Binary and ternary (liquid + liquid) equilibrium for {methylcyclohexane (1) + toluene (2) + 1-hexyl-3-methylimidazolium tetracyanoborate (3)/1-butyl-3-methylimidazolium tetracyanoborate (3)}

This paper focuses on the study of the solubility behaviour of 1-hexyl-3-methylimidazolium tetracyanoborate [HMIM][TCB] and 1-butyl-3-methylimidazolium tetracyanoborate [BMIM][TCB] in combination with methylcyclohexane and toluene as representatives for non-aromatic and aromatic components. Binary and ternary (liquid + liquid) equilibrium data were collected at three different temperatures and at atmospheric pressure (0.1 MPa). The experimental data were well-correlated with the NRTL and UNIQUAC thermodynamic models; however, the UNIQUAC model gave better predictions than the NRTL, with a root mean square error below 0.97%. The non-aromatic/aromatic selectivities of the ionic liquids make them suitable solvents to be used in extractive distillation processes.     

Excess molar volumes of (an alkane + 1-chloroalkane) atT = 298.15 K

The densities of the following: (pentane  +  1-chloropropane, or 1-chlorobutane, or 1-chloropentane, or 1-chlorohexane), (hexane  +  1-chloropropane, or 1-chlorobutane, or 1-chloropentane, or 1-chlorohexane), (heptane  +  1-chloropropane, or 1-chlorobutane, or 1-chloropentane, or 1-chlorohexane), (octane  +  1-chloropropane, or 1-chlorobutane, or 1-chloropentane, or 1-chlorohexane), were measured at T =  298.15 K by means of a vibrating-tube densimeter. The excess molar volumes V_m^E, calculated from the density data, are negative for (pentane  +  1-chloropentane, or 1-chlorohexane) and (hexane  +  1-chlorohexane) over the entire range of composition. (Pentane  +  1-chlorobutane), (hexane  +  1-chloropentane) and (heptane  +  1-chlorohexane) exhibit an S-shapedV_m^E dependence. For all the other systems,V_m^E is positive. The V_m^Eresults were correlated using the fourth-order Redlich–Kister equation, with the maximum likelihood principle being applied for determining the adjustable parameters.     

Separation of toluene from alkanes using 1-ethyl-3-methylpyridinium ethylsulfate ionic liquid at T = 298.15 K and atmospheric pressure

In this paper, the separation of toluene from aliphatic hydrocarbons (heptane, or octane, or nonane) was analyzed by solvent extraction with 1-ethyl-3-methylpyridinium ethylsulfate ionic liquid, [EMpy][ESO₄]. Liquid?liquid equilibrium (LLE) data for the ternary systems {heptane (1) + toluene (2) + [EMpy][ESO₄] (3)}, {octane (1) + toluene (2) + [EMpy][ESO₄] (3)}, and {nonane (1) + toluene (2) + [EMpy][ESO₄] (3)} were obtained by measurements at T = 298.15 K and atmospheric pressure. The selectivity, % removal of aromatic, and solute distribution ratio, obtained from experimental equilibrium results, were used to determine the ability of [EMpy][ESO₄] as a solvent. The degree of consistency of the experimental LLE values was ascertained using the Othmer–Tobias and Hand equations. The experimental results for the ternary systems were correlated with the NRTL model. Finally, the results obtained were compared with other ionic liquids and other solvents.     

The influence of 1-butyl-3-methyl-imidazolium bromide on the partitioning of l-tyrosine within the {polyethylene glycol 600 + potassium citrate} aqueous biphasic system at T = 298.15 K

Partitioning in aqueous biphasic systems (ABS) is widely recognized today as a rapid, gentle, and highly efficient technique for the separation of soluble as well as particulate biomaterials. This technique has gained increasing attention as the separation method of choice in biotechnology. In recent years, a new approach has been proposed based on the use of ionic liquids (ILs) as adjuvants for the separation and purification of bio-molecules using polymer-based ABS. In this regard, the influence of IL 1-butyl-3-methylimidazolium bromide ([C₄mim]Br) on the phase behavior and extraction capability of {PEG 600 + tri-potassium citrate (K₃C₆H₅O₇)} ABS for l-tyrosine (Tyr) is investigated here. For this purpose, phase diagrams and the liquid–liquid equilibrium (LLE) data for the {PEG 600 + K₃C₆H₅O₇} ABS with the addition of small quantities of IL were determined at T = 298.15 K. It was found that, for the studied polymer-based ABS, the addition of 5 wt% of [C₄mim]Br to ABS caused the expansion of two-phase area in the salt-rich region; while, for the PEG-rich region no change was observed. The partition coefficients of l-tyrosine (K_Tyr) within the studied system were determined at T = 298.15 K. The results obtained indicate that the addition of small quantities of [C₄mim]Br to the {PEG 600 + K₃C₆H₅O₇} ABS could enhance the extraction efficiency for l-tyrosine. In addition, the experimental data are correlated using the NRTL model. The comparisons between the correlation and the experimental data reveal a good agreement.     

(Liquid + liquid) equilibrium of (imidazolium ionic liquids + organic salts) aqueous two-phase systems at T = 298.15 K and the influence of salts and ionic liquids on the phase separation

(Liquid + liquid) equilibrium for the {1-ethyl-3-methylimidazolium tetrafluoroborate ([C₂mim]BF₄)/1-propyl-3-methylimidazolium tetrafluoroborate ([C₃mim]BF₄) + organic salt + H₂O} aqueous two-phase systems (ATPSs) have been experimentally ascertained at T = 298.15 K. Three empirical equations were used to correlate the binodal data. On the basis of the empirical equation of the binodal curve with the highest accuracy and lever rule, the (liquid + liquid) equilibrium data were calculated by MATLAB. The reliability of the tie line compositions was proved by the empirical correlation equations given by the Othmer–Tobias and Bancroft equations. The effective excluded volume (EEV) values obtained from the binodal model for these systems were determined. The EEV and the binodal curves plotted in molality both indicate that the salting-out abilities of the four salts follow the order: Na₃C₆H₅O₇ > (NH₄)₃C₆H₅O₇ > Na₂C₄H₄O₄ ≈ Na₂C₄H₄O₆, while the phase-separation abilities of the investigated ILs are in the order of [C₃mim]BF₄ > [C₂mim]BF₄. In the systems investigated, the effect of salts on the phase-forming capability was also evaluated in the shape of the salting-out coefficient obtained from fitting the tie-line data to a Setschenow-type equation. The phase-forming ability increases with the increase of salting-out coefficient.     

Energetic characterisation of the system (water + 1-propoxypropan-2-ol) at T = 298.15 K

Given the importance that enthalpic and entropic contributions have in the interplay between thermodynamics and self-assembly of aqueous amphiphile systems, the energetic characterisation of the system {water + 1-propoxypropan-2-ol (1-pp-2-ol)} at T = 298.15 K was made by directly measuring excess partial molar enthalpies of 1-pp-2-ol and water, over the entire composition range, at T = 298.15 K and atmospheric pressure. Derivatives of the partial molar properties with respect to the composition are used to improve the understanding of molecular interactions in the water-rich region. The present results were compared with those for the well-studied system {water + 2-butoxyethanol (nC₄E₁)}, the two amphiphiles being structural isomers.     

(Liquid + liquid) equilibrium of aqueous biphasic systems composed of 1-benzyl or 1-hexyl-3-methylimidazolium chloride ionic liquids and inorganic salts

(Liquid + liquid) equilibria for {1-benzyl-3-methylimidazolium chloride ([BzMIM]Cl) or 1-hexyl-3-methylimidazolium chloride ([HMIM]Cl) + inorganic salts (potassium phosphate K₃PO₄, potassium carbonate K₂CO₃, or dipotassium hydrogen phosphate K₂HPO₄) + H₂O} aqueous biphasic systems (ABSs) are presented at T = 298.15 K. An empirical equation was used to correlate the binodal data. The experimental tie lines were appropriately correlated by the Othmer–Tobias and Brancroft empirical equations. The influence of the selected inorganic salts in the phase segregation was investigated by means the calculated effective excluded volume (EEV) and Setschenow-type equation. The salting-out ability of salts was also evaluated in terms of the Gibbs energy of hydration of salt (ΔG_hyd) and assessed with EEV values.     

The P, V, T,x properties of binary aqueous chloride solutions up to T = 573 K and 100 MPa

A highly accurate P, V, T,x model is developed for aqueous chloride solutions of the binary systems, viz. (LiCl + H₂O), (NaCl + H₂O), (KCl + H₂O), (MgCl₂ + H₂O), (CaCl₂ + H₂O), (SrCl₂ + H₂O), and (BaCl₂ + H₂O). The applied ranges of temperature, pressure, and concentrations for the systems (LiCl + H₂O), (NaCl + H₂O), (KCl + H₂O), (MgCl₂ + H₂O), (CaCl₂ + H₂O), (SrCl₂ + H₂O), and (BaCl₂ + H₂O) are (273 K to 564 K, 0.1 MPa to 40 MPa, and 0 to 10 molal), (273 K to 573 K, 0.1 MPa to 100 MPa, and 0 to 6.0 molal), (273 K to 543 K, 0.1 MPa to 50 MPa, and 0 to 4.5 molal), (273 K to 543 K, 0.1 MPa to 40 MPa, and 0 to 3.0 molal), (273 K to 523 K, 0.1 MPa to 60 MPa, and 0 to 6.0 molal), (298 K to 473 K, 0.1 MPa to 2 MPa, and 0 to 2.0 molal) and (273 K to 473 K, 0.1 MPa to 20 MPa, and 0 to 1.6 molal), respectively. Comparison of the model with thousands of experimental data points concludes that the average deviation over the above T, P, m range is 0.020% to 0.066% in density (or volume) for these systems, which indicates high accuracy. From this model, various volumetric properties, such as the apparent molar volume at infinite dilution and isochores of fluid inclusions, can be calculated, thus having a wide range of geological applications, such as reservoir fluid flow simulation and fluid-inclusion study. A computer code is developed for this model and can be downloaded from the website: www.geochem-model.org/programs.htm and online calculations is made available on: www.geochem-model.org/models.htm     

(Liquid + liquid) equilibria in {water + acrylic acid + (1-butanol,or 2-butanol,or 1-pentanol)} systems at T = 293.2 K,T = 303.2 K,and T = 313.2 K and atmospheric pressure

(Liquid + liquid) equilibrium (LLE) data for {water + acrylic acid + (1-butanol, or 2-butanol, or 1-pentanol)} at T = 293.2 K, T = 303.2 K, and T = 313.2 K and atmospheric pressure (≈95 kPa) were determined by Karl Fischer titration and densimetry. All systems present type I binodal curves. The size of immiscibility region changes little with an increase in temperature, but increases according to the solvent, following the order: 2-butanol < 1-butanol < 1-pentanol. Values of solute distribution and solvent selectivities show that 1-pentanol is a better solvent than 1-butanol or 2-butanol for acrylic acid removal from water solutions. Quality of data was ascertain by Hand and Othmer-Tobias equations, giving R² > 0.916, mass balance and accordance between tie lines and cloud points. The NRTL model was used to correlate experimental data, by estimating new energy parameters, with root mean square deviations below 0.0053 for all systems.     

Volumetric properties of (an organic compound + di-n-butyl ether) atT = 298.15 K

Excess molar volumes V_m^Eof {di- n -butyl ether (DBE)  +  a monofunctional organic compound} have been determined atT =  298.15 K over the whole composition range by means of a vibrating-tube densimeter. TheV_m^E values were either positive (propylamine, or butylamine, or acetone, or tetrahydrofuran  +  DBE) or negative (methanol, or butanol, or diethyl ether, or cyclopentanone, or acetonitrile  +  DBE). Markedly asymmetric V_m^Ecurves were displayed by (DBE  +  methanol) and (DBE  +  acetonitrile). Partial molar volumes __ V_m^oat infinite dilution in DBE, both from this work and the literature, were analysed in terms of an additivity scheme, and the group contributions thus obtained were discussed and compared with analogous results in water. DBE revealed a greater capability of distinguishing between polar and non-polar solutes, as well as in discriminating differently shaped molecules (unbranched, branched, cyclic). The limiting slopes of apparent excess molar volumes are evaluated and briefly discussed in terms of solute–solute and solute–solvent interactions.     

Diffusion coefficients of nickel chloride in aqueous solutions of lactose at T = 298.15 K and T = 310.15 K

Binary mutual diffusion coefficients (interdiffusion coefficients) of nickel chloride in water at T = 298.15 K and T = 310.15 K, and at concentrations between (0.000 and 0.100) mol · dm^?3, using a Taylor dispersion method have been measured. These data are discussed on the basis of the Onsager–Fuoss and Pikal models. The equivalent conductance at infinitesimal concentration of the nickel ion in these solutions at T = 310.15 K has been estimated using these results. Through the same technique, ternary mutual diffusion coefficients (D₁₁, D₂₂, D₁₂, and D₂₁) for aqueous solutions containing NiCl₂ and lactose, at T = 298.15 K and T = 310.15 K, and at different carrier concentrations were also measured. These data permit us to have a better understanding of the structure of these systems and the thermodynamic behaviour of NiCl₂ in different media.     

(Liquid + liquid) equilibrium of (water + 2-propanol + 1-butanol + salt) systems at T = 313.15 K and T = 353.15 K: Experimental data and correlation

(Liquid + liquid) equilibrium data for the quaternary systems (water + 2-propanol + 1-butanol + potassium bromide) and (water + 2-propanol + 1-butanol + magnesium chloride) were measured at T = 313.15 K and T = 353.15 K. The overall salt concentrations were 5 and 10 mass percent. Ternary (liquid + liquid) equilibrium data for the salt-free system (water + 2-propanol + 1-butanol) were also determined and found to be in good agreement with data from the literature. The NRTL model for the activity coefficient was used to correlate the data. New interaction parameters were estimated, using the Simplex minimization method and a concentration-based objective function. The results are very satisfactory, with root mean square deviations between experimental and calculated compositions of both phases being less than 0.5%.     

Synthesis,structure and magnetic properties of 2-D and 3-D [cation]{M[Au(CN)2]3} (M = Ni, Co) coordination polymers

In order to study the templating effect of the cation and the resulting impact on the magnetic properties, reactions of M(II) salts with [cation][Au(CN)₂] were conducted, yielding a series of coordination polymers of the form [cation]{M[Au(CN)₂]₃} (cation = ⁿBu₄N⁺, PPN⁺ (bis(triphenylphosphoranylidene)ammonium); M = Ni(II) and Co(II)). The structures of ⁿBu₄N{M[Au(CN)₂]₃} and PPN{M[Au(CN)₂]₃} (M = Ni and Co) contain two distinct 3-D anionic frameworks of {M[Au(CN)₂]₃}⁻, hence the framework was sensitive to the cation, but not to the identity of the metal center. In ⁿBu₄N{M[Au(CN)₂]₃}, the metal centers are connected by [Au(CN)₂] units to form six 2-D (4, 4) rectangular grids that are fused through the M centers to yield a complex three-dimensional framework which accommodates the ⁿBu₄N⁺ cations. In PPN{M[Au(CN)₂]₃}, the framework adopts a simpler non-interpenetrated Prussian-blue-type pseudo-cubic array, with the PPN⁺ cations occupying each cavity; no reduction in dimensionality occurs despite the large cation size. In the presence of water, {Co(H₂O)₂[Au(CN)₂]₂} · ⁿBu₄N[Au(CN)₂] was obtained, a 2-D layered polymer that contains neutral sheets of {Co(H₂O)₂[Au(CN)₂]₂} which are separated by ⁿBu₄N[Au(CN)₂] layers; aurophilic interactions of 3.4250(13) Å and hydrogen-bonding connect the layers. The magnetic properties of all compounds were investigated by SQUID magnetometry. The Ni(II) polymers have similar magnetic behaviour, which are dominated by zero-field splitting with very weak antiferromagnetic interactions at low temperature (D ∼ 2–3 cm⁻¹, zJ < 1 cm⁻¹). The magnetic behaviour of all of the Co(II) polymers were found to be very similar, and dominated by single-ion effects (i.e. a large first-order orbital contribution). No significant magnetic coupling is observed in any of these coordination polymers, suggesting that the [Au(CN)₂] bridging unit behaves as a poor mediator of magnetic exchange in these high-dimensionality systems.     

Density and activity of pertechnetic acid aqueous solutions at T = 298.15 K

The previous isopiestic investigations of HTcO₄ aqueous solutions at T = 298.15 K are believed to be unreliable, because of the formation of a ternary mixture at high molality. Consequently, published isopiestic molalities for aqueous HTcO₄ solutions at T = 298.15 K were completed and corrected. Binary data (variation of the osmotic coefficient and activity coefficient of the electrolyte in solution in the water) at T = 298.15 K for pertechnetic acid HTcO₄ were determined by direct water activity measurements. These measurements extend from molality m = 1.4 mol · kg⁻¹ to m = 8.32 mol · kg⁻¹. The variation of the osmotic coefficient of this acid in water is represented mathematically. Density variations at T = 298.15 K are also established and used to express the activity coefficient values on both the molar and molal concentration scale. The density law leads to the partial molar volume variations for aqueous HTcO₄ solutions at T = 298.15 K, which are compared with published data.     

(Liquid + liquid) equilibria of ternary and quaternary systems containing n-hexane,toluene, m-xylene,propanol, sulfolane,and water at T = 303.15 K

(Liquid + liquid) equilibrium data for ternary and quaternary systems containing n-hexane (C₆H₁₄), toluene (C₇H₈), m-xylene (C₈H₁₀), propanol (C₃H₈O), sulfolane (C₄H₈SO₂), and water (H₂O) were measured at T = 303.15 K. Phase diagrams of {w₁C₄H₈SO₂ + w₂C₇H₈ + (1 − w₁ − w₂)C₆H₁₄}, {w₁C₄H₈SO₂ + w₂C₈H₁₀ + (1 − w₁ − w₂)C₆H₁₄}, {w₁C₄H₈SO₂ + w₂C₃H₈O + w₃C₇H₈ + (1 − w₁ − w₂ − w₃)C₆H₁₄} and also systems containing water: {w₁C₄H₈SO₂ + w₂H₂O + w₃C₇H₈ + (1 − w₁ − w₂ − w₃)C₆H₁₄} and {w₁C₄H₈SO₂ + w₂H₂O + w₃C₈H₁₀ + (1 − w₁ − w₂ − w₃)C₆H₁₄} (w = mass fraction) were obtained at T = 303.15 K. The (liquid + liquid) equilibrium data of the systems were used to obtain interaction parameters in non-random two-liquid (NRTL) and universal quasi-chemical theory (UNIQUAC) activity coefficient models. These parameters can be used to predict equilibrium data of ternary and quaternary systems. The root mean square deviations (RMSDs) using these models were calculated and reported. The partition coefficients and the selectivity factors of solvents for extraction of toluene or m-xylene from n-hexane at T = 303.15 K are calculated and presented. The experimental selectivity factors of sulfolane for the system {w₁C₄H₈SO₂ + w₂C₇H₈ + (1 − w₁ − w₂)C₆H₁₄} at T = 298.15 K and T = 323.15 K were taken from the literature and the influence of temperature on the extraction of toluene was also investigated. The phase diagrams for the ternary and quaternary systems including both the experimental and correlated tie lines are presented. The tie-line data of the studied systems were also correlated using the Hand equation and the correlation parameters are calculated and reported.     

(Liquid + liquid) equilibrium for the ternary systems {heptane + toluene + 1-allyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide} and {heptane + toluene + 1-methyl-3-propylimidazolium bis(trifluoromethylsulfonyl)imide} ionic liquids

The (liquid + liquid) equilibrium (LLE) data for two systems containing heptane, toluene, and 1-methyl-3-propylimidazolium bis(trifluoromethylsulfonyl)imide ([mpim][Tf₂N]) or 1-allyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([amim][Tf₂N]) ionic liquids (ILs) were determined at T = 313.2 K and atmospheric pressure. The effect of a double bond in an alkyl side chain in the imidazolium cation was evaluated in terms of selectivity and extractive capacity. The results show a decrease of the amount of toluene and heptane dissolved in the IL with the allyl group. Thus, the distribution ratios of toluene and heptane of [mpim][Tf₂N] IL are higher than those of [amim][Tf₂N] IL. On the other hand, the separation factor of the [amim][Tf₂N] IL increases comparing to [mpim][Tf₂N] IL. The NRTL model was used to correlate satisfactorily the experimental LLE data for the two studied ternary systems.     

Conductometric study of some alkali metal halides in (dimethyl sulfoxide + acetonitrile) at T = 298.15 K

Electrolytic conductivities of some alkali metal halides, MX (M⁺ = Li⁺, Na⁺, and K⁺; X^? = Cl^?, Br^?, and I^?), NaBPh₄ and Bu₄NBr have been investigated in (20, 40, and 60) mass% {dimethyl sulfoxide (DMSO) in DMSO + acetonitrile} at T = 298.15 K. The conductance results have been analyzed by the Fuoss-conductance-concentration equation in terms of the limiting molar conductance Λ° the association constant K_A and the association diameter R. The ionic contributions to the limiting molar conductance have been estimated using Bu₄NBPh₄ as the “reference electrolyte”. The association constant K_A tends to increase in the order mass percent 20 < 40 < 60 DMSO in (DMSO + acetonitrile) which is explained by the thermodynamic parameter ΔG° and Walden product Λ°η. The results have been interpreted in terms of ion–solvent interactions and structural changes in the mixed solvents.     

Thermodynamics of dibenzo-p-dioxin and 1,2,3,4-tetrachlorodibenzo-p-dioxin fromT = 5 K toT = 490 K

In adiabatic low-pressure and dynamic calorimeters the temperature dependence of the standard molar heat capacity C_{p, m}^oof dibenzo- p -dioxin and 1,2,3,4-tetrachlorodibenzo- p -dioxin have been determined at temperatures in the range T =  5 K to T =  490 K: from T =  5 K to T =  340 K with an accuracy of about 0.2 per cent and with an accuracy of 0.5 per cent to 1.5 per cent between T =  340 K and T =  490 K. The temperatures, enthalpies, and entropies of melting of the above compounds have been determined. The experimental data were used to calculate the thermodynamic functions C_{p, m}^o / R, Δ₀^TH_m^o / (R·K), Δ₀^TS_m^o / R, and Φ_m^o = Δ₀^TS_m^o − Δ₀^TH_m^o / T(where R is the universal gas constant) in the range T →  0 to T =  490 K. The isochoric heat capacity C_{V, m}of both dioxins has been estimated over the range T →  0 to T_fus. The effect of substitution of four hydrogen atoms by chlorine atoms on the lattice and atomic components of the isochoric heat capacity was considered.     

Volumetric properties of ternary (IL + 2-propanol or 1-butanol or 2-butanol + ethyl acetate) systems and binary (IL + 2-propanol or 1-butanol or 2-butanol) and (1-butanol or 2-butanol + ethyl acetate) systems

The experimental densities for the binary or ternary systems were determined at T = (298.15, 303.15, and 313.15) K. The ionic liquid methyl trioctylammonium bis(trifluoromethylsulfonyl)imide ([MOA]⁺[Tf₂N]⁻) was used for three of the five binary systems studied. The binary systems were ([MOA]⁺[Tf₂N]⁻ + 2-propanol or 1-butanol or 2-butanol) and (1-butanol or 2-butanol + ethyl acetate). The ternary systems were {methyl trioctylammonium bis(trifluoromethylsulfonyl)imide + 2-propanol or 1-butanol or 2-butanol + ethyl acetate}. The binary and ternary excess molar volumes for the above systems were calculated from the experimental density values for each temperature. The Redlich–Kister smoothing polynomial was fitted to the binary excess molar volume data. Virial-Based Mixing Rules were used to correlate the binary excess molar volume data. The binary excess molar volume results showed both negative and positive values over the entire composition range for all the temperatures.The ternary excess molar volume data were successfully correlated with the Cibulka equation using the Redlich–Kister binary parameters.     

Calorimetric investigation of mixing enthalpy of liquid (Co + Cu + Zr) alloys at T = 1873 K

The enthalpies of mixing of liquid (Co + Cu + Zr) alloys have been determined using the high-temperature isoperibolic calorimeter. The measurements have been performed along three sections (x_Co/x_Cu = 3/1, 1/1, 1/3) with x_Zr = 0 to 0.55 at T = 1873 K. Over the investigated composition range, the partial mixing enthalpies of zirconium are negative. The limiting partial enthalpies of mixing of undercooled liquid zirconium in liquid (Co + Cu) alloys are (−138 ± 18) kJ · mol⁻¹ (the section x_Co/x_Cu = 3/1), (−155 ± 10) kJ · mol⁻¹ (the section x_Co/x_Cu = 1/1), and (−130 ± 22) kJ · mol⁻¹ (the section x_Co/x_Cu = 1/3). The integral mixing enthalpies are sign-changing. The isenthalpic curves have been plotted on the Gibbs triangle. The main features of the composition dependence of the integral mixing enthalpy of liquid ternary alloys are defined by the pair (Co + Zr) and (Cu + Zr) interactions.