Module 3: Applied Fluid Phase Equilibria

Introduction
This module focuses on the practical applications of thermodynamic principles and
equations of state to solve real-world fluid phase equilibria problems. Building on the
foundations established in Modules 1 and 2, this module explores methods for modeling
binary and multicomponent mixtures, identifying non-ideal behaviors, and conducting
various equilibrium calculations essential for process design and analysis.

Learning Outcomes
1. Model binary and multicomponent mixtures using fugacity and activity
coefficient models (e.g., Van Laar, Margules, NRTL, Peng-Robinson).

2. Identify non-ideal behaviors and their impacts on phase equilibria.

3. Conduct equilibrium calculations (bubble point, dew point, liquid-liquid

equilibrium) with computational tools.

4. Validate theoretical predictions against experimental data and explain

discrepancies.

5. Generate and interpret equilibrium diagrams (T-x-y, P-x-y, LLE) using

computational tools.

Key Concepts

Principle of Phase Equilibrium

The fundamental principle of phase equilibrium states that for a component to be at

equilibrium between multiple phases, its fugacity must be equal in all phases:

fᵢᵅ = fᵢᵝ = ... = fᵢⁿ

Where fᵢ is the fugacity of component i in each phase (α, β, ..., n).

This principle is the basis for all phase equilibrium calculations and can be applied to
various types of equilibria: - Vapor-liquid equilibrium (VLE) - Liquid-liquid equilibrium
(LLE) - Solid-liquid equilibrium (SLE) - Vapor-liquid-liquid equilibrium (VLLE)
Modeling Pure Compounds

For pure compounds, the vapor pressure is a critical property for phase equilibrium
calculations. Several equations are used to estimate vapor pressure as a function of
temperature:

Antoine Equation

log₁₀(P°) = A - B/(T + C)

Where P° is the vapor pressure, T is temperature, and A, B, and C are compound-specific

constants.

Wagner Equation

ln(P°/Pc) = (a·τ + b·τ^1.5 + c·τ^3 + d·τ^6)·Tc/T

Where τ = 1 - T/Tc, and a, b, c, and d are compound-specific constants.

Modeling Mixtures

Ideal Solutions

For ideal solutions, Raoult's Law applies: yᵢP = xᵢP°ᵢ

Where: - yᵢ is the vapor phase mole fraction - xᵢ is the liquid phase mole fraction - P is the
total pressure - P°ᵢ is the vapor pressure of pure component i

Non-Ideal Solutions

For non-ideal solutions, modified forms of Raoult's Law are used:

Activity Coefficient Approach

yᵢφᵢP = xᵢγᵢP°ᵢ

Where: - φᵢ is the fugacity coefficient in the vapor phase - γᵢ is the activity coefficient in
the liquid phase

Equation of State Approach

yᵢφᵢᵛP = xᵢφᵢˡP

Where φᵢᵛ and φᵢˡ are the fugacity coefficients in the vapor and liquid phases,
respectively, calculated from an equation of state.
Activity Coefficient Models

Several models are used to calculate activity coefficients for non-ideal liquid mixtures:

Margules Equation

For a binary mixture: ln(γ₁) = A₁₂x₂² ln(γ₂) = A₂₁x₁²

For the two-suffix Margules equation, A₁₂ = A₂₁.

Van Laar Equation

ln(γ₁) = A₁₂[B₂₁x₂/(A₁₂x₁ + B₂₁x₂)]² ln(γ₂) = B₂₁[A₁₂x₁/(A₁₂x₁ + B₂₁x₂)]²

NRTL (Non-Random Two-Liquid) Model

ln(γ₁) = x₂²[τ₂₁(G₂₁/(x₁ + x₂G₂₁))² + τ₁₂G₁₂/(x₂ + x₁G₁₂)²]

Where: - G₁₂ = exp(-α₁₂τ₁₂) - G₂₁ = exp(-α₂₁τ₂₁) - τ₁₂ = (g₁₂ - g₂₂)/RT - τ₂₁ = (g₂₁ - g₁₁)/RT - α is
the non-randomness parameter

The NRTL model is particularly useful for highly non-ideal systems, including those
exhibiting liquid-liquid immiscibility.

UNIQUAC (Universal Quasi-Chemical) Model

The UNIQUAC model combines combinatorial and residual terms: ln(γᵢ) = ln(γᵢᶜ) + ln(γᵢʳ)

The combinatorial term accounts for molecular size and shape differences, while the
residual term accounts for energy interactions.

Non-Ideal Behaviors

Azeotropes

An azeotrope is a mixture that has the same composition in the vapor and liquid phases
at equilibrium (yᵢ = xᵢ). Azeotropes can be: - Minimum-boiling (positive deviation from
Raoult's Law) - Maximum-boiling (negative deviation from Raoult's Law)

Azeotropes present challenges for separation processes like distillation, as the

composition cannot be changed by simple boiling or condensation at the azeotropic
point.
Liquid-Liquid Immiscibility

Some liquid mixtures exhibit partial miscibility, forming two liquid phases. This occurs
when the Gibbs free energy of mixing has multiple minima.

The condition for liquid-liquid equilibrium is: xᵢᵅγᵢᵅ = xᵢᵝγᵢᵝ

Where α and β represent the two liquid phases.

Retrograde Condensation

Retrograde condensation is a phenomenon where a decrease in pressure at constant

temperature causes a vapor to condense, contrary to the usual behavior. This occurs
near the critical point and is important in natural gas processing.

Equilibrium Calculations

Bubble Point Calculation

The bubble point is the condition where the first vapor bubble forms from a liquid
mixture. The calculation involves finding the temperature or pressure that satisfies:
Σ(xᵢKᵢ) = 1

Where Kᵢ = yᵢ/xᵢ is the equilibrium ratio or K-value.

Dew Point Calculation

The dew point is the condition where the first liquid droplet forms from a vapor mixture.
The calculation involves finding the temperature or pressure that satisfies: Σ(yᵢ/Kᵢ) = 1

Flash Calculations

Flash calculations determine the amounts and compositions of vapor and liquid phases
in equilibrium. The Rachford-Rice equation is commonly used: Σ[zᵢ(Kᵢ-1)/(1+V/F(Kᵢ-1))] = 0

Where: - zᵢ is the overall mole fraction - V/F is the vapor fraction - Kᵢ is the equilibrium
ratio

Liquid-Liquid Equilibrium Calculations

For liquid-liquid equilibrium, the distribution coefficient is defined as: Dᵢ = xᵢᵅ/xᵢᵝ

The calculation involves finding compositions that satisfy the equal activity criterion:
xᵢᵅγᵢᵅ = xᵢᵝγᵢᵝ
Equilibrium Diagrams

Temperature-Composition (T-x-y) Diagrams

T-x-y diagrams show the relationship between temperature and composition at constant
pressure. They include: - Bubble point curve (liquid compositions) - Dew point curve
(vapor compositions) - Azeotropic points (where the curves meet)

Pressure-Composition (P-x-y) Diagrams

P-x-y diagrams show the relationship between pressure and composition at constant
temperature. They are analogous to T-x-y diagrams but with pressure as the vertical axis.

Liquid-Liquid Equilibrium (LLE) Diagrams

LLE diagrams show the compositions of coexisting liquid phases as a function of

temperature. Key features include: - Binodal curve (boundary of the two-phase region) -
Tie lines (connecting equilibrium compositions) - Critical solution temperature (where
the two phases become identical)

Ternary Diagrams

For three-component systems, triangular diagrams are used to represent compositions.

These can show: - Vapor-liquid equilibrium regions - Liquid-liquid equilibrium regions -
Three-phase regions

Computational Approaches

Equation-Based Methods

Equation-based methods solve the governing equations directly: - Newton-Raphson

method for nonlinear equations - Successive substitution for iterative solutions -
Homotopy continuation for difficult convergence cases

Gibbs Energy Minimization

This approach finds the equilibrium state by minimizing the total Gibbs energy of the
system: G = Σᵢ Σⱼ nᵢⱼμᵢⱼ

Where nᵢⱼ is the number of moles of component i in phase j, and μᵢⱼ is the chemical
potential.
Process Simulators

Commercial process simulators (Aspen Plus, HYSYS, PRO/II) incorporate various

thermodynamic models and calculation methods for phase equilibria.

Example Problems

Problem 1: Bubble Point Calculation

Question: A liquid mixture contains 40 mol% n-hexane and 60 mol% n-heptane at 1 atm.
The vapor pressures at 80°C are P°hexane = 1.2 atm and P°heptane = 0.6 atm. Assuming
ideal solution behavior, calculate the bubble point temperature.

Solution: For an ideal solution at the bubble point: Σ(xᵢP°ᵢ) = P

Substituting the given values: 0.4 × 1.2 + 0.6 × 0.6 = 0.48 + 0.36 = 0.84 atm

Since the total pressure (1 atm) is greater than the calculated pressure (0.84 atm), the
mixture is not at its bubble point at 80°C. The temperature must be increased to raise the
vapor pressures.

Using the Antoine equation and iterative calculations, the bubble point temperature is
found to be approximately 87°C.

Problem 2: Activity Coefficient Calculation

Question: A binary mixture of components 1 and 2 follows the NRTL model with
parameters τ₁₂ = 1.5, τ₂₁ = 0.8, and α = 0.3. Calculate the activity coefficients for a mixture
with x₁ = 0.4 and x₂ = 0.6.

Solution: Step 1: Calculate G₁₂ and G₂₁ G₁₂ = exp(-α·τ₁₂) = exp(-0.3 × 1.5) = exp(-0.45) =
0.6376 G₂₁ = exp(-α·τ₂₁) = exp(-0.3 × 0.8) = exp(-0.24) = 0.7866

Step 2: Calculate ln(γ₁) ln(γ₁) = x₂²[τ₂₁(G₂₁/(x₁ + x₂G₂₁))² + τ₁₂G₁₂/(x₂ + x₁G₁₂)²] ln(γ₁) =

(0.6)²[0.8(0.7866/(0.4 + 0.6×0.7866))² + 1.5×0.6376/(0.6 + 0.4×0.6376)²] ln(γ₁) =
0.36[0.8(0.7866/0.872)² + 1.5×0.6376/(0.6 + 0.255)²] ln(γ₁) = 0.36[0.8(0.9021)² +
1.5×0.6376/0.855²] ln(γ₁) = 0.36[0.8×0.8138 + 1.5×0.6376/0.731] ln(γ₁) = 0.36[0.651 +
1.5×0.872] ln(γ₁) = 0.36[0.651 + 1.308] ln(γ₁) = 0.36 × 1.959 = 0.705

Therefore, γ₁ = exp(0.705) = 2.024

Step 3: Calculate ln(γ₂) ln(γ₂) = x₁²[τ₁₂(G₁₂/(x₂ + x₁G₁₂))² + τ₂₁G₂₁/(x₁ + x₂G₂₁)²] ln(γ₂) =

(0.4)²[1.5(0.6376/(0.6 + 0.4×0.6376))² + 0.8×0.7866/(0.4 + 0.6×0.7866)²] ln(γ₂) =
0.16[1.5(0.6376/0.855)² + 0.8×0.7866/0.872²] ln(γ₂) = 0.16[1.5(0.746)² + 0.8×0.7866/0.76]
ln(γ₂) = 0.16[1.5×0.557 + 0.8×1.035] ln(γ₂) = 0.16[0.836 + 0.828] ln(γ₂) = 0.16 × 1.664 =
0.266

Therefore, γ₂ = exp(0.266) = 1.305

Problem 3: Flash Calculation

Question: A feed stream containing 40 mol% propane and 60 mol% n-butane at 350 K
and 6 bar undergoes a flash separation. If the K-values at these conditions are Kpropane
= 2.5 and Kn-butane = 0.8, determine the vapor fraction and the compositions of the
vapor and liquid phases.

Solution: Step 1: Use the Rachford-Rice equation to find the vapor fraction (ψ = V/F)
Σ[zᵢ(Kᵢ-1)/(1+ψ(Kᵢ-1))] = 0

Substituting the given values: [0.4(2.5-1)/(1+ψ(2.5-1))] + [0.6(0.8-1)/(1+ψ(0.8-1))] = 0

[0.4(1.5)/(1+1.5ψ)] + [0.6(-0.2)/(1-0.2ψ)] = 0 [0.6/(1+1.5ψ)] - [0.12/(1-0.2ψ)] = 0

Solving this equation iteratively: For ψ = 0.6: [0.6/(1+1.5×0.6)] - [0.12/(1-0.2×0.6)] =

[0.6/1.9] - [0.12/0.88] = 0.316 - 0.136 = 0.18

For ψ = 0.625: [0.6/(1+1.5×0.625)] - [0.12/(1-0.2×0.625)] = [0.6/1.9375] - [0.12/0.875] =

0.31 - 0.137 = 0.173

For ψ = 0.63: [0.6/(1+1.5×0.63)] - [0.12/(1-0.2×0.63)] = [0.6/1.945] - [0.12/0.874] = 0.308 -

0.137 = 0.171

Continuing this process, the vapor fraction is approximately ψ = 0.625 (or 62.5%).

Step 2: Calculate the compositions of the vapor and liquid phases For component i: xᵢ =
zᵢ/[1+ψ(Kᵢ-1)] yᵢ = Kᵢxᵢ

For propane: x₁ = 0.4/[1+0.625(2.5-1)] = 0.4/[1+0.625×1.5] = 0.4/1.9375 = 0.206 y₁ = 2.5 ×

0.206 = 0.516

For n-butane: x₂ = 0.6/[1+0.625(0.8-1)] = 0.6/[1+0.625×(-0.2)] = 0.6/0.875 = 0.686 y₂ = 0.8

× 0.686 = 0.549

Step 3: Normalize the compositions Liquid phase: x₁ = 0.206/(0.206+0.686) = 0.231, x₂ =

0.769 Vapor phase: y₁ = 0.516/(0.516+0.549) = 0.484, y₂ = 0.516

Therefore, the flash separation produces a vapor stream containing 48.4% propane and
51.6% n-butane, and a liquid stream containing 23.1% propane and 76.9% n-butane.
Summary
Module 3 applies the fundamental concepts from previous modules to practical phase
equilibria problems:

1. The principle of phase equilibrium, based on equal fugacity, is the foundation for
all equilibrium calculations.

2. Various models (Margules, Van Laar, NRTL, UNIQUAC) are used to account for non-
ideal behavior in liquid mixtures.

3. Non-ideal behaviors like azeotropes, liquid-liquid immiscibility, and retrograde

condensation present challenges and opportunities in separation processes.

4. Equilibrium calculations (bubble point, dew point, flash, LLE) are essential tools for
process design and analysis.

5. Equilibrium diagrams provide visual representations of phase behavior and are

valuable for understanding and designing separation processes.

6. Computational approaches, including equation-based methods and Gibbs energy

minimization, enable the solution of complex phase equilibria problems.

These applied concepts and calculation methods are essential for solving real-world
problems in chemical engineering, petroleum refining, environmental science, and
many other fields.