The Effect of Sodium and Chloride Salts on Tetrahydrofuran Hydrate Formation by Using a Differential Scanning Calorimetry

1. Introduction

Clathrate hydrates are inclusion compounds in which small guest molecules are captured in water cavities made up of hydrogen-bonded water molecules. Typical guest molecules include methane, ethane, carbon dioxide and some liquids that are stabilized in the cavities of sI, sII and sH hydrate via van der Waals interaction forces (Carroll, 2002; Erfani and Varaminian, 2017; Giavarini and Hester, 2011; Naeiji and Varaminian, 2013; Sloan, 2003; Sloan and Koh, 2008; Sun et al, 2011). Clathrate hydrates can be formed at low temperatures close to the freezing point of water and at high pressures that are also common in oil and gas transmission (Carroll, 2002; Mohebbi and Behbahani, 2014; Sloan, 2003; Sun et al, 2011; Vysniauskus and Bishnoi, 1983). Gas hydrates have been extensively investigated because of their importance in the petroleum industry especially because they are a major possible reason of pipeline blockage during the transport of gas or petroleum (Ellison et al., 2000; Hammerschmidt, 1934; Holder and Bishnoi, 2000; Karimi et al., 2014; Ng and Robinson, 1994).

Numerous techniques have been employed to investigate gas hydrates (Chazallon et al., 2007; Makogon, 1997). NMR and Raman spectroscopy are usually applied to characterize the gas hydrates (Chazallon et al., 2007; Kini and Sloan, 2002; Rovetto et al., 2008; Sum et al., 1997). Ripmeester and Ratcliffe (1990) have developed an NMR technique to assess the clathrate hydrate forming ability of new potential guest molecules and to determine their structure. Also, Gupta et al. (2007) have studied the methane hydrate dissociation mechanism on the molecular scale using NMR spectroscopy. They found a relationship between NMR methane gas chemical shift, pressure, and temperature of the system (Gupta et al., 2007). Hydrate properties, such as structure and composition have been measured using Raman spectroscopy reported by Hester et al. (2007). Moreover, Susilo et al. (2007) have synthesized and characterized structure I and H methane hydrate with PXRD, NMR, Raman spectroscopy and DSC.

Differential Scanning Calorimetry (DSC) is one of the most common techniques of thermal analysis of Physico-chemical transformation that it consists of monitoring the heat exchange between the sample and a reference. It has been used to study hydrates since 1980 (Giavarini and Hester, 2011). This has proved to be a valuable technique in the investigation of gas hydrate formation and dissociation. Dalmazzone et al. (2003) have been presented experiments using DSC to determine the thermodynamic equilibrium properties of methane hydrate in complex solutions. Lachance et al. (2008) have studied the effect of gas hydrate formation/dissociation on water-in-oil emulsions stability using DSC. In addition, it has been used to evaluate hydrate nucleation and kinetic hydrate inhibitor performance trends (McNamee, 2011).

It has been previously found that some liquids such as furan, tetrahydrofuran (THF) and 1,4-dioxane are non-gaseous hydrate former that they can form sII hydrate under atmospheric pressure (Chun and Lee, 2000; Ding et al., 2010; Dirdal et al., 2012; Erfani et al., 2015; Erfani and Varaminian, 2016; Gough and Davidson, 1971; Karamoddin and Varaminian, 2013, 2014a, 2014b; Naeiji et al., 2014a, 2014b). They can be used as suitable alternative molecules to study the gas hydrates without the requirement of highpressure equipment. On the other hand, their formation heat is comparable to the fusion heat of ice, so that they can be considered as suitable alternative materials in air-conditioning systems (Fournaison et al., 2004; Li et al., 2006; Mahmoudi et al., 2016a, 2016b; Marinhas et al., 2006). THF is a cyclic ether completely miscible with water that can form hydrate at 4 ^°C and atmospheric pressure (Karamoddin and Varaminian, 2014a; Naeiji et al., 2014a). DSC analysis has been also employed to determine the thermal properties and enthalpy of THF hydrate formation by Zhang et al. (2004). Hydrate formation conditions of water-THF-CO₂ systems have been investigated by DSC to show the effect of THF on equilibrium pressure and dissociation enthalpy of CO₂ hydrate applied to secondary refrigeration (Delahaye et al., 2006). Li et al. (2010, 2012, and 2013) have also widely studied on the hydrates as cold storage materials by using DSC. They have obtained the phase change temperature and fusion heat of tetra-n-butyl ammonium bromide (TBAB) and THF hydrate (Li et al., 2010).

In this study, it is tried to evaluate THF hydrate formation by the DSC technique. It has been previously reported by Parlouër et al. (2004) that additives influence the methane hydrate formation so that the introduction of sodium chloride and ethylene glycol allows to shift the dissociation temperature obtained from DSC at a higher value for a significative pressure. Similarly, Koh et al. (1998) have also used DSC to test hydrates inhibitors at atmospheric pressure. Consequently, the thermal properties including the heat of formation, phase change temperature and specific heat capacity of THF hydrate formation in the presence of sodium and chloride salts have been determined.

2. Experimental Method

2.1. Materials

The samples have been made using freshly deionized and distilled water, and tetrahydrofuran liquid (purity: 99.5 vol%) provided by the Merck company. The salts used are sodium chloride, sodium fluoride, sodium sulfite, potassium chloride and calcium chloride with a normal purity of 99.5% (Merck Co.).

2.2. Apparatus and Procedure

A DSC setup has been used in this study (see Figure 1). The experimental system consisted of two cylindrical stainless steel cells (about 1 cm³) placed into a copper segment. This segment includes three rectangular cube form sub-segments, so that sample and reference cells placed into the middle segment. The heating/cooling process is conducted by peltier effect, so that a 4×4 (cm×cm) peltier is in direct contact with the middle segment on the one side and with the lateral copper segment on another side. Input and output streams of refrigerant fluid (ethylene glycol solution-50 mol%) provided by a refrigerated circulator is inserted into the lateral copper segment to reduce the temperature of the hot surface of peltier. The copper segment placed into a plexiglass chamber which has been isolated by perlite granules and vacuum. Two platinum thermometers (Pt 100) with temperature accuracy of ±0.01 ^°C have been placed under each of the cells to monitor/control the temperature of the calorimeter. The temperature of the calorimetry is controlled through a temperature program with a cooling rate of 0.5 ^°C.min^-1.

For the test, a sample of 0.5±0.025 cc of the water-THF solution along with salt at the concentrations of 0.1-1 wt% has been prepared. THF forms structure II hydrate in the molar composition of THF.17H₂O at atmospheric condition (P ≈ 1 atm and low temperatures) (Naeiji et al., 2014a). The equation of the hydration reaction is as follows (Karamoddin and Varaminian, 2014a):

where Q is the heat of formation since THF hydrate formation is an exothermic process. The prepared sample has been injected into the sample cell. The reference cell has been prepared in the same way as the sample cell but without the sample. Then, the sample and reference cells have been placed in the specified cavities. The temperature ramp method (Hohne et al., 2003) has been used to cool the sample from 10 to -5 ^°C at 0.5 ^°C.min^-1. The runs have been repeated two or three times to observe the results repeatability.

It should be noticed that the calibration runs have to perform before the main experimental runs. In this work, the calibration runs were done based on the freezing/melting temperature and heat of fusion of ice, corresponding more closely with the experimental temperature range. Since the heat capacity of ice, C, and heating/cooling rate, β, is available, the true heat flow can be determined by (Hohne et al., 2003):

where Φ and K_Φ are the measured heat flow and the calibration factor, respectively. In the DSC, the measurement signal, ΔT, is obtained as electric voltage (with temperature accuracy of ±0.01 °C). The measured heat flow, Φ_m, is assigned to this signal by:

where K is given by the heat conduction path properties between the furnace and the sample. This factor was calculated by two decimals. The heat flow rate is also reported with two decimals so that its accuracy is 0.1 mW. Generally, the standard error of the value obtained from the DSC setup was around ±0.2-0.7.

Figure 1. Schematic diagram of the experimental apparatus of DSC.

3. Results and Discussion

The ramp method has been used to measure the thermal properties of THF hydrate formation with or without the presence of sodium and chloride salts. As previously mentioned, the temperature and heat calibration has been performed based on the thermophysical properties of ice. The DSC curve of pure ice is shown in Figure 2. Only one exothermic peak observes in the heat flow curve for only the single substance (ice) under the process of freezing by DSC. The phase change temperature of ice investigated by DSC is 273.05±0.20 K that the deviation is only ±0.10 K compared with the standard value of ice (273.15 K). According to Figure 2, the temperature of peak and heat of formation corresponding to the peak area of pure ice have been determined to be 271.84±0.22 K and 164.01±0.21 J respectively. So, the deviation for the heat of formation is 5.53 J/g compared with the standard value (333.55 J/g).

Figure 2. DSC freezing curve of ice.
(Tp: temperature of peak, Te: phase change temperature)

The DSC curve of THF hydrate formation is given in Figure 3. With regard to the THF hydrate, the phase change temperature is 276.36 K. Since THF and water are also mutually soluble, it is also easily explained by the fact that the phase change temperature is lower than the critical formation temperature point (277.55 K) in the THF hydrate phase equilibrium diagram (Sloan and Koh, 2008). The heat of formation that equals the enthalpy of formation is 66.15±0.47 J or 132.30 J/g that is comparable with that of reported in the literature (Li et al., 2010). Figure 4 shows the DSC curve of THF hydrate formation in the presence of sodium chloride at various concentrations (0.1-1 wt%). The results are also given in Table 1. It is observed that THF hydrate is inhibited in the presence of NaClso that the DSC peak shifts to lower temperature. In other words, a delay in the phase change temperature occurs, so that it approximately shifts to 272.30 K in 1.0 wt% NaCl solution. Besides, a broadening of the DSC peak can be seen, however the height of peak decreases. The results are shown in Table 1 also confirm that the heat of formation decreases in the presence of NaCl and more decreases by adding NaCl. The r_Q represents the reduction percent of formation heat of salt solution compared to that of without salt. Increasing this criterion reveals that salt solution can inhibit THF hydrate formation so that 1.0 wt% NaCl solution with r_Q of 41.35% represents the highest ability in the inhibition performance.

Figure 3. DSC curve of THF hydrate formation. (Tp: temperature of peak, Te: phase change temperature)

Figure 4. DSC curve of THF hydrate formation at various NaCl concentrations.

Table 1. The results of THF hydrate formation in the presence of NaCl.

(a) T_e: phase change temperature, (b) T_p: temperature of peak, (c) Q: formation heat, (d) r_Q: reduction percent of formation heat of salt solution compared to that of without salt, and (e) standard error.

The DSC curve of THF hydrate formation in the presence of sodium fluoride at various concentrations is represented in Figure 5. When NaF is added to the solution, the hydrate formation is postponed and the phase change temperature shifts from 276.36 K of without salt to 270.76 K of 1.0 wt% NaF aqueous solution. Similar to the NaCl solutions, it is clear that the height of peaks decreases and their widths is extended. So, it is to be expected the hydrate growth occurs slow. According to Table 2, the heat of formation of NaF solutions is lower than that of without salt. The criterion of r_Q represents a reduction of 7.80-31.43%.

Figure 6 and Table 3 are given the results of DSC for Na₂SO₃ solutions. It is found that Na₂SO₃ canalso act on the nucleation of THF hydrate, since it shifts the phase change temperature from 276.36 K of without salt to 270.87 K of 0.1 wt% Na₂SO₃ aqueous solutions, and decreases the heat flow rate. Contrary to NaCl and NaF solutions, an increase in the salt concentration causes an increase in the formation of heat and a decrease in r_Q criterion. So that the solution of 0.1 wt% Na₂SO₃ represents a reduction of 56.68% in the formation heat compared to that of without salt. It means that the inhibition performance of hydrate formation is weakened when the salt concentration increases. It is likely related to the fact that heavier salt may be precipitated as the hydrate formation proceeds.

Figure 5. DSC curve of THF hydrate formation at various NaF concentrations.

Table 2. The results of THF hydrate formation in the presence of NaF.

(a) T_e: phase change temperature, (b) T_p: temperature of peak, (c) Q: formation heat, (d) r_Q: reduction percent of formation heat of salt solution compared to that of without salt, and (e) standard error.

Figure 6. DSC curve of THF hydrate formation at various Na2SO3 concentrations.

Table 3. The results of THF hydrate formation in the presence of Na2SO3.

(a) T_e: phase change temperature, (b) T_p: temperature of peak, (c) Q: formation heat, (d) r_Q: reduction percent of formation heat of salt solution compared to that of without salt, and (e) standard error.

The results obtained from THF hydrate formation in the presence of potassium chloride and calcium chloride is represented in Figs. 7-8 and Table 4-5. Both chloride salts can postpone the hydrate formation and decrease the heat flow rate. It is clear that the phase change temperature shifts from 276.36 K of without salt to 271.74 K and 273.79 at 1.0 wt% potassium chloride and 0.1 wt% calcium chloride salt aqueous solutions, respectively. The criterion of r_Q has a maximum value of 56.28% at 1.0 wt% KCl solution, while it reduces as the concentration of CaCl₂ into the solution increases.

Comparison of sodium and chloride salts shows that they can influence both nucleation and growth of THF hydrate formation. It has been previously proved that electrolytes (salts) as additives disrupt hydrogen bonding in water and the freezing point of water is lowered, so that ice is no longer stable at 273.15 K and one can vary the amount of freezing point depression achieved based on the additive chosen (Giavarini and Hester, 2011). This change can be considered to predict hydrate formation condition like ice. Salts affect hydrate formation as thermodynamic inhibitors and decrease the stable hydrate temperature for a given pressure (Barduhn et al., 1962; Chun and Lee, 2000; Englezos and Bishnoi, 1988; Kubota et al., 1984).

The results show different performances of salts on the inhibition of THF hydrate formation. The salt solutions of NaF and Na₂SO₃ have exhibited the best performance on delaying nucleation of the hydrate. These salts can postpone the hydrate formation by lowering the phase change temperature from 276.36 K to about 270 K. On the contrary, KCl and CaCl₂ solutions have demonstrated that can reduce the heat of formation to around 65% compared to that of without salt. It can be explained by the fact that the dual moment of the clathrate of hydrate is theoretically zero and they are neutral. But, they are experimentally biased because the water protons locate irregularly at hydrate cages so that it causes a partial positive charge temporary assigns for each cage. Consequently, anions of salts are placed near the hydrate surface and can disrupt the hydrate clathrate, while cations of salts can prevent the accumulation of hydrate clusters in the growth stage. The anions of sodium salt solutions show a more effect on the disruption of water cages in the nucleation stage.

Park et al. (2011) have proved that the inhibition effect of salts on hydrate formation is directly related to the number of electrical charges, while it has an inverse relation to the ionic size. As seen in Tables 1-3, by comparing the temperature of peak and the phase change temperature, it is clear that the order of inhibition strength among chloride salts isNa₂SO₃ > NaF > NaCl. Also, the highest value of r_Q belongs to the Na₂SO₃ solution. Similarly, the inhibition effect of chloride salts on hydrate nucleation is: CaCl₂ > NaCl > KCl, as seen in Tables 1, 4 and 5. Also, the solution of CaCl₂ has the most reduction in the formation of heat compared to that of without salt. The obtained results are in agreement with the results reported by Karamoddin and Varaminian (2013). Moreover, the freezing point is one of the colligative properties of solutions that depend on the concentration of solute molecules or ions, and not on the type of chemical species (McQuarrie et al., 2011). The measurement of the difference between the freezing point of a pure solvent and that of a solution including some solutes is expressed as follows (McQuarrie et al., 2011):

where k_f is the freezing point depression constant and m is the molality. k_f is present for a well-known solvent like water, but it is unknown for THF hydrate formation process. So, it can be determined by reported data from DSC. Table 6 gives the freezing point depression constant, k_f, for sodium and chloride salt solutions. The large k_f implies that the given salt is an effective material for inhibiting hydrate formation. It can be found that the highest amount of the freezing point depression constant belongs to the solution of Na₂SO₃ and CaCl₂ among sodium and chloride salts, respectively.

Figure 7. DSC curve of THF hydrate formation at various KCl concentrations.

Table 4. The results of THF hydrate formation in the presence of KCl.

(a) T_e: phase change temperature, (b) T_p: temperature of peak, (c) Q: formation heat, (d) r_Q: reduction percent of formation heat of salt solution compared to that of without salt, and (e) standard error.

Figure 8. DSC curve of THF hydrate formation at various CaCl₂ concentrations.

Table 5. The results of THF hydrate formation in the presence of CaCl₂.

(a) T_e: phase change temperature, (b) T_p: temperature of peak, (c) Q: formation heat, (d) r_Q: reduction percent of formation heat of salt solution compared to that of without salt, and (e) standard error.

Table 6. The freezing point depression constant, k_f, of THF hydrate formation.

(a) standard error.