 at 298.15 K, respectively. After the increasing mole fraction HBDs in TTMs, these values for PC-GL 1:16 (TTM4) and PC-EG 1:16 (TTM2) decreased to 2.1813 and 2.0784 at the same temperature, respectively. On the other hand, by adding AMPD to binary TTMs, the values of E for both TTTMs almost increased compared with those of binary TTMs. For example, the E value for PC-GL 1:10 (TTM1) is 2.0931while that of PC-EG-AMPD 1:10:1 (TTTM1) is 2.1074 at 298.15 K. [Insert Figure 9] Figure 9. Electronic polarization, E, of TTMs and TTTMs with various mole ratios at several

temperatures. The symbols are the same as symbols depicted in Figure 3. Figure 10 displays the specific refraction, RD, of TTMs and TTTMs. As can be seen, the mole ratio salt to HBDs shows a significant effect on the specific refraction of TTMs. As the mole ratio increases this property decreases at all temperatures. However, temperature exerted a minor effect on RD and an increase in temperature from 293.15 K to 343.15 K, brought about a minimal increase in the amount of RD for all TTMs and TTTMs. As can be seen from Figure 10, there is a slight upward trend on the specific refraction of all TTMs and TTTMs by increasing temperature. In the case of TTTMs, their RD values decrease in comparison to their corresponding binary TTMs, though this decrease in RD value for PCGL-AMPD 1:16:1(TTTM2) occurred after 313.15 K. [Insert Figure 10]

Figure 10. Specific refraction, RD, of TTMs and TTTMs with various mole ratio at several

temperatures. The symbols are the same as symbols depicted in Figure 3.

Table S.2 in the supporting information represents the numerical values of molar refraction, Rm, of TTMs and TTTMs and Figure 11 depicts these values against various temperatures. Although, temperature shows a weak effect on Rm of TTMs and TTTMs, the effect of mole ratio is clear on Rm of TTMs. In the case of PC-EG TTMs, as the mole ratio increases from 1:10 to 1:16 salt to HBD, the Rm values decrease marginally. However, there is an increase in the Rm values of PC-GL TTMs when the mole ratio increases. From the data listed in Table S.2 and shown in Figure 11, it is obvious that by adding AMPD as a third component to binary TTMs, the Rm values of both TTTMs increases compared to their corresponding binary TTM. For example, PC-GL 1:16 (TTM4) has the Rm value of 20.4623 cm3∙mol-1 at 298.15 K, in contrast the Rm value for PC-GL-AMPD 1:16:1 (TTTM2) was 20.6104 cm3·mol-1 at the same temperature.

[Insert Figure 11] Figure 11. Molar refraction, Rm, of TTMs and TTTMs with various values of mole ratio at

several temperatures. The symbols are the same as symbols depicted in Figure 3.

In our previous work, Eq. (11) was used for calculating the free volume, fm, phosphoniumbased DESs. Here, Eq. (11) was used to calculate fm values of TTMs and TTTMs:

f m = V − Rm

(11)

where V is molar volume in cm3·mol-1; Rm is molar refraction in cm3·mol-1. Table 9 represents free volume of all TTMs and TTTMs at several temperatures. From Table 9, it is clear that with an increase in temperature all TTMs and TTTMs witness a gradual rise in the free volume. Further, by increasing mole ratio (from 1:10 to 1:16), the free volume all TTMs increases. Amongst TTMs with the same mole ratio, TTMs with GL in their structure have the higher free volume. This may be explained by the presence of the higher alkyl chain on HBDs. From Figure 1, it can be seen that GL has an additional CH in its structure and consequently leads to the higher free volume. Moreover, by adding AMPD in binary TTMs to form TTTMs, the free volume of both TTTMs increases compared to their corresponding binary TTMs. [Insert Table 9] Table 9

Calculated free volume, fm, of TTMs and TTTMs within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa. a

Extraction of the values of refraction index and polarizability, δ, for the TTMs and TTTMs as a solvent media is important because having access to these data can give vital information about the behaviour of TTMs and TTTMs such as dispersion forces. Indeed, a solvent with a large δ value has strong dispersion forces [41]. The δ values of TTMs and TTTMs are shown in Table 10. From Table 10, it is clear that there is a decrease in the δ value of TTMs as the mole ratio of TTMs increases. Also, the δ values for both TTTMs are lower than those of their corresponding binary TTMs. For example, the δ values for PC-GL 1:16 (TTM4) and PCGL-AMPD 1:16:1 (TTTM2) are (1.7078 10-23 and 1.611310-23) cm3·mol-1 at 298.15 K. Furthermore, there is a marginal increase in the δ values of TTMs and TTTMs, as the temperature increases.

[Insert Table 10] Table 10

Calculated polarizability constant, δ, of TTMs and TTTMs within the temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a

4.

Conclusions

In this work, the binary TTMs were PC-GL with mole ratios 1:10 and 1:16 and PC-EG with the same mole ratios. TTTMs were prepared by adding AMPD in binary TTMs such as PC-EG-AMPD 1:10:1 (TTTM1) and PC-GL-AMPD 1:16:1 (TTTM2). The experimental density and refractive index of all TTMs, TTTMs and their HBDs were measured at temperatures from 293.15 Kto 343.15 K with an interval of 5 K. The effect of temperature, mole ratio, type of HBD and molar mass was investigated on the properties. The results revealed that as the temperature increases, the density and refractive index values of TTMs and TTTMs decrease. Moreover, the mole ratio had an effect on these properties TTMs. By increasing the amount of HBDs in TTMs the density and refractive index values of TTMs decrease toward the properties of their HBD. By adding AMPD to the binary TTM of PC-GL 1:16 (TTM4) the density and refractive index of PC-GL-AMPD 1:16:1 (TTTM2) increase at all temperatures except for density at 293.15 K and 298.15 K. The density of PC-EG-AMPD 1:10:1 (TTTM1) decreased at all temperatures by adding AMPD to the binary TTM of PC-EG 1:10 (TTM1), while the refractive index of this TTTM1 increased at the same temperatures. It was found that TTMs with the GL in their structures had the greater density and refractive index values in comparison to TTMs having EG in their structures with same mole ratios. Finally, there was a decrease in the density and refractive index of TTMs (with the same components but different mole ratios) when the molar mass decreased.

References [1]

[2]

[3]

[4]

[5] [6]

H.G. Morrison, C.C. Sun, S. Neervannan, Characterization of thermal behaviour of deep eutectic solvents and their potential as drug solubilization vehicles, International Journal of Pharmaceutics, 378 (2009) 136-139. M.C. Gutierrez, D. Carriazo, C.O. Ania, J.B. Parra, M.L. Ferrer, F. del Monte, Deep eutectic solvents as both precursors and structure directing agents in the synthesis of nitrogen doped hierarchical carbons highly suitable for CO2 capture, Energy & Environmental Science, 4 (2011) 3535-3544. W. Guo, Y. Hou, S. Ren, S. Tian, W. Wu, Formation of Deep Eutectic Solvents by Phenols and Choline Chloride and Their Physical Properties, Journal of Chemical & Engineering Data, 58 (2013) 866-872. F. Chemat, H. Anjum, A.M. Shariff, P. Kumar, T. Murugesan, Thermal and physical properties of (Choline chloride + urea + l-arginine) deep eutectic solvents, Journal of Molecular Liquids, 218 (2016) 301-308. F.S. Mjalli, J. Naser, B. Jibril, S.S. Al-Hatmi, Z.S. Gano, Ionic liquids analogues based on potassium carbonate, Thermochimica Acta, 575 (2014) 135-143. M. Francisco, A. van den Bruinhorst, M.C. Kroon, Low-Transition-Temperature Mixtures (LTTMs): A New Generation of Designer Solvents, Angewandte Chemie International Edition, 52 (2013) 3074-3085.

[7] [8]

[9]

[10]

[11] [12]

[13]

[14]

[15]

[16] [17]

[18]

[19]

[20]

[21]

[22]

C.A. Nkuku, R.J. LeSuer, Electrochemistry in deep eutectic solvents, The Journal of Physical Chemistry B, 111 (2007) 13271-13277. Z. Maugeri, P. Dominguez de Maria, Novel choline-chloride-based deep-eutectic-solvents with renewable hydrogen bond donors: levulinic acid and sugar-based polyols, RSC Advances, 2 (2012) 421-425. Y. Dai, J. van Spronsen, G.-J. Witkamp, R. Verpoorte, Y.H. Choi, Natural deep eutectic solvents as new potential media for green technology, Analytica Chimica Acta, 766 (2013) 6168. H. Wang, Y. Jia, X. Wang, Y. Yao, Y. Jing, Physical–chemical properties of nickel analogs ionic liquid based on choline chloride, Journal of Thermal Analysis and Calorimetry, 115 (2014) 1779-1785. Y.-T. Liu, Y.-A. Chen, Y.-J. Xing, Synthesis and characterization of novel ternary deep eutectic solvents, Chinese Chemical Letters, 25 (2014) 104-106. L.L. Sze, S. Pandey, S. Ravula, S. Pandey, H. Zhao, G.A. Baker, S.N. Baker, Ternary Deep Eutectic Solvents Tasked for Carbon Dioxide Capture, ACS Sustainable Chemistry & Engineering, 2 (2014) 2117-2123. T. Khezeli, A. Daneshfar, Synthesis and application of magnetic deep eutectic solvents: Novel solvents for ultrasound assisted liquid-liquid microextraction of thiophene, Ultrasonics Sonochemistry, 38 (2016) 590-597. S. Sarkar, S. Sampath, Ambient temperature deposition of gallium nitride/gallium oxynitride from a deep eutectic electrolyte, under potential control, Chemical Communications, 52 (2016) 6407-6410. G. Imperato, S. Hoger, D. Lenoir, B. Konig, Low melting sugar-urea-salt mixtures as solvents for organic reactions-estimation of polarity and use in catalysis, Green Chemistry, 8 (2006) 1051-1055. F. Ilgen, B. Konig, Organic reactions in low melting mixtures based on carbohydrates and lcarnitine-a comparison, Green Chemistry, 11 (2009) 848-854. H. Ghaedi, M. Ayoub, S. Sufian, S.M. Hailegiorgis, G. Murshid, S. Farrukh, S.N. Khan, Experimental and prediction of volumetric properties of aqueous solution of (allyltriphenylPhosphonium bromide—Triethylene glycol) deep eutectic solvents, Thermochimica Acta, 657 (2017) 123-133. H. Ghaedi, M. Ayoub, S. Sufian, A.M. Shariff, G. Murshid, S.M. Hailegiorgis, S.N. Khan, Density, excess and limiting properties of (water and deep eutectic solvent) systems at temperatures from 293.15K to 343.15K, Journal of Molecular Liquids, 248 (2017) 378-390. H. Ghaedi, M. Ayoub, S. Sufian, B. Lal, Y. Uemura, Thermal stability and FT-IR analysis of Phosphonium-based deep eutectic solvents with different hydrogen bond donors, Journal of Molecular Liquids, 242 (2017) 395-403. H. Ghaedi, M. Ayoub, S. Sufian, S.M. Hailegiorgis, S. Krishnan, Toxicity of Several Potassium Carbonate and Phosphonium-Based Deep Eutectic Solvents towards Escherichia coli and Listeria monocytogenes Bacteria, Journal of Environtal & Analytical Toxicology, 7 (2017) 2161-0525. H. Ghaedi, M. Ayoub, S. Sufian, S.M. Hailegiorgis, G. Murshid, S.N. Khan, Thermal stability analysis, experimental conductivity and pH of phosphonium-based deep eutectic solvents and their prediction by a new empirical equation, The Journal of Chemical Thermodynamics, 116 (2018) 50-60. H. Ghaedi, M. Ayoub, S. Sufian, A.M. Shariff, B. Lal, The study on temperature dependence of viscosity and surface tension of several Phosphonium-based deep eutectic solvents, Journal of Molecular Liquids, 241 (2017) 500-510.

[23]

[24]

[25]

[26] [27] [28]

[29]

[30]

[31]

[32]

[33] [34]

[35]

[36]

[37]

[38]

H. Ghaedi, M. Ayoub, S. Sufian, B. Lal, A.M. Shariff, Measurement and correlation of physicochemical properties of phosphonium-based deep eutectic solvents at several temperatures (293.15 K to 343.15 K) for CO2 capture, The Journal of Chemical Thermodynamics, 113 (2017) 41-51. H. Ghaedi, M. Ayoub, S. Sufian, G. Murshid, S. Farrukh, A.M. Shariff, Investigation of various process parameters on the solubility of carbon dioxide in phosphonium-based deep eutectic solvents and their aqueous mixtures: Experimental and modeling, International Journal of Greenhouse Gas Control, 66 (2017) 147-158. H. Ghaedi, M. Ayoub, S. Sufian, A.M. Shariff, S.M. Hailegiorgis, S.N. Khan, CO2 capture with the help of Phosphonium-based deep eutectic solvents, Journal of Molecular Liquids, 243 (2017) 564-571. G. García, S. Aparicio, R. Ullah, M. Atilhan, Deep Eutectic Solvents: Physicochemical Properties and Gas Separation Applications, Energy & Fuels, 29 (2015) 2616-2644. M.K. AlOmar, M. Hayyan, M.A. Alsaadi, S. Akib, A. Hayyan, M.A. Hashim, Glycerol-based deep eutectic solvents: Physical properties, Journal of Molecular Liquids, 215 (2016) 98-103. D.M. Cristancho, D.R. Delgado, F. Martínez, M.A. Abolghassemi Fakhree, A. Jouyban, Volumetric properties of glycerol + water mixtures at several temperatures and correlation with the Jouyban-Acree model, Revista Colombiana de Ciencias Químico - Farmacéuticas, 40 (2011) 92-115. G.I. Egorov, D.M. Makarov, A.M. Kolker, Densities and Volumetric Properties of Ethylene Glycol + Dimethylsulfoxide Mixtures at Temperatures of (278.15 to 323.15) K and Pressures of (0.1 to 100) MPa, Journal of Chemical & Engineering Data, 55 (2010) 3481-3488. W. Afzal, A.H. Mohammadi, D. Richon, Volumetric Properties of Mono-, Di-, Tri-, and Polyethylene Glycol Aqueous Solutions from (273.15 to 363.15) K: Experimental Measurements and Correlations, Journal of Chemical & Engineering Data, 54 (2009) 12541261. Q.-S. Li, M.-G. Su, S. Wang, Densities and Excess Molar Volumes for Binary Glycerol + 1Propanol, + 2-Propanol, + 1,2-Propanediol, and + 1,3-Propanediol Mixtures at Different Temperatures, Journal of Chemical & Engineering Data, 52 (2007) 1141-1145. K.R. Siongco, R.B. Leron, M.-H. Li, Densities, refractive indices, and viscosities of N,Ndiethylethanol ammonium chloride–glycerol or –ethylene glycol deep eutectic solvents and their aqueous solutions, The Journal of Chemical Thermodynamics, 65 (2013) 65-72. C.L. Yaws, Chemical Properties Handbook, McGraw-Hill Company, New York, 1999, pp. 185. O. Ciocirlan, O. Croitoru, O. Iulian, Density and Refractive Index of Binary Mixtures of Two 1-Alkyl-3-methylimidazolium Ionic Liquids with 1,4-Dioxane and Ethylene Glycol, Journal of Chemical & Engineering Data, 59 (2014) 1165-1174. M. Moosavi, A.A. Rostami, Densities, Viscosities, Refractive Indices, and Excess Properties of Aqueous 1,2-Etanediol, 1,3-Propanediol, 1,4-Butanediol, and 1,5-Pentanediol Binary Mixtures, Journal of Chemical & Engineering Data, 62 (2017) 156-168. A. Marchetti, C. Preti, M. Tagliazucchi, L. Tassi, G. Tosi, The N,N-dimethylformamide + ethane-1,2-diol solvent system. Dielectric constant, refractive index, and related properties at various temperatures, Journal of Chemical & Engineering Data, 36 (1991) 365-368. U.S. Vural, V. Muradoglu, S. Vural, Excess molar volumes, and refractive index of binary mixtures of glycerol+ methanol and glycerol+ water at 298.15 K and 303.15 K, Bulletin of the Chemical Society of Ethiopia, 25 (2011) 111-118. V. Aniya, D. De, A. Singh, B. Satyavathi, Isobaric Phase Equilibrium of tert-Butyl Alcohol + Glycerol at Local and Subatmospheric Pressures, Volumetric Properties, and Molar

[39]

[40]

[41]

Refractivity from 303.15 to 333.15 K of tert-Butyl Alcohol + Glycerol, tert-Butyl Alcohol + Water, and Water + Glycerol Binary Systems, Journal of Chemical & Engineering Data, 61 (2016) 1850-1863. R.B. Leron, A.N. Soriano, M.-H. Li, Densities and refractive indices of the deep eutectic solvents (choline chloride + ethylene glycol or glycerol) and their aqueous mixtures at the temperature ranging from 298.15 to 333.15 K, Journal of the Taiwan Institute of Chemical Engineers, 43 (2012) 551-557. Arun B. Nikumbh, M.V. Rathi, Study of molar refraction and polarisability constant of aqueous solutions of NH4NO3 and KBrO3 at different temperatures, AJER, 5 (2016) 195200. M. Tariq, P.A.S. Forte, M.F.C. Gomes, J.N.C. Lopes, L.P.N. Rebelo, Densities and refractive indices of imidazolium- and phosphonium-based ionic liquids: Effect of temperature, alkyl chain length, and anion, The Journal of Chemical Thermodynamics, 41 (2009) 790-798.

Supporting Information Density and refractive index measurement of transition-temperature mixture (deep eutectic analogues) based on potassium carbonate with dual hydrogen bond donors

Hosein Ghaedi, Muhammad Ayoub, Suriati Sufian, Azmi Mohd Shariff, Bhajan Lal, Cecilia Devi Wilfred

Predicted density by Eq. (1)/g·cm-3

1.38 1.35 1.32 1.29 1.26 1.23 1.2 1.17 1.17 1.20 1.23 1.26 1.29 1.32 1.35 1.38 Experimental density data/g·cm-3 Figure S.1. Comparison between the experimental and

calculated values of density, ρ, for all TTMs and TTTMs. Symbols represent experimental results and dark lines denote fitted values.

Table S.1

Calculated molar volume, V, of TTMs and TTTMs within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a TTM1 T/K 293.15 298.15 303.15 308.15 313.15 318.15 323.15 328.15

54.7845 54.9272 55.0714 55.2164 55.3626 55.5101 55.6592 55.8100

TTM2

TTTM1

TTM3

TTM4

TTTM2

54.8711 55.0186 55.1737 55.3135 55.4809 55.6452 55.7954 55.9681

V/(m3·mol-1) 58.0258 58.1791 58.3356 58.4934 58.6516 58.8116 58.9734 59.1366

70.2681 70.4192 70.5720 70.7259 70.8811 71.0369 71.1950 71.3554

72.2297 72.4295 72.6303 72.8158 73.0304 73.2292 73.4576 73.6740

72.7338 72.8989 73.0664 73.2364 73.4083 73.5862 73.7620 73.9410

55.9626 56.1181 333.15 338.15 56.1151 56.2823 343.15 56.2693 56.4451 a The standard uncertainties are u (T) =

59.3017 71.5219 73.8974 74.1226 59.4668 71.6837 74.1163 74.3033 59.6337 71.8517 74.3424 74.4867 ±0.03 K and u(P) = 1 kPa. The TTM and TTTM

compositions are given in Table 2 and standard uncertainty for TTM and TTTM mass fraction composition is 0.002. The relative standard uncertainties are ur (V) = 0.0031 for TTM1, ur (V) = 0.0028 for TTM2, ur (V) = 0.0027 for TTTM1, ur (V) = 0.0022 for TTM3, ur (V) = 0.0029 for TTM4, ur (V) = 0.0024 for TTTM2.

Predicted nD by Equation (5)

1.49 1.48 1.47 1.46 1.45 1.44 1.43

1.42 1.42 1.43 1.44 1.45 1.46 1.47 1.48 1.49 Experimental nD Figure S.2. Comparison between the experimental nD and calculated nD by Eq. (5) for all TTMs and TTTMs. Symbols represent experimental results and dark lines denote fitted values. The symbols are the same as symbols depicted in Figure 3.

Table S.2

Calculated molar refraction, Rm, of TTMs and TTTMs within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa. a TTM1

TTM2

TTTM1

TTM3

Rm /(cm3.mol-1) 15.6815 20.1401

TTM4

TTTM2

20.4460

20.6018

T/K 293.15

14.6635

14.5464

298.15

14.6684

14.5477

15.6854

20.1473

20.4623

20.6104

303.15

14.6720

14.5544

15.6900

20.1567

20.4774

20.6176

308.15

14.6759

14.5460

15.6936

20.1649

20.4869

20.6255

313.15

14.6795

14.5556

15.6977

20.1732

20.5069

20.6332

318.15

14.6830

14.5603

15.7019

20.1873

20.5202

20.6468

323.15

14.6838

14.5602

15.7055

20.1962

20.5444

20.6597

328.15

14.6885

14.5682

15.7085

20.2072

20.5608

20.6676

333.15

14.6915

14.5667

15.7101

20.2139

20.5809

20.6750

338.15

14.6962

14.5752

15.7131

20.2249

20.5977

20.6829

343.15

14.6991

14.5794

15.7147

20.2317

20.6166

20.6904

a

The standard uncertainties are u (T) = ±0.03 K and u(P) = 1 kPa. The TTM and TTTM

compositions are given in Table 2 and standard uncertainty for TTM and TTTM mass fraction composition is 0.002. The relative standard uncertainties are ur (Rm) = 0.0003 for

TTM1, ur (Rm) = 0.0002 for TTM2, ur (Rm) = 0.0002 for TTTM1, ur (Rm) = 0.0005 for TTM3, ur (Rm) = 0.0008 for TTM4, ur (Rm) = 0.0004 for TTTM2.

List of Figures

Potassium Carbonate (PC)

Ethylene glycol (EG)

Glycerol (GL)

2-amino-2methyl-1-3-propanediol (AMPD)

Figure 1. A schematic of chemical structure of the salt and HBDs used in this work.

Figure 2. DSC curves of TTMs and TTTMs in this work. Note t/oC

1.45

ρ/g.cm-3

1.37 1.29 1.21 1.13 1.05 291.15

304.15

317.15

330.15

343.15

T/ K Figure 3. Experimental density, ρ, of the pure GL, pure EG, TTMs

and TTTMs in this work, as a function of temperature: (■), pure GL; (□), pure EG; (○), TTM1; (∆), TTM2; (◊), TTTM1; (●), TTM3; (▲), TTM4; (♦), TTTM2 in this work ; and (*), PC-GL 1:6; (), PCEG 1:6 reported by Mjalli et al. [5].

(b) 1.50

1.36 1.32 1.28 1.24 1.20 1.16 1.12 1.08 1.04 1.00

Density of mixture (ρ/g.cm-3)

Density of mixture (ρ/g.cm-3)

(a)

0

0.1

0.2

0.3

1.45 1.40 1.35 1.30 1.25 1.20 0

Mass fraction of PC

0.1 0.2 Mass fraction of PC

Temperature T/K:

0.3

Temperature T/K: 293.15 298.15 303.15 293.15 298.15 303.15 308.15 313.15 318.15 308.15 313.15 318.15 323.15 328.15 333.15 323.15 328.15 333.15 338.15 343.15 338.15 343.15 Figure 4. Composition behaviour of density for (a), PC-EG mixture and (b), PC-GL mixture. The red lines represent the data taken from Ref. [5]. 78

V / (cm3.mol-1)

74 70 66 62 58 54 50 291.15

304.15

317.15

330.15

343.15

T/K Figure 5. Molar volume, V, of TTMs and TTTMs as a function of temperature. The symbols are the same as symbols depicted in Figure 3.

0.000605 0.00058 αP / K-1

0.000555 0.00053 0.000505 0.00048 0.000455 0.00043 291.15 304.15 317.15 330.15 343.15 T/K Figure 6. Isobaric thermal expansion, αp, of TTMs

and TTTMs at several temperatures. The symbols are the as symbols depicted in Figure 3.

Refractive index / nD

1.49 1.47 1.45 1.43 1.41 291.15 304.15 317.15 330.15 343.15 T/ K Figure 7. Refractive index, nD, as a function of temperature

for pure GL, pure EG, TTMs and TTTMs in this work and PC-GL 1:6 and PC-EG 1:6 reported by Mjalli et al. [5]. The symbols are the same as symbols depicted in Figure 3.

(b) 1.493

1.463

Refractive index of mixture / nD

Refractive index of mixture / nD

(a)

1.455 1.448 1.440 1.433 1.425 1.418 1.410 0 0.05 0.1 0.15 0.2 0.25 0.3

1.486 1.478 1.471 1.463 1.456 0

0.1

0.2

0.3

Mass fraction of PC Mass fraction of PC Temperature T/K: Temperature T/K: 293.15 298.15 303.15 293.15 298.15 303.15 308.15 313.15 318.15 308.15 313.15 318.15 323.15 328.15 333.15 323.15 328.15 333.15 338.15 343.15 338.15 343.15 Figure 8. Composition behaviour of refractive index for (a), PC-EG mixture and (b), PC-GL mixture. The red lines represent the data taken from Ref. [5].

Electronic polarization , E

2.25 2.20 2.15 2.10 2.05 2.00 291.15

304.15

317.15

330.15

343.15

T/K Figure 9. Electronic polarization, E, of TTMs and TTTMs with various molar ratios at several temperatures. The symbols are the same as symbols depicted in Figure 3.

0.2200

RD / g-1.cm3

0.2180 0.2160 0.2140 0.2120 0.2100 0.2080 291.15

304.15

317.15

330.15

343.15

T/ K Figure 10. Specific refraction, RD, of TTMs and TTTMs with various molar ratio at several temperatures. The symbols are the same as symbols depicted in Figure 3.

Rm / cm3.mol-1

22

20

18

16

14 291.15

304.15

317.15

330.15

343.15

T/K Figure 11. Molar refraction, Rm, of TTMs and

TTTMs with various molar ratio at several temperatures. The symbols are the same as symbols depicted in Figure 3.

List of Tables Table 1

Chemical name and Linear formula Potassium carbonate K2CO3 2-amino-2methyl-1-3propanediol (HOCH2)2C(NH2)CH3

Specification of chemical samples used in this study. CAS No. Supplier Supplier Water purity a content 584-08-7 Sigma-Aldrich >0.999 None b 115-69-5

Sigma-Aldrich

>0.99

None b

Melting point 891 °C d 109-111 °C d

Glycerol 56-81-5 R & M Chemicals > 0.998 < 0.0030 c 18.0 °C e CH2OH-CHOH-CH2OH Ethylene glycol 107-21-1 R & M Chemicals >0.995 < 0.0025 c -13.0 °C e HOCH2CH2OH a The mass fraction purity of all components are provided by the suppliers. b Supplier did not provide this information. c This number is water content in mass fraction and was measured by Karl Fischer (KF) method. d Provided by supplier. e This data taken from safety data sheet (SDS). f Gas chromatography.

Purity Analysis Trace Metals Analysis Titration with HClO4 GC f GC f

Table 2 The abbreviations, molecular mass M and mole ratios for TTMs, TTTMs and their single components in this work.a TTM/TTTM Salt HBD Mole ratio Mass fraction (wsalt and wHBD) b c d c e Symbol MTTM/TTTM/ Abb. Msalt/ Abb. MHBD/ Salt HBD salt HBD g·mol-1

PC-EG (TTM1) PC-EG (TTM2) PC-EG- AMPD (TTTM1) PC-GL(TTM3) PC-GL(TTM4) PC-GL- AMPD (TTTM2) a

g·mol-1

Glass transition temperature (Tg/oC) m

0.0060 0.0054 0.0068

-121.5 -112.5 -83.3

0.0044 0.0059 0.0071

-90.5 -82.0 -121.5

g·mol-1

68.992 66.549 72.004

PC PC PC

138.21 138.21 138.21

EG EG EG & AMPD

96.283 94.803 95.377

PC PC PC

138.21 138.21 138.21

GL GL GL & AMPD

62.07 62.07 62.07 105.14 92.09 92.09 92.09 105.14

1 1 1

10 16 10:1f

0.1821 0.1222 0.1599

1 1 1

10 16 16:1g

0.1305 0.0858 0.0805

The standard uncertainty for TTM and TTTM mass fraction composition is 0.0071. MTTM/TTTM molar mass of TTM and TTTM in g·mol−1. c Abbreviation. d Molar mass of salt in g· mol−1. e Molar mass of HBD in g·mol−1. f (10) is mole ratio of EG and (1) is mole ratio of AMPD. g (16) is mole ratio of GL and (1) is mole ratio of AMPD. h Mass fraction of EG in TTTM1. i Mass fraction of AMPD in TTTM1. j Mass fraction of GL in TTTM2. k Mass fraction of AMPD in TTTM2. l Water content of samples in mass fraction. m The standard uncertainty in glass transition temperature is 0.2 °C. b

Water content l

0.8179 0.8778 0.7184 h 0.1217 i 0.8695 0.9142 0.8583 j 0.0612 k

Table 3 Comparison of the experimental and the literature data of the density and refractive index of EG and GL.a

Ethylene glycol (EG) Lit. %AAD b Exp. -3 Density ρ/g·cm 293.15 1.1133 1.1133 [25] 0.0023 1.2620 298.15 1.1098 1.1098 [24] 1.2590 303.15 1.1063 1.1063 [25] 1.2558 308.15 1.1028 1.1028 [24] 1.2527 313.15 1.0992 1.0992 [25] 1.2496 323.15 1.0921 1.0921 [24] 1.2432 328.15 1.0885 1.0884 [25] 1.2400 333.15 1.0849 1.0848 [25] 1.2367 Refractive index 298.15 1.43001 1.4300 [27] 0.0082 1.47182 303.15 1.42853 1.4285 [27] 1.47035 308.15 1.42709 1.4269 [30] 1.46887 313.15 1.42567 1.4258 [27] 1.46747 318.15 1.42437 1.4244 [30] 1.46597 323.15 1.42309 1.4229 [31] 1.46456 328.15 1.42155 1.4213 [31] 1.46312 333.15 1.42010 1.4200 [31] 1.46161 a The standard uncertainties are u(P) = 1 kPa and u (T) for ρ= ±0.01 T/K

Exp.

Glycerol (GL) Lit. %AAD b Density ρ/g·cm-3 1.2611 [23] 0.0270 1.2589 [26] 1.2556 [26] 1.2527 [26] 1.2495 [26] 1.2429 [26] 1.2396 [26] 1.2360 [26] Refractive index 1.46970 [28] 0.0414 1.47185 [29] 1.46941 [29] 1.46727 [29] 1.46573 [29] 1.46448 [29] 1.46323 [29] 1.46169 [29] K and for nD= ±0.03 K. The

relative standard uncertainties in density are ur (ρ) = 0.0033 for EG, ur (ρ) = 0.0026 for GL and those of refractive index are ur (nD) = 0.0015 for EG, ur (nD) = 0.00099 for GL. b

%AAD is the average absolute deviation and calculated: % AAD =

100 i = n Yexp − Ylit/pred ×∑ n Yexp i =1

where Yexp and Ylit/pred are experimental results and literature or predicted data, respectively; n represents the number of data point.

Table 4

Experimental density, ρ, of pure GL, pure EG, TTMs and TTTMs. Measurements were conducted within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a Pure EG T/K 293.15

TTM1

TTM2

TTTM1

Pure GL

TTM3

TTM4

TTTM2

1.2620

1.3702

1.3125

1.3113

-3

1.1133

1.2593

1.2128

ρ/g·cm 1.2409

298.15 303.15 308.15 313.15 318.15 323.15 328.15 333.15 338.15 343.15 a

1.1098 1.1063 1.1028 1.0992 1.0957 1.0921 1.0885 1.0849 1.0812 1.0775

1.2561 1.2528 1.2495 1.2462 1.2429 1.2395 1.2362 1.2328 1.2295 1.2261

1.2096 1.2062 1.2031 1.1995 1.1960 1.1927 1.1891 1.1859 1.1824 1.1790

1.2376 1.2343 1.2310 1.2277 1.2243 1.2210 1.2176 1.2142 1.2108 1.2074

1.2590 1.2558 1.2527 1.2496 1.2464 1.2432 1.2400 1.2367 1.2334 1.2301

1.3673 1.3643 1.3614 1.3584 1.3554 1.3524 1.3493 1.3462 1.3432 1.3400

1.3089 1.3053 1.3020 1.2981 1.2946 1.2906 1.2868 1.2829 1.2791 1.2752

1.3084 1.3054 1.3023 1.2993 1.2961 1.2930 1.2899 1.2868 1.2836 1.2805

The standard uncertainties are u (T) = ±0.01 K and u(P) = 1 kPa. The TTM and TTTM compositions are

given in Table 2 and standard uncertainty for TTM and TTTM mass fraction composition is 0.002. The relative standard uncertainties are ur (ρ) = 0.0033 for EG, ur (ρ) = 0.0027 for TTM1, ur (ρ) = 0.0028 for TTM2, ur (ρ) = 0.0027 for TTTM1, ur (ρ) = 0.0026 for GL, ur (ρ) = 0.0022 for TTM3, ur (ρ) = 0.0029 for TTM4, ur (ρ) = 0.0024 for TTTM2.

Table 5 The constant parameters of a, b used in Eq. (1), %AAD, RMSE and R2 values. TTM/TTTM a 103 × b 105 × RMSE a 102 × %AAD b R2 TTM1 1.454 -0.6646 5.926 0.99999 0.1780

a

TTM2

1.412

-0.6791

14.4

0.9998

TTTM1 TTM3

1.437 1.547

-0.6695 -0.6032

5.615 10.59

0.1231

0.99999 0.9999

TTM4

1.532

-0.7466

26.4

0.2224

0.9996

TTTM2

1.493

-0.6182

10.61

0.1911

0.9999

0.0831 0.1914

 ∑ (Yexp − Ypred )2  RMSE is root mean square and calculated by: RMSE =   K −L  

0.5

where Yexp and Ypred are experimental and predicted data, respectively; K equals the number of experimental data points, and L equals the number of parameters. b

% AAD value is calculated according to the formula presented in Table 3.

Table 6

Experimental refractive index, nD, of pure EG, pure GL, TTMs and TTTMs at several temperatures. Measurements were conducted within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a Pure EG

TTM1

TTM2

TTTM1

Pure GL

TTM3

TTM4

TTTM2

Refractive index / nD

T/K

293.15 1.43146 1.44791 1.44298 1.45293 1.47333 1.48503 1.47801 1.47836 298.15 1.43002 1.44674 1.44166 1.45168 1.47182 1.48401 1.47691 1.47733 303.15 1.42853 1.44552 1.44046 1.45043 1.47035 1.48305 1.47578 1.47624 308.15 1.42709 1.44431 1.43889 1.44915 1.46887 1.48205 1.47462 1.47517 313.15 1.42567 1.44308 1.43770 1.44788 1.46747 1.48105 1.47353 1.47407 318.15 1.42437 1.44184 1.43638 1.44662 1.46597 1.48020 1.47239 1.47310 323.15 1.42309 1.44051 1.43503 1.44532 1.46456 1.47920 1.47132 1.47213 328.15 1.42155 1.43930 1.43376 1.44400 1.46312 1.47825 1.47015 1.47101 333.15 1.42010 1.43803 1.43238 1.44263 1.46161 1.47713 1.46903 1.46986 338.15 1.41868 1.43682 1.43122 1.44131 1.46029 1.47618 1.46786 1.46874 343.15 1.41733 1.43555 1.42994 1.43994 1.45867 1.47506 1.46670 1.46759 a The standard uncertainties are u (T) = ±0.03 K and u(P) = 1 kPa. The TTM and TTTM compositions are given in Table 2 and standard uncertainty for TTM and TTTM mass fraction composition is 0.002. The relative standard uncertainties are ur (nD) = 0.0015 for EG, ur (nD) = 0.00086 for TTM1, ur (nD) = 0.00091 for TTM2, ur (nD) = 0.00090 for TTTM1, ur (nD) = 0.00099 for GL, ur (nD) = 0.00067 for TTM3, ur (nD) = 0.00077 for TTM4, ur (nD) = 0.00073 for TTTM2.

Table 7 Estimated constant parameters of Eqs. (5) and (6), %AAD, RMSE and R2.

A0 A1 A2 5 10 ×RMSE a

TTM1

TTM2

TTTM1

TTM3

TTM4

TTTM2

-3.829 -126.2 14190 3.44

Eq. (5) -3.723 -3.863 -146.7 -119.6 17122 13060 14.46 2.65

-4.255 -111.1 12640 3.48

-4.071 -180.7 21860 12.06

-4.182 -123.6 14230 4.40

%AAD b R2

0.0036 0.99999

a 103×b 5 10 ×RMSE a %AAD b R2

1.521 -0.2483 4.373 0.0154 0.9999

0.0111 0.0017 0.9996 0.99999 Eq. (6) 1.520 1.529 -0.262 -0.2593 8.17 6.942 0.0170 0.0053 0.9996 0.9997

0.0106 0.99999

0.0087 0.9998

0.0061 0.9999

1.543 -0.1971 7.204 0.0123 0.9995

1.544 -0.2257 3.123 0.0134 0.9999

1.541 -0.2139 7.59 0.0074 0.9996

a

RMSE value is calculated according to the formula presented in Table 5.

b

% AAD value is calculated according to the formula presented in Table 3.

Table 8 Calculated free volume, fm, of TTMs and TTTMs within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa. a

TTM1

TTM2

TTTM1

TTM3

TTM4

TTTM2

50.1280

51.7837

52.1319

T/K 293.15

40.1210

fm/(cm3·mol-1) 40.3248 42.3442

298.15

40.2587

40.4709

42.4937

50.2719

51.9672

52.2885

303.15

40.3994

40.6193

42.6456

50.4153

52.1530

52.4488

308.15

40.5405

40.7675

42.7998

50.5610

52.3290

52.6109

313.15

40.6832

40.9254

42.9538

50.7079

52.5234

52.7751

318.15

40.8271

41.0848

43.1096

50.8496

52.7090

52.9393

323.15

40.9754

41.2352

43.2679

50.9989

52.9133

53.1023

328.15

41.1216

41.3998

43.4281

51.1483

53.1132

53.2734

333.15

41.2711

41.5515

43.5916

51.3080

53.3165

53.4475

338.15

41.4189

41.7070

43.7536

51.4588

53.5186

53.6204

343.15

41.5702

41.8658

43.9191

51.6200

53.7258

53.7963

a

The standard uncertainties are u (T) = ±0.03 K and u(P) = 1 kPa. The TTM and TTTM

compositions are given in Table 2 and standard uncertainty for TTM and TTTM mass

fraction composition is 0.002. The relative standard uncertainties are ur (fm) = 0.0042 for TTM1, ur (fm) = 0.0038 for TTM2, ur (fm) = 0.0037 for TTTM1, ur (fm) = 0.0029 for TTM3, ur (fm) = 0.0037 for TTM4, ur (fm) = 0.0031 for TTTM2.

Table 9

Calculated polarizability constant, δ, of TTMs and TTTMs within temperature range from T = (293.15 to 343.15) K and P = 0.1 MPa.a TTM1

TTM2

T/K

a

TTTM1 23

3

TTM3

TTM4

TTTM2

-1

10 δ/(cm ·mol )

293.15

1.6848

1.3099

1.2064

2.0281

1.7074

1.6111

298.15

1.6856

1.3102

1.2066

2.0289

1.7078

1.6113

303.15

1.6865

1.3105

1.2070

2.0297

1.7081

1.6115

308.15

1.6874

1.3108

1.2072

2.0306

1.7085

1.6120

313.15

1.6883

1.3111

1.2076

2.0314

1.7088

1.6124

318.15

1.6892

1.3113

1.2078

2.0322

1.7092

1.6127

323.15

1.6901

1.3117

1.2080

2.0329

1.7095

1.6132

328.15

1.6909

1.3120

1.2081

2.0337

1.7099

1.6136

333.15

1.6918

1.3123

1.2084

2.0343

1.7102

1.6140

338.15

1.6927

1.3126

1.2088

2.0351

1.7105

1.6144

343.15

1.6936

1.3129

1.2090

2.0359

1.7109

1.6148

The standard uncertainties are u (T) = ±0.03 K and u(P) = 1 kPa. The TTM and TTTM

compositions are given in Table 2 and standard uncertainty for TTM and TTTM mass fraction composition is 0.002. The relative standard uncertainties are ur (δ) = 0.0006 for TTM1, ur (δ) = 0.0002 for TTM2, ur (δ) = 0.0002 for TTTM1, ur (δ) = 0.0004 for TTM3, ur (δ) = 0.0002 for TTM4, ur (δ) = 0.0002 for TTTM2.

Highlights •

New potassium carbonate-based TTTMs were prepared with dual HBDs.

•

As the temperature increased the density and refractive index decreased.

•

TTMs with higher amount of HBD had the lower density and refractive index values.

•

As the molecular weight of TTMs decreased the density and refractive index decreased.

•

TTMs with GL in their structures had the higher density refractive index values.