Ternary Phase Equilibrium & Extraction theory

The thermodynamic phase rule formulated by J. Willard Gibbs establishes the number of intensive independent variables (degrees of freedom, F) required to completely specify the equilibrium state of a multi-component, multi-phase system:



F = C − P + 2



where C is the number of independent chemical components, P is the number of coexisting phases, and the numeral 2 represents temperature and pressure. For a three-component system (C = 3) such as water + acetic acid + chloroform undergoing liquid-liquid phase separation into two coexisting liquid phases (an aqueous phase and an organic phase, P = 2):



F = 3 − 2 + 2 = 3



Under standard laboratory conditions where both temperature and pressure are fixed externally (T = 25.0 °C and P = 1.0 atm), two degrees of freedom are constrained, reducing the variance to:



F_{T, P} = C − P = 3 − 2 = 1



This single degree of freedom has profound physical consequences: inside the two-phase miscibility gap, the system is univariant. If the investigator fixes the concentration of any one component in either phase (for example, the acetic acid concentration in the aqueous phase, C_aq), the concentrations of all three components in both the aqueous phase and the coexisting organic phase become rigorously fixed by thermodynamics. Every point in the two-phase region lies on a unique equilibrium tie line that connects these two invariant conjugate phase compositions.

Representing a three-component system requires three compositional variables (x₁, x₂, x₃), but because their sum is constrained to unity (x₁ + x₂ + x₃ = 1), only two independent composition variables exist. J. Willard Gibbs devised an equilateral triangular coordinate diagram (often termed a Gibbs triangle or ternary phase diagram) that represents all three components symmetrically.



The geometric foundation of the Gibbs triangle relies on Viviani's theorem: the sum of the perpendicular distances from any interior point P to the three sides of an equilateral triangle is constant and equal to the height H of the triangle. By normalizing H = 100% (or 1.0 mole fraction):



1. Vertices: Each of the three vertices of the triangle represents 100% of a pure component (e.g., top vertex = 100% Acetic Acid, bottom left = 100% Water, bottom right = 100% Chloroform).

2. Edges: The three outer boundary edges represent binary mixtures of the components at their respective vertices (e.g., the bottom edge represents binary water-chloroform mixtures with 0% acetic acid).

3. Interior Grid: Any point inside the triangle represents a ternary mixture. The percentage of component i is read by projecting a line parallel to the edge opposite vertex i, measuring the fractional distance from that edge toward the vertex.

4. Tie Lines: Inside the two-phase region under the binodal solubility curve, the straight line connecting the coexisting aqueous phase composition and organic phase composition is called a tie line. According to the lever rule (the center-of-gravity principle), any overall mixture prepared along a tie line splits into conjugate phases whose relative mass quantities are inversely proportional to the lengths of the segments between the overall mixture point and the phase endpoints.

The molecular behavior of carboxylic acids differs dramatically between polar aqueous environments and non-polar organic media. Liquid water is a protic solvent with a huge dielectric constant (ε ≈ 78.5) and immense hydrogen-bond donor and acceptor capabilities. In water, acetic acid molecules are individually solvated by water molecules forming strong solute-solvent hydrogen bonds; furthermore, acetic acid partially ionizes (K_a = 1.75 × 10⁻⁵). Consequently, acetic acid exists almost exclusively as solvated monomers in the aqueous phase.



In contrast, chloroform (CHCl₃) is a weakly polar, aprotic solvent with a very low dielectric constant (ε ≈ 4.8). In such non-polar media, the solvent cannot effectively solvate the polar carboxylic acid group (−COOH), nor can it screen electrostatic dipole interactions. To minimize their free energy, two acetic acid molecules spontaneously associate via two strong, reciprocal, planar hydrogen bonds between their carbonyl oxygen atoms and hydroxyl hydrogens, forming an eight-membered cyclic dimer:



2 CH₃COOH ⇌ (CH₃COOH)₂



This cyclic dimer structure is thermodynamically favored because it completely satisfies the hydrogen-bonding capacity of both molecules internally, effectively shielding the polar hydrophilic moieties within a non-polar hydrocarbon-like exterior. The dimerization equilibrium is governed by the mass action constant:



K_dim = [(HAc)₂]_org / [HAc]_org,mon²



Because of this strong dimerization (K_dim ≈ 100 L/mol in chloroform), as more acetic acid enters the organic phase, the dimer concentration increases quadratically relative to the monomer. According to the Nernst distribution law, only the monomeric species can cross the phase boundary in thermodynamic equilibrium with aqueous monomers. Consequently, the apparent distribution ratio D = C_org / C_aq is not constant (as it would be for an ideal solute) but increases steeply with concentration, reflecting the massive thermodynamic sink created by dimerization in the organic phase.

To establish the quantitative relationship governing the distribution of acetic acid between water (aq) and an organic solvent (org), we apply thermodynamic mass action principles. In the aqueous phase, ionization is negligible at moderate concentrations (pH < 3), so total aqueous concentration equals monomer concentration: C_aq ≈ [HAc]_aq.



According to Nernst's distribution law, thermodynamic equilibrium requires the chemical potentials of the neutral monomeric species to be equal in both phases: μ_mon^aq = μ_mon^org. Assuming activity coefficients are unity in the dilute regime, the ratio of monomer concentrations across the phase boundary is an invariant thermodynamic constant, the true monomer partition coefficient K_D:



K_D = [HAc]_org,mon / [HAc]_aq ==> [HAc]_org,mon = K_D · C_aq



In the organic phase, acetic acid exists in dynamic equilibrium between monomers and cyclic dimers: 2 HAc_org ⇌ (HAc)₂_org, governed by the dimerization constant:



K_dim = [(HAc)₂]_org / [HAc]_org,mon² ==> [(HAc)₂]_org = K_dim · [HAc]_org,mon²



Substituting the Nernst relation [HAc]_org,mon = K_D·C_aq into the dimer concentration yields:



[(HAc)₂]_org = K_dim · (K_D · C_aq)² = K_dim · K_D² · C_aq²



The total analytical concentration of acetic acid in the organic phase, C_org (measured by NaOH titration), represents the sum of monomer and dimer moles of monomer units:



C_org = [HAc]_org,mon + 2 · [(HAc)₂]_org



Substituting our expressions for monomer and dimer concentrations:



C_org = K_D · C_aq + 2 · K_dim · K_D² · C_aq²



Dividing both sides of this equation by the aqueous concentration C_aq gives the apparent distribution ratio D:



D = C_org / C_aq = K_D + (2 · K_dim · K_D²) · C_aq



This is the celebrated Nernst-Christian distribution equation. It demonstrates that plotting the experimental distribution ratio D (y-axis) against the aqueous acid concentration C_aq (x-axis) yields a strictly linear relationship where the y-intercept directly equals the monomer partition coefficient K_D, and the slope m equals 2·K_dim·K_D². Dimerization constant K_dim is then calculated directly: K_dim = m / (2·K_D²).

On a ternary liquid-liquid phase diagram, the binodal curve delineates the boundary between the homogeneous single-phase region above and the heterogeneous two-phase region below. Inside this two-phase dome, an infinite family of tie lines connects coexisting aqueous and organic conjugate phases. As the concentration of the consolute component (acetic acid) increases, the mutual solubility of water and chloroform expands, and the compositions of the two coexisting phases become progressively closer together. Consequently, the length of the tie lines progressively shrinks.



The unique critical point on the binodal curve where the tie line length shrinks to exactly zero is called the plait point (or critical solution point). At the plait point, the compositions of the coexisting aqueous and organic phases become identical in every physical and chemical property (density, refractive index, surface tension, and composition), the interfacial tension vanishes, and the two phases merge into a single critical phase.



Experimentally, locating the plait point is challenging because tie lines do not converge to the apex of the binodal curve; rather, due to asymmetric solubility, the plait point is almost always displaced to one side of the dome. Two classic graphical and empirical methods are used to determine the plait point precisely: (1) The Hand Plot: plotting ln(w_acid^aq / w_water^aq) against ln(w_acid^org / w_chloroform^org) yields a straight line. The condition where both phase compositions become identical corresponds to w_acid^aq = w_acid^org and w_water^aq = w_chloroform^org; the intersection of this line with the independently determined binodal curve locates the exact coordinates of the plait point; (2) The Conjugate Curve (Trevoy-Scheibel method): connecting the midpoints of the tie lines constructs a smooth conjugate locus that terminates precisely at the plait point on the binodal boundary.

High-precision liquid-liquid equilibrium metrology is subject to several severe systematic experimental errors that must be rigorously controlled:



1. Incomplete Phase Separation and Micro-Emulsions: Vigorous shaking produces micro-droplets of the organic phase suspended in the aqueous layer, or vice versa (cloudy emulsions). Draining a turbid phase introduces suspended droplets of the opposite phase into the titration aliquot, distorting concentration measurements. To eliminate this, the separatory funnels must be allowed to settle undisturbed for at least 30 minutes in a draft-free constant-temperature bath until both layers achieve optical clarity.



2. Chloroform Volatility and Vapor Headspace Evaporation: Chloroform has a high vapor pressure (P_vap ≈ 26 kPa at 25 °C). If sampling is performed slowly in open beakers, chloroform rapidly evaporates, causing the apparent concentration of acetic acid in the organic phase to be systematically overestimated. Using pre-chilled, airtight glass syringes and transferring aliquots directly into covered titration flasks containing water prevents evaporation.



3. Stopcock Bore Cross-Contamination: The narrow bore of the separatory funnel stopcock contains dense chloroform from the previous draining step. Before collecting an analytical sample, the operator must drain and discard at least 0.5 mL of the phase to flush the stopcock bore and neck.



4. Temperature Sensitivity of the Binodal Boundary: Mutual solubility is highly sensitive to temperature. A 1.0 °C temperature change shifts the binodal curve and alters partition coefficients by 2% to 4%. All equilibrations, phase separations, and titrations must be maintained strictly at 25.00 ± 0.05 °C.



5. Water Extraction into the Organic Phase: Glacial acetic acid carries a small quantity of water with it into the organic phase as mutual solubility increases. By performing titrations in a mixed water-ethanol solvent, end-point sharpness with phenolphthalein is preserved without phase separation during titrant addition.

Liquid-liquid extraction (solvent extraction) is an essential separation operation used extensively in chemical processing, hydrometallurgy, biotechnology, and nuclear fuel reprocessing, particularly when distilling heat-sensitive compounds or separating components with close boiling points or azeotropes (such as recovering acetic acid from dilute industrial wastewater).



In industrial extraction, a liquid feed mixture (consisting of solute A and carrier solvent B, such as acetic acid in water) is contacted with an immiscible extracting solvent C (such as ethyl acetate or chloroform). The process is executed across multiple countercurrent stages (in mixer-settlers or packed extraction columns):



1. Triangular Operating Diagrams: The ternary phase diagram directly dictates the theoretical feasibility and maximum recovery of the extraction. The feed composition F and solvent S define a straight operating line across the two-phase dome. The mixture separates into an extract phase E (rich in extracting solvent C and solute A) and a raffinate phase R (rich in carrier solvent B, depleted in solute A).



2. Hunter-Nash Graphical Construction: By projecting the tie lines and material balance operating lines to a common difference point (pole point Δ), chemical engineers step off theoretical equilibrium stages across the ternary diagram, exactly analogous to the McCabe-Thiele method for distillation. The number of required stages, minimum solvent-to-feed ratio (S/F)_min, and maximum achievable solute purity are determined directly from the slope of the tie lines and the position of the plait point.



3. Selectivity and Solvent Selection: The effectiveness of solvent C is quantified by the separation factor (selectivity) β = (x_A / x_B)_extract / (x_A / x_B)_raffinate = D_A / D_B. A high distribution coefficient D_A combined with low mutual solubility of B and C ensures sharp phase separation, low solvent circulation rates, and minimal energy requirements in downstream solvent recovery strippers.