Phase behaviour and conductivity study of electrolytes in supercritical hydrofluorocarbons

Abstract

The purpose of this work was to characterise supercritical hydrofluorocarbons (HFC) that can be used as solvents for electrodeposition. The phase behaviour of CHF₃, CH₂F₂, and CH₂FCF₃ containing [NBuⁿ₄][BF₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄] was studied and the conditions for forming a single supercritical phase established. Although all three HFCs are good solvents for [NBuⁿ₄][BF₄] the results show that the CH₂F₂ system has the lowest p_r for dissolving a given amount of [NBuⁿ₄][BF₄]. The solubility of Na[B{3,5-C₆H₃(CF₃)₂}₄] in CH₂F₂ was found to be unexpectedly high. Studies of the phase behaviour of CH₂F₂ containing [NBuⁿ₄][BF₄] and [Cu(CH₃CN)₄][BF₄] showed that the copper complex was unstable in the absence of CH₃CN. For CHF₃, [Cu(hfac)₂] was more soluble and more stable than [Cu(CH₃CN)₄][BF₄] and only increased the phase-separation pressure by a moderate amount. Studies of the conductivity of [NBuⁿ₄][B(C₆F₅)₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], [NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄] (R_f = (CH₂)₃C₇F₁₅), and Na[B{3,5-C₆H₃(CF₃)₂}₄] were carried out in scCH₂F₂. The results show that these salts are more conducting than [NBuⁿ₄][BF₄] under the same conditions although the increase is much less significant than that reported in previous work in supercritical CO₂ + CH₃CN. Consequently, either [NBuⁿ₄][BF₄] or the corresponding BARF salts would be suitable background electrolytes for electrodeposition from scCH₂F₂.

---

1. Introduction

Supercritical fluids (fluids at temperatures and pressures above their critical point) are potentially attractive media for materials electrodeposition because they have low or zero surface tension enabling them to readily penetrate nanopores, they show enhanced mass transport rates,¹ and, given a suitable choice of fluid/electrolyte, they can have wide electrochemical windows² allowing the deposition of reactive materials at elevated temperatures. The properties of supercritical fluids (dielectric constant, viscosity, density, etc.) are intermediate between those of the gas and the liquid states and they can be varied by altering the temperature and pressure, making them “tuneable” solvents.³

Despite these potential advantages there are very few reports in the literature of electrodeposition from supercritical fluids because they pose a number of challenges. Electrodeposition places significant demands on the supercritical fluid in terms of the solubility of the reagents and conductivity provided by the supporting electrolyte in order to achieve deposition at meaningful rates and without significant problems caused by large uncompensated solution resistance. The major obstacle to carrying out electrodeposition in many supercritical fluids is the low dielectric constant of the fluid (ε < 8) which limits the solubility and conductivity for many electrolytes.⁴

In a recent publication we reported the first generally applicable method for electrodeposition from a homogeneous supercritical fluid.⁵ In that paper we described the electrodeposition of Cu, Ag and Co from a single-phase supercritical CO₂–MeCN. This was achieved by the use of appropriately designed metal complexes and a range of tailored electrolytes to give high solution conductivities.⁶ The inclusion of MeCN as a co-solvent limits the accessible potential range and introduces complications. The use of supercritical fluorinated hydrocarbons^2,7–16 is an attractive alternative which avoids these problems.

In this paper we present results of a study of the phase behaviour of trifluoromethane (CHF₃), difluoromethane (CH₂F₂), and 1,1,1,2-tetrafluoroethane (CH₂FCF₃) containing tetrabutylammonium tetrafluoroborate ([NBuⁿ₄][BF₄]), [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄]. In addition we present results for the phase behaviour of CH₂F₂ containing [NBuⁿ₄][BF₄] and [Cu(CH₃CN)₄][BF₄] and for CHF₃ containing [NBuⁿ₄][BF₄] and [Cu(hfac)₂]. Finally, we report electrical conductivity data for scCH₂F₂ with four BARF salts (i.e.[NBuⁿ₄][B(C₆F₅)₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], [NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄], and Na[B{3,5-C₆H₃(CF₃)₂}₄]) and compare these results with the published data⁸ for [NBuⁿ₄][BF₄].

The purpose of this work is to characterise supercritical fluids that can be used as electrolytes for electrodeposition of a range of materials and we therefore concentrate on finding the conditions which give single supercritical phases with good electrical conductivity.

2. Experimental

2.1 Materials

Trifluoromethane (CHF₃, Aldrich, 98+%), difluoromethane (CH₂F₂, INEOS Fluor Ltd, UK, 99.90 wt/wt%), and 1,1,1,2-tetrafluoroethane (CH₂FCF₃, Air Product, technical grade, 99.9 wt/wt%) were used as supplied. Tetrabutylammonium tetrafluoroborate [NBuⁿ₄][BF₄], Fluka, electrochemical grade, 99.0+%), tetrakis(acetonitrile)copper(I) tetrafluoroborate ([Cu(CH₃CN)₄][BF₄], Aldrich, 97%), acetonitrile (Fisher Scientific, 99.98+%) were used without further purification. Bis(hexafluoroacetyl acetonato)copper(II) hydrate ([Cu(hfac)₂]·H₂O) was supplied by Aldrich. [NBuⁿ₄][B(C₆F₅)₄], Na[B{3,5-C₆H₃(CF₃)₂}₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], and [{CF₃(CF₂)₇(CH₂)₃}NBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄] were synthesised and purified as described elsewhere.^5,6

2.2 Methods

The phase behaviour was studied by using a visual method in a variable-volume, view-cell. The details of the high pressure cell and the experimental procedure were described elsewhere.^6,17 The conductivity measurements were performed in the same cell. Unlike our previously study on ternary systems, the conductivity study in this work was focused on binary systems. It is more convenient to measure the conductivity of the binary systems in a fixed-volume cell, in which the molar concentration of solute can be kept as a constant during the course of a series of measurements under different pressures. Therefore, the movable piston of the high-pressure cell was replaced by a stainless steel plug fitted with a double acting hydraulic type seal to minimise leaking of the samples, especially at the higher temperatures (e.g. 363 K) required to perform the measurements for the CH₂F₂ systems. A schematic diagram of the apparatus for measuring the conductivity of binary systems is presented in Fig. 1.

Fig. 1 Schematic diagram of the experimental setup for measuring the conductivity of binary systems: BC, back cap; C, gas cylinder; CB, cell body; CM, conductivity meter; E, platinum electrodes; EH, electrode holder; G, glass tube; HS, hydraulic seal; MP, manual high-pressure pump; O, O-ring; P, pressure transducer, S, magnetic stirrer; T, thermocouple; TC, temperature controller; V1–V5, valves.

The experimental procedure for the conductivity measurements are described briefly as follows. A known weight of a solid (e.g. supporting electrolyte) was first placed at the bottom of the cell. Then the cell was fitted with the electrode holder and purged with a suitable fluid. After heating the cell to the desired temperature, the supercritical fluid was delivered gradually into the cellvia a hand operated high-pressure pump. At each pressure step, the mixture was left with stirring to equilibrate for at least 30 min before the conductivity was recorded. The conductivity data were collected by JENWAY 4510 conductivity meter that was calibrated using the standard conductivity solution of KCl.

The accuracy of the mixture composition is ∼±0.1% for HFCs and CH₃CN, and ∼±1% for supporting electrolytes and metal precursors. The estimated uncertainties are ±0.2 K for the temperature measurement, ±0.04 MPa for the pressure measurement, ∼±0.5% for the internal volume of the cell, and ∼±1% for the conductivity.

3. Results and discussion

3.1 Phase behaviour

CHF₃, CH₂F₂ and CH₂FCF₃ have a dipole moment greater than 1.6 D (see Table 1), which is comparable to low-polarity organic solvents, such as tetrahydrofuran (1.75 D) and ethyl acetate (1.78 D).¹⁸ Thus, in contrast to the CO₂ system reported in our previous study,⁶ addition of co-solvents may not be necessary to enhance the solubility of the supporting electrolytes in the HFC systems. Additionally, the critical temperatures (T_c) of CHF₃, CH₂F₂ and CH₂FCF₃ vary between 299 and 375 K, which also allows more choices of working temperatures because the experimental conditions often require that the bath temperature is not far away from the critical temperature of the media used for supercritical fluid electrodeposition. Our first aim of this study was to examine how different the three HFCs are in terms of the phase equilibrium of their mixtures with supporting electrolytes and metallic complexes.

Table 1 Physical properties of the hydrofluorocarbons used in this study

Physical property	CHF3	CH2F2	CH2FCF3
a See ref. 21. b See ref. 18.
T c/K^a	299.29	351.26	374.21
p c/bar^a	48.32	57.82	40.49
ρ c/kg m^−3 ^a	526.5	424.00	511.90
μ/D^b	1.65	1.98 ± 0.02	1.80 ± 0.22

---

3.1.1 Phase behaviour of supporting electrolytes in CHF₃, CH₂F₂ and CH₂FCF₃. The supporting electrolytes used in the phase behaviour study were [NBuⁿ₄][BF₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], and Na[B{3,5-C₆H₃(CF₃)₂}₄]. All of the three electrolytes, which are solid at ambient conditions, started to liquefy even at room temperature after they were exposed to gaseous HFCs (e.g. 1.3 MPa for CH₂F₂). This suggests that the solubility of the HFCs in the liquefied electrolytes is high, resulting in a large depression of melting point.^19,20 It also implies that there must be a three-phase equilibrium (solid–liquid–vapour) for the binary systems of electrolyte + HFC at a specified condition (composition, T and p). However, the data presented below are limited to the phase boundary between one-, and two-phase region since our attention has been focussed on the homogeneous, single-phase region, which provides the preferential conditions for conducting supercritical fluid electrodeposition through nano-channels.

Fig. 2 illustrates the p–T phase boundary of three binary systems of [NBuⁿ₄][BF₄] with CHF₃, CH₂F₂, and CH₂FCF₃, respectively. The mole fraction of all of the 13 mixtures and the phase boundary data can be found from Table S1 in the ESI.† The measurements were carried out in the temperature region covering both of the sub- and supercritical states, i.e. 293–328 K for the CHF₃ system, 295–363 K for the CH₂F₂ system, and 322–83 K for the CH₂FCF₃ system. Clearly, the boundaries of the three binary systems ([NBuⁿ₄][BF₄] + CHF₃, [NBuⁿ₄][BF₄] + CH₂F₂, and [NBuⁿ₄][BF₄] + CH₂FCF₃) lie in three separate regions from top left to bottom right in p, T space, which is related to the position of their critical points. The other main features of the phase diagrams may be described as follows: (i) for a mixture with a given composition, the phase transition pressure rises as the temperature rises; (ii) when the temperature is below the corresponding T_c of each HFC (indicated by the arrows in Fig. 2), the p–T boundaries are overlapped for the mixtures with different x_[NBu4][BF4]. In particular, the phase transition pressures of the CH₂F₂ and CH₂FCF₃ systems are only 0.3–0.7 MPa higher than the vapour pressures²¹ of the corresponding HFC at T < T_c,HFC. (iii) Above T_c of each HFC, the p–T boundary shifts to high pressures with increasing x_NBu4BF4, except for the mixture with the highest composition of [NBuⁿ₄][BF₄] in CH₂F₂.

Fig. 2 p–T phase diagram of the mixture of [NBuⁿ₄][BF₄] (1) + HFC (2). HFC = CHF₃: □, x₁ = 0.82 × 10⁻³; ○, x₁ = 1.59 × 10⁻³; △, x₁ = 2.44 × 10⁻³; ▽, x₁ = 4.83 × 10⁻³; ◊, x₁ = 6.90 × 10⁻³. HFC = CH₂F₂: +, x₁ = 1.67 × 10⁻³; ×, x₁ = 2.55 × 10⁻³; , x₁ = 3.29 × 10⁻³; ⊕, x₁ = 4.17 × 10⁻³; HFC = CHF₂CF₃: ■, x₁ = 1.33 × 10⁻³; ●, x₁ = 2.62 × 10⁻³; ▲, x₁ = 3.91 × 10⁻³; ▼, x₁ = 5.14 × 10⁻³. The arrows indicate the critical temperature of the HFCs.

Fig. 3 p–x phase diagram of the mixture of [NBuⁿ₄][BF₄] (1) + HFC (2) at the same reduced temperature (T/T_c,HFC = 1.01). The y-axis is shown as the reduced pressure, which is defined as p_r = p/p_c,HFC. The critical temperature (T_c,HFC) and the critical pressure (p_c,HFC) of each HFC are listed in Table 1. □, HFC = CHF₃, T = 302.3 K; ○, HFC = CH₂F₂, T = 354.8 K; and ◊, HFC = CHF₂CF₃, T = 378.0 K. Piecewise cubic Hermite polynomials were used to interpolate the data shown in Fig. 2 to the fixed temperature.

It is not unexpected that the p–T boundaries are overlapped below T_c for the mixtures with the different compositions of the electrolyte. Since [NBuⁿ₄][BF₄] is completely miscible with the liquid HFC in the composition region studied here, the phase equilibrium established below T_c may be regarded as equilibrium between a gas phase that consists of virtually no [NBuⁿ₄][BF₄] and a liquid phase with dissolved [NBuⁿ₄][BF₄]. Therefore, such a low mole fraction (x_[NBu4][BF4] < 6.9 × 10⁻³) of solute cannot significantly change the vapour pressure of the liquid CH₂F₂ or CH₂FCF₃. However, when the system temperature is increased above a certain temperature (often around T_c of the pure HFC), the nature of the phase transition switches from vapour–liquid‡ equilibrium to fluid–liquid§ equilibrium. Because the density of the fluid phase is more pressure dependent than the density of the liquid phase, the higher the pressure of the system, the larger the density, and hence the higher the solubility of [NBuⁿ₄][BF₄] in the HFC fluid phase.

We also note that the pressures required to form a homogeneous solution are much lower for the HFC system than for the CO₂ + CH₃CN system (x_CO2/x_CH3CN = 0.14)⁶ when [NBuⁿ₄][BF₄] is used as the supporting electrolyte. For example, the phase separation pressure is 22.2 MPa at 318.15 K for the CO₂ + CH₃CN mixture (x_[NBu4][BF4] = 2.70 × 10⁻³), while the phase separation pressure is only 11.7 MPa at the same temperature for the CHF₃ mixture with a much higher mole fraction of [NBuⁿ₄][BF₄] (x_[NBu4][BF4] = 4.83 × 10⁻³).

Although the p–T phase boundary data shown in Fig. 2 demonstrate that all of the three HFCs are good solvents to dissolve a sufficient amount of [NBuⁿ₄][BF₄] for carrying out electrodeposition, it is difficult to ascertain directly from the figure which HFC out of the three has the best solvation power since the vapour–liquid equilibria of the three HFCs are very different. To overcome this problem, we present our results in a modified p–x phase diagram. Rather than plotting the phase diagrams of the three HFC systems at a constant temperature, we have plotted the diagrams at the same reduced temperature, T_r = T/T_c,HFC, in which T is the system temperature, and T_c,HFC is the critical temperature of each HFC. Moreover, the pressure on the phase diagram is replaced by the reduced pressure (p_r) to eliminate the effect of different p_c for the three HFCs (p_r = p/p_c,HFC, where p is the pressure on the phase boundary, and p_c,HFC is the corresponding critical pressure of each HFC.).

Fig. 3 shows the modified p–x phase diagram for all three HFC systems at a constant T_r = 1.01, which corresponds to the commonly used working temperatures for applying HFCs to electrodeposition⁵ and other applications.²² The data shown in Fig. 3 have been obtained by interpolating the p–T boundary of the isopleths (Fig. 2) to a specified temperature for each HFC, i.e. 302.3 K for the CHF₃ system, 354.8 K for the CH₂F₂ system, and 378.0 K for the CH₂FCF₃ system. We can see from the figure that for the CHF₃ and CH₂FCF₃ systems the solubility (x_[NBu4][BF4]) increases as the increase of p_r, and that for the CH₂F₂ system p_r does not change dramatically as a function of x_[NBu4][BF4] within the composition range measured here. More importantly, it is clear that the CH₂F₂ system has the lowest p_r for dissolving the same amount of [NBuⁿ₄][BF₄]. This result may be explained by the fact that CH₂F₂ has the highest dipole moment among the three HFCs (see Table 1), and hence it is a more polar solvent than either CHF₃ or CH₂FCF₃.

Fig. 4 shows the p–T phase diagram of three binary mixtures of CH₂F₂ and the supporting electrolyte (i.e.[NBuⁿ₄][BF₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], and Na[B{3,5-C₆H₃(CF₃)₂}₄], respectively). The mole fractions of the test mixtures were kept virtually identical for the three electrolytes (x_electrolyte = ∼1.65 × 10⁻³). It can be seen from the figure that the p–T phase boundary for all of the three mixtures are almost superimposed within the temperature region between 295 to 353 K. The slight difference in the phase separation pressure can only be observed when T > 353 K. It is not surprising that two binary mixtures with [NBuⁿ₄][BF₄] and [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] show similar phase behaviour because CH₂F₂ is such a good solvent for the tetrabutylammonium salts. Thus the effect of introducing the BARF anion on the phase boundary is not as significant as that for the CO₂ systems (see Fig. 6 in our previous paper).⁵ However, the behaviour of Na[B{3,5-C₆H₃(CF₃)₂}₄] in CH₂F₂ is somewhat unexpected due to the fact that sodium salts often have low solubility in organic solvents. The BARF anion is a key component to enhance the solubility of Na[B{3,5-C₆H₃(CF₃)₂}₄] in CH₂F₂, and the large size difference between Na⁺ and [B{3,5-C₆H₃(CF₃)₂}₄]⁻ may play a role on improving solubility in CH₂F₂.

Fig. 4 p–T phase diagram of the mixtures of electrolyte (1) + CH₂F₂ (2): ■, [NBuⁿ₄][BF₄], x₁ = 1.67 × 10⁻³; ●, [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], x₁ = 1.65 × 10⁻³; ◆, Na[B{3,5-C₆H₃(CF₃)₂}₄], x₁ = 1.62 × 10⁻³; and of electrolyte (1) + CHF₂CF₃ (2): □, [NBuⁿ₄][BF₄], x₁ = 2.62 × 10⁻³; ○, [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], x₁ = 2.58 × 10⁻³; ◊, Na[B{3,5-C₆H₃(CF₃)₂}₄], x₁ = 2.59 × 10⁻³.

Similarly, the phase equilibrium measurements for the same three supporting electrolytes were made in CH₂FCF₃ at a fixed mole fraction of ∼2.60 × 10⁻³, see Fig. 4. Just as for the CH₂F₂ + electrolyte mixtures, the p–T phase boundaries of the three CH₂FCF₃ systems are overlapped below the temperature around T_c,CH2FCF3, while the phase separation pressure for the Na[B{3,5-C₆H₃(CF₃)₂}₄] mixture is lower than that of the other two mixtures when T > T_c,CH2FCF3. It is worth noting that Na[B{3,5-C₆H₃(CF₃)₂}₄] is an intermediate in the preparation of [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄]. The use of Na[B{3,5-C₆H₃(CF₃)₂}₄] may minimise impurities introduced from the metathesis step and make the supporting electrolyte less expensive.

3.1.2 Phase behaviour of the mixtures with copper precursors. Unlike supporting electrolytes, there are enormous numbers of metal precursors which can be potentially used for supercritical fluid electrodeposition, and thus systematic studies are required on different types of complexes (e.g. ionic, or neutral species), and on different metals and their oxidation states. Studies on the solubility of metal complexes in HFCs are very limited in the literature. It is reported by Shadrin et al.²³ that uranium dioxide can be extracted by CH₂FCF₃ with addition of nitric acid and tributyl phosphate, showing 99% of the uranium dioxide being converted to uranyl nitrate and dissolved in liquid CH₂FCF₃. Here we present some recently obtained experimental results on the two copper complexes: [Cu(CH₃CN)₄][BF₄] and [Cu(hfac)₂], representing an ionic compound and a neutral complex, respectively.

Our initial experiments on a binary mixture of [Cu(CH₃CN)₄][BF₄] and CH₂F₂ showed that [Cu(CH₃CN)₄][BF₄] was not stable enough for use in electrodeposition from pure CH₂F₂ in near- or supercritical conditions. In these experiments some blue solid (rather than white powder for pure [Cu(CH₃CN)₄][BF₄]) precipitated on the inner surface of the windows of the view cell when the system temperature was ∼333 K indicating decomposition of the copper complex. Consequently, a small amount of CH₃CN was added to the CH₂F₂ not only to increase the solubility of [Cu(CH₃CN)₄][BF₄], but also to stabilise it. Fig. 5 shows the p–T phase boundary of a ternary mixture of [Cu(CH₃CN)₄][BF₄] + CH₃CN + CH₂F₂, together, for comparison, with the phase boundaries of a ternary mixture of [NBuⁿ₄][BF₄] + CH₃CN + CH₂F₂, and a quaternary mixture of [Cu(CH₃CN)₄][BF₄] + [NBuⁿ₄][BF₄] + CH₃CN + CH₂F₂. The mole fraction of CH₃CN was kept at ∼0.034 for all of the three mixtures, and the mole fractions of [Cu(CH₃CN)₄][BF₄] (∼3.0 × 10⁻⁴) and [NBuⁿ₄][BF₄] (∼1.2 × 10⁻³) in the quaternary mixture were chosen to match the values of the two corresponding ternary systems, respectively.

Fig. 5 p–T phase diagram of the [Cu(CH₃CN)₄][BF₄] mixtures in CH₂F₂: ◊, [NBuⁿ₄][BF₄] (1) + CH₃CN (2) + CH₂F₂ (3), x₁ = 1.17 × 10⁻³, x₂ = 0.0312; □, [Cu(CH₃CN)₄][BF₄] (1) + CH₃CN (2) + CH₂F₂ (3), x₁ = 3.0 × 10⁻⁴, x₂ = 0.0375; ○, [Cu(CH₃CN)₄][BF₄] (1) + [NBuⁿ₄][BF₄] (2) + CH₃CN (3) + CH₂F₂ (4), x₁ = 3.0 × 10⁻⁴, x₂ = 1.22 × 10⁻³, x₃ = 0.0324.

One can see from Fig. 5 that below 353 K the p–T boundaries overlap each other for all of the three mixtures, and that in the temperature region 353–368 K, the difference in the phase-separation pressure is still small between the two ternary systems with either [Cu(CH₃CN)₄][BF₄] or [NBuⁿ₄][BF₄]. Surprisingly, the p–T boundary of the quaternary mixture (with both [Cu(CH₃CN)₄][BF₄] and [NBuⁿ₄][BF₄]) exhibits a steep rise between 353 and 363 K, indicating the change of the nature of the phase transition. In fact, we observed some white particles precipitated out of the solution at both 363 and 368 K when the system pressure was below the phase separation pressure. This is in contrast to the typical vapour–liquid phase transition observed for the same mixture at temperatures below 353 K. Therefore, it is clear that addition of [NBuⁿ₄][BF₄] to the [Cu(CH₃CN)₄][BF₄] system drastically shifts the phase boundary to high pressures at T > T_c,CH2F2. This phenomenon may be understood as follows. Our measured data on the ternary mixture with [Cu(CH₃CN)₄][BF₄] indicate that 3.75% of CH₃CN should be adequate to stabilise [Cu(CH₃CN)₄][BF₄] in scCH₂F₂ at temperatures between 333 and 368 K. However, when [NBuⁿ₄][BF₄] is also present in the mixture at ∼4 times as much as [Cu(CH₃CN)₄][BF₄], there may not be enough free CH₃CN molecules in CH₂F₂ available to stabilise the copper–acetonitrile complex because it is known that co-solvent molecules (e.g.CH₃CN) form clusters around the solute molecules and dissociated ions (e.g.[NBuⁿ₄][BF₄], [NBuⁿ₄]⁺, and [BF₄]⁻) in supercritical fluids with enhanced local concentration of co-solvents.^24–27 In order to compensate the depletion of the CH₃CN molecules in the bulk phase, the density, and hence the pressure of the system, needs to be increased considerably for the quaternary mixture to form a homogenous phase.

Similar measurements were made for the [Cu(hfac)₂] system in CHF₃. Since [Cu(hfac)₂] is very soluble and stable in CHF₃ at the required temperatures for electrodeposition (<330 K), no co-solvents were added to our test mixtures. Fig. 6 illustrates the p–T phase diagram of a ternary mixture of [Cu(hfac)₂] + [NBuⁿ₄][BF₄] + CHF₃, together with the phase diagrams of two corresponding binary mixtures: [NBuⁿ₄][BF₄] + CHF₃, and [Cu(hfac)₂] + CHF₃. Again, the mole fractions of [Cu(hfac)₂] (∼3.1 × 10⁻⁴) and [NBuⁿ₄][BF₄] (∼8.1 × 10⁻⁴) in the ternary mixture were chosen to match the values of the two corresponding binary systems, respectively.

Fig. 6 p–T phase diagram of the Cu(hfac)₂ mixtures in CH₂F₂: ◊, [NBuⁿ₄][BF₄] (1) + CHF₃ (2), x₁ = 8.23 × 10⁻⁴; □, Cu(hfac)₂ (1) + CHF₃ (2), x₁ = 3.0 × 10⁻⁴; ○, [Cu(hfac)₂](1) + [NBuⁿ₄][BF₄](2) + CHF₃(3), x₁ = 3.1 × 10⁻⁴, x₂ = 8.03 × 10⁻⁴.

In Fig. 6 the binary mixture of [Cu(hfac)₂] + CHF₃ exhibits the lowest phase-separation pressure between 298 and 328 K among the three mixtures. As can also be seen from the figure, the pressure of this mixture increases with increasing temperature, and then gradually levels off when the temperature approaches 328 K. This is in contrast to the behaviours of the mixtures of [NBuⁿ₄][BF₄] + CHF₃ and [Cu(hfac)₂] + [NBuⁿ₄][BF₄] + CHF₃, the boundaries of which are approximately two straight lines in parallel. Furthermore, when a small amount of [Cu(hfac)₂] (x_Cu(hfac)2 = 3.1 × 10⁻⁴) is added to the binary mixture of [NBuⁿ₄][BF₄] + CHF₃, the p–T boundary is shifted to higher pressure by only ∼0.5 MPa. This result is consistent with our previous report⁶ on the mixtures of [Cu(hfac)₂] and [NBuⁿ₄][BF₄] in a solution of CH₃CN + CO₂. Returning to the phase diagrams of the [Cu(CH₃CN)₄][BF₄] mixtures in Fig. 5, one can see the significant difference in the shift of the p–T boundary by adding different copper precursors to the solution of HFC + [NBuⁿ₄][BF₄]. Clearly, [Cu(hfac)₂] is more soluble and stable than [Cu(CH₃CN)₄][BF₄] in HFCs, and therefore it only increases the phase-separation pressure by a moderate amount (0.5 MPa in the example shown above) when added to the CHF₃ solution with the supporting electrolyte. This result may be understood by the fact that [Cu(hfac)₂] is a neutral complex with fluorinated, bidentate ligands, while [Cu(CH₃CN)₄][BF₄] is an ionic compound and the coordination bond between CH₃CN and Cu is rather weak.

3.2 Conductivity of BARF salts in HFCs

We have reported in our previous paper⁶ the conductivity of a variety of BARF electrolytes in a mixed fluid of CO₂ and CH₃CN, indicating that the three electrolytes [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], [NR_fMe₃][B{3,5-C₆H₃(CF₃)₂}₄], and [NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄] (R_f = CF₃(CF₂)₇(CH₂)₃) have the highest molar conductivities (22–26 S cm² mol⁻¹) at 328.15 K and 20 MPa. In this work, we extend our conductivity measurements to scCH₂F₂. The supporting electrolytes employed in this study were four BARF salts (i.e.[NBuⁿ₄][B(C₆F₅)₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], [NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄], and Na[B{3,5-C₆H₃(CF₃)₂}₄]), and [NBuⁿ₄][BF₄], which has been measured by Abbott and Eardley⁸ in both liquid and scCH₂F₂. Na[B{3,5-C₆H₃(CF₃)₂}₄] has been included as a candidate mainly because the phase behaviour described in the previous section of this paper demonstrates that it is surprisingly soluble in scCH₂F₂.

Fig. 7 illustrates the molar conductivity of the five electrolytes in scCH₂F₂ as a function of pressure at a constant temperature of 363 K and a fixed molar concentration of ∼9.0 mol m⁻³. One can see that, as the system pressure increases, the molar conductivity of all of the five electrolytes increases within the pressure range between 10–32 MPa. In particular, the [NBuⁿ₄][BF₄] mixture shows the most pronounced pressure effect on the molar conductivity, which is consistent with the trend reported in the literature⁸ at the same temperature, but for a mixture with a higher concentration (c_NBu4BF4 = 20 mol m⁻³). Moreover, the molar conductivity of [NBuⁿ₄][BF₄] obtained in our measurement is reasonably close to data reported in the literature. For example, Abbott and Eardley reported a molar conductivity of 0.013 S m² mol⁻¹ at 26.0 MPa and c_NBu4BF4 = 10.2 mol m⁻³ (estimated from Fig. 6 in ref. 8); our result is 0.0144 S m² mol⁻¹ at c_NBu4BF4 = 9.0 mol m⁻³, obtained by interpolating the data shown in Fig. 7 to 26.0 MPa.

Fig. 7 Molar conductivity (Λ) of the supporting electrolytes in CH₂F₂ at 363.15 K. The molar concentration of the electrolytes is ∼9.0 mol m⁻³. ■, [NBuⁿ₄][BF₄]; ●, [NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄]; ▲, [NBuⁿ₄][B(C₆F₅)₄]; ▼, [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄]; ◆, Na[B{3,5-C₆H₃(CF₃)₂}₄].

It can also be seen from Fig. 7 that when the system pressure is below 20 MPa, the molar conductivity of all of the five electrolytes increases considerably with increasing pressure, reflecting the fact that the dielectric constant of CH₂F₂ increases from 7.39 at 10.0 MPa to 8.99 at 20.1 MPa at a constant temperature of 363 K.²⁸ However, when the system pressure is above 20 MPa, the molar conductivity is almost unchanged with pressure for the three electrolytes having [B{3,5-C₆H₃(CF₃)₂}₄]⁻ as the anion. Although at 363 K the dielectric constant of CH₂F₂ increases from 8.99 at 20.1 MPa to 9.72 at 28.1 MPa,²⁸ such a small change in the dielectric constant may have less significant effect on association/dissociation of the [B{3,5-C₆H₃(CF₃)₂}₄]⁻ salts than on [NBuⁿ₄][BF₄], and its effect on the molar conductivity may be cancelled out by the change in the viscosity of the fluid mixtures.⁹ Nevertheless, the pressure effect on the molar conductivity of the five electrolytes largely depends on the anions. Comparing the results for [NBuⁿ₄][BF₄], [NBuⁿ₄][B(C₆F₅)₄] and [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] in Fig. 7, as the anion gets bigger the molar conductivity of the electrolyte increases and the molar conductivity shows less pressure dependence within the pressure range measured in this study (10–32 MPa).

We have interpolated the molar conductivity data shown in Fig. 7 to a fixed pressure of 20.0 MPa by using piecewise cubic Hermite polynomials, and the results are listed in Table 2. In Table 2, the molar conductivity of all of the five electrolytes lies between 95 and 255 S cm² mol⁻¹ for the mixtures with c = ∼9.0 mol m⁻³, which is at least 3.6 times as high as that of the same electrolyte in the mixed fluid of CO₂ + CH₃CN (see Table 2 in ref. 6). Thus, the CH₂F₂ system provides a higher conductivity than that of the CO₂ + CH₃CN system (x_CH3CN/x_CO2 = 0.12).

Table 2 Molar conductivity of the supporting electrolytes in CH₂F₂ at 363 K and 20 MPa^a

Electrolyte	Λ/S cm^2 mol^−1
a The concentration of all of the electrolytes is ∼9.0 mol m^−3. Piecewise cubic Hermite polynomials were used to interpolate the data shown in Fig. 7 to the fixed pressure (20 MPa). b Rf = CF3(CF2)7(CH2)3.
[NBu^n4][BF4]	124
[NRfBu^n3][B{3,5-C6H3(CF3)2}4] ^b	95
[NBu^n4][B(C6F5)4]	170
[NBu^n4][B{3,5-C6H3(CF3)2}4]	197
Na[B{3,5-C6H3(CF3)2}4]	255

---

Fig. 7 also shows that within the pressure range between 10–32 MPa, the general trend of the molar conductivity for the four BARF electrolytes is as follows:

[NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄] < [NBuⁿ₄][B(C₆F₅)₄] < [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] < Na[B{3,5-C₆H₃(CF₃)₂}₄].

The radii of the cations are in the order Na⁺ < [NBuⁿ₄]⁺ < [NR_fBuⁿ₃]⁺. Thus, for the three electrolytes with the same anion ([B{3,5-C₆H₃(CF₃)₂}₄]⁻) the molar conductivities increase as the radius of the cation decreases. We also note that Na[B{3,5-C₆H₃(CF₃)₂}₄] has a much higher molar conductivity than [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], i.e. 255 S cm² mol⁻¹ for Na[B{3,5-C₆H₃(CF₃)₂}₄], and 197 S cm² mol⁻¹ for [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄]), respectively, at 363 K and 20 MPa (see Table 2). This may also be related to the greater dissociation of Na[B{3,5-C₆H₃(CF₃)₂}₄] in CH₂F₂.

Since Na[B{3,5-C₆H₃(CF₃)₂}₄] and [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] have the highest conductivity among the four BARF salts, these two compounds were chosen for a detailed conductivity study at various pressures and concentrations. Fig. 8 and 9 present results, respectively, for [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄] at 363 K and the molar concentration up to 60 mol m⁻³. All of the conductivity data shown in the figures were recorded under the conditions corresponding to a homogeneous single phase present in the high-pressure cell. Two main features of the conductivity diagrams can be summarised as follows: (i) the higher the molar concentration of the supporting electrolyte, the larger the conductivity the mixture has at a given pressure for both of the [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄] systems; (ii) the system pressure has a moderate effect on the conductivity for all of the mixtures with a molar concentration ranging from 0.8 to 60 mol m⁻³ (See also the previous discussion on the pressure effect on molar conductivity using the mixtures with c = 9.0 mol m⁻³ as examples (Fig. 7)). We stress here that the considerable increase of conductivity with pressure, which has been observed between 10–20 MPa from a variety of mixtures cannot be seen in Fig. 8 for the mixture with 60 mol m⁻³ of [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], partially because the conductivity must be measured in a homogeneous, single-phase solution, resulting in the unavailability of the conductivity data below 18 MPa.

Fig. 8 Conductivity (κ) of [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] in scCH₂F₂ at 363.15 K. The molar concentrations of [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] are: , 0.78 mol m⁻³; ×, 1.04 mol m⁻³; +, 1.35 mol m⁻³; ◊, 2.54 mol m⁻³; ▽, 4.93 mol m⁻³; △, 8.98 mol m⁻³; ○, 30.6 mol m⁻³; □, 60.2 mol m⁻³.

Fig. 9 Conductivity (κ) of Na[B{3,5-C₆H₃(CF₃)₂}₄] in scCH₂F₂ at 363.15 K. The molar concentrations of Na[B{3,5-C₆H₃(CF₃)₂}₄] are: ◊, 1.05 mol m⁻³; ▽, 2.59 mol m⁻³; △, 9.04 mol m⁻³; ○, 15.0 mol m⁻³; □, 20.2 mol m⁻³.

To understand the different behaviour in conductivity between [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄], the data shown in Fig. 8 and 9 were interpolated to a fixed pressure of 20 MPa using piecewise cubic Hermite polynomials, and the results are plotted against the square root of the concentration in Fig. 10, together with the data for the [NBuⁿ₄][BF₄] system reported by Abbott and Eardley⁸ at the same temperature (363 K) but a slightly higher pressure of 22 MPa. It can be seen from the figure that the molar conductivity decreases with increasing √c for both [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄], and no minimum can be seen in the concentration range of 0.7–60 mol m⁻³ for [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], or 1–20 mol m⁻³ for Na[B{3,5-C₆H₃(CF₃)₂}₄]. In addition, the molar conductivity of [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] starts to level off when c approaches 30 mol m⁻³, indicating the significant contribution of triple ions to the conductivity at the high concentrations, while the Λ−√c curve for the Na[B{3,5-C₆H₃(CF₃)₂}₄] system shows a linear relationship within the concentration region given above. The limiting molar conductivity of Na[B{3,5-C₆H₃(CF₃)₂}₄] cannot be calculated from the curve because molar conductivity data are not available at low concentrations (<1 mol m⁻³) from this study.

Fig. 10 Molar conductivity (Λ) of three supporting electrolytes in scCH₂F₂ at 363.15 K: ■, [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], 20 MPa; ●, Na[B{3,5-C₆H₃(CF₃)₂}₄], 20 MPa; and ◆, [NBuⁿ₄][BF₄], 22 MPa, estimated from Fig. 6 of ref. 8. The molar conductivity of the two BARF salts were obtained by interpolating the data shown in Fig. 8 and 9 to the fixed pressure (20 MPa) using piecewise cubic Hermite polynomials. The lines are merely to guide the eye. The error bars represent the estimated experimental error.

Compared with the [NBuⁿ₄][BF₄] system, we can see that the molar conductivities of the [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] system are higher at all concentrations. Furthermore, the molar conductivity of [NBuⁿ₄][BF₄] decreases rapidly with the increase of √c when c < 8 mol m⁻³ and then it starts to level off at ∼16 mol m⁻³, that is lower than the corresponding concentration (30 mol m⁻³) for the [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] system. Thus [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] can produce more free ions in scCH₂F₂ than [NBuⁿ₄][BF₄], making it a better choice of a supporting electrolyte than [NBuⁿ₄][BF₄] in terms of conductivity. However, this effect is less significant for the scCH₂F₂ system than for the CO₂ + CH₃CN system when replacing [NBuⁿ₄][BF₄] with the BARF salts: the molar conductivity increases by 60% in scCH₂F₂ when replacing [NBuⁿ₄][BF₄] with [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] (Table 2), whereas in CO₂ + CH₃CN, the conductivity increases by at least 900%⁶ when using [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄]. Thus [NBuⁿ₄][BF₄] is a satisfactory electrolyte for the use in scCH₂F₂, particularly for those applications that do not require passage of large currents.

Although the difference in the conductivity of the three electrolytes is mainly ascribed to the association/dissociation ability of the electrolytes in scCH₂F₂, the viscosity has an effect on the mobility of the ions and hence on the conductivity. The viscosity of CH₂F₂ can be increased by a factor of 8–32 at 363 K when adding 5–30 mol m⁻³[NBuⁿ₄][BF₄].⁹ It is expected that both [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄] will increase the viscosity of the CH₂F₂ even more because [B{3,5-C₆H₃(CF₃)₂}₄]⁻ is a much larger anion than [BF₄]⁻. Nevertheless, as the concentration increases, the viscosity increases for all three electrolytes, resulting in the decrease in the molar conductivity, which is consistent with the general trend shown in Fig. 10.

4. Conclusions

In this paper we have presented results of a study of the phase behaviour of CHF₃, CH₂F₂, and CH₂FCF₃ containing tetrabutylammonium tetrafluoroborate ([NBuⁿ₄][BF₄]), [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄] and Na[B{3,5-C₆H₃(CF₃)₂}₄]. For all three HFCs we find that when the temperature is below the corresponding T_c the p–T boundaries are overlapped for the mixtures with different composition of [NBuⁿ₄][BF₄] but at higher temperature the solubility of [NBuⁿ₄][BF₄] in the HFC fluid phase increases with increasing pressure. Compared to the CO₂ + CH₃CN system (x_CO2/x_CH3CN = 0.14) studied previuously,⁶ when [NBuⁿ₄][BF₄] is used as the supporting electrolyte, the pressures required to form a homogeneous solution are much lower for the HFCs.

Although all three HFCs are good solvents to dissolve [NBuⁿ₄][BF₄], our results show that the CH₂F₂ system has the lowest p_r for dissolving the same amount of [NBuⁿ₄][BF₄]. This may be explained by the fact that CH₂F₂ has the highest dipole moment among the three HFCs. The behaviour of the Na[B{3,5-C₆H₃(CF₃)₂}₄] in CH₂F₂ is somewhat unexpected since sodium salts often have low solubility in organic solvents. Here the BARF anion appears to be responsible for the enhanced solubility of the Na[B{3,5-C₆H₃(CF₃)₂}₄] in CH₂F₂, possibly due to the large mismatch in the size of the two ions.

Studies of the phase behaviour of CH₂F₂ containing [NBuⁿ₄][BF₄] and [Cu(CH₃CN)₄][BF₄] showed that the copper complex was unstable in the absence of CH₃CN. Our results show that the addition of [NBuⁿ₄][BF₄] to the [Cu(CH₃CN)₄][BF₄] system drastically shifts the phase boundary to high pressures at T > T_c,CH2F2. We believe that this is caused by selective solvation of the [NBuⁿ₄]⁺ and [BF₄]⁻ ions by CH₃CN, depleting the CH₃CN molecules in the bulk phase, so that the density, and hence the pressure of the system, needs to be increased considerably for the quaternary mixture to form a homogenous phase.

For CHF₃ and [Cu(hfac)₂] we find that [Cu(hfac)₂] is more soluble and more stable than [Cu(CH₃CN)₄][BF₄] in HFCs and that it only increases the phase-separation pressure by a moderate amount when added to the CHF₃ solution with [NBuⁿ₄][BF₄] as the supporting electrolyte. An example of using [Cu(hfac)₂] as a precursor for the electrodeposition of copper from scCHF₃ has been reported in our recent work.²⁹

Since CH₂F₂ was found to have the lowest p_r for dissolving the same amount of [NBuⁿ₄][BF₄] studies of the conductivity for four different BARF salts, [NBuⁿ₄][B(C₆F₅)₄], [NBuⁿ₄][B{3,5-C₆H₃(CF₃)₂}₄], [NR_fBuⁿ₃][B{3,5-C₆H₃(CF₃)₂}₄], and Na[B{3,5-C₆H₃(CF₃)₂}₄], were carried out. Our results show that the BARF salts are more conducting than [NBuⁿ₄][BF₄] under the same conditions although the effect is much less significant than for our previous work in supercritical CO₂ + CH₃CN (a factor of 1.6 increase in molar conductivity as opposed to 9).

The main purpose of this work was to characterise supercritical fluids that can be used as electrolytes for electrodeposition. Our results show that scCH₂F₂ is a suitable system for electrodeposition using either [NBuⁿ₄][BF₄] or the corresponding BARF salts as electrolytes and identify the condition under which a single supercritical phase is formed. The results also confirm that the visual method used in this study can provide crucial information on the stability of metal precursors under supercritical conditions, and is a powerful method for the acquisition of information on phase equilibria.

Acknowledgements

This work was supported by RCUK via a Basic Technology Grant. We are grateful to Prof. M. Poliakoff and Dr D. C. Smith for useful advice and Messrs. M. Guyler, R. Wilson, P. Fields, D. Litchfield, and J. Warren for technical support. We are grateful to Dr R. E. Low (INEOS Fluor Ltd, UK) for supplying R32. Royal Society Wolfson Merit Award to MWG is gratefully acknowledged. JK wish to thank Drs S. E. Davies, F. Qiu, D. Walsh, and J. Yang for helpful discussion.

References

1. I. Souvignet and S. V. Olesik, Anal. Chem., 1998, 70, 2783–2788.
2. A. P. Abbott, C. A. Eardley, J. C. Harper and E. G. Hope, J. Electroanal. Chem., 1998, 457, 1–4.
3. T. Clifford, Fundamentals of Supercritical Fluids, Oxford University Press, Oxford, 1999.
4. A. P. Abbott and J. C. Harper, J. Chem. Soc., Faraday Trans., 1996, 92, 3895–3898.
5. J. Ke, W. Su, S. M. Howdle, M. W. George, D. Cook, M. Perdjon-Abel, P. N. Bartlett, W. Zhang, F. Cheng, W. Levason, G. Reid, J. Hyde, J. Wilson, D. Smith, K. Mallik and P. Sazio, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 14768–14772.
6. P. N. Bartlett, D. C. Cook, M. W. George, J. Ke, W. Levason, G. Reid, W. Su and W. Zhang, Phys. Chem. Chem. Phys., 2010, 12, 492–501.
7. A. P. Abbott and N. E. Durling, Phys. Chem. Chem. Phys., 2001, 3, 579–582.
8. A. P. Abbott and C. A. Eardley, J. Phys. Chem. B, 2000, 104, 9351–9355.
9. A. P. Abbott, E. G. Hope and D. J. Palmer, Anal. Chem., 2005, 77, 6702–6708.
10. D. L. Goldfarb and H. R. Corti, Electrochem. Commun., 2000, 2, 663–670.
11. D. L. Goldfarb and H. R. Corti, J. Phys. Chem. B, 2004, 108, 3368–3375.
12. D. L. Goldfarb and H. R. Corti, J. Phys. Chem. B, 2004, 108, 3358–3367.
13. S. A. Olsen and D. E. Tallman, Anal. Chem., 1994, 66, 503–509.
14. S. A. Olsen and D. E. Tallman, Anal. Chem., 1996, 68, 2054–2061.
15. M. Atobe, S. Iizuka, T. Fuchigami and H. Yamamoto, Chem. Lett., 2007, 36, 1448–1449.
16. M. Atobe, H. Ohsuka and T. Fuchigami, Chem. Lett., 2004, 33, 618–619.
17. P. Licence, M. P. Dellar, R. G. M. Wilson, P. A. Fields, D. Litchfield, H. M. Woods, M. Poliakoff and S. M. Howdle, Rev. Sci. Instrum., 2004, 75, 3233.
18. R. L. David, CRC Handbook of Chemistry and Physics, CRC Press/Taylor and Francis, Boca Raton, FL, 89th edn, 2009.
19. K. Fukne-Kokot, A. Konig, Z. Knez and M. Skerget, Fluid Phase Equilib., 2000, 173, 297–310.
20. A. M. Scurto, E. Newton, R. R. Weikel, L. Draucker, J. Hallett, C. L. Liotta, W. Leitner and C. A. Eckert, Ind. Eng. Chem. Res., 2008, 47, 493–501.
21. E. W. Lemmon, M. L. Huber and M. O. McLinden, NIST Standard Reference Database 23: Reference Fluid Thermodynamic and Transport Properties—REFPROP, Version 8.0, National Institute of Standards and Technology, Standard Reference Data Program, Gaithersburg, 2007.
22. D. R. Brannegan, M. Ashraf-Khorassani and L. T. Taylor, Chromatographia, 2001, 54, 399–401.
23. A. Shadrin, V. Kamachev, A. Murzin and D. Shafikov, J. Supercrit. Fluids, 2007, 42, 347–350.
24. S. W. Kim and K. P. Johnston, AIChE J., 1987, 33, 1603–1611.
25. Y. P. Sun, G. Bennett, K. P. Johnston and M. A. Fox, J. Phys. Chem., 1992, 96, 10001–10007.
26. I. Skarmoutsos, D. Dellis and J. Samios, J. Chem. Phys., 2007, 126, 10.
27. J. W. Ford, M. E. Janakat, J. Lu, C. L. Liotta and C. A. Eckert, J. Org. Chem., 2008, 73, 3364–3368.
28. A. P. Abbott, C. A. Eardley and R. Tooth, J. Chem. Eng. Data, 1999, 44, 112–115.
29. D. Cook, P. N. Bartlett, W. Zhang, W. Levason, G. Reid, J. Ke, W. Su, M. W. George, J. Wilson, D. Smith, K. Mallik, E. Barrett and P. Sazio, Phys. Chem. Chem. Phys., 2010, 12, 11744–11752.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental phase equilibrium data of the binary systems of electrolytes and hydrofluorocarbons. See DOI: 10.1039/c0cp01202e

‡ The liquid phase mainly consists of CH₂F₂ or CHF₂CF₃.

§ The liquid phase consists of a large amount of [NBuⁿ₄][BF₄].

---