Equations of State: Peng-Robinson and SRK

Equations of state are indispensable tools in chemical engineering for predicting the pressure-volume-temperature (PVT) behavior of fluids, which is critical for designing separation processes, optimizing reactors, and simulating entire plants. The Peng-Robinson (PR) and Soave-Redlich-Kwong (SRK) equations represent the pinnacle of cubic equations of state, offering a pragmatic balance between simplicity and accuracy, especially for hydrocarbon systems and vapor-liquid equilibria calculations. Understanding their derivation, parameterization, and application allows you to reliably model real fluid behavior beyond the ideal gas law.

Foundations of Cubic Equations of State

A cubic equation of state is any thermodynamic relationship that can be rearranged into a cubic polynomial in terms of molar volume or compressibility factor. This cubic form inherently predicts the possibility of phase changes, making these equations powerful for calculating vapor-liquid equilibria. The journey began with the van der Waals equation, which introduced concepts of molecular attraction and repulsion through parameters $a$ and $b$. While foundational, van der Waals' model lacked accuracy for engineering calculations. The Redlich-Kwong equation improved upon it by making the attraction term temperature-dependent. The SRK and PR equations are direct, refined descendants that further modify this temperature dependence using the acentric factor $\omega$, a dimensionless parameter that quantifies molecular non-sphericity and polarity. This evolution means you can use a single, relatively simple equation to model complex real-fluid behavior from the critical point to subcooled liquids.

The Peng-Robinson and Soave-Redlich-Kwong Equations

Both the Peng-Robinson and Soave-Redlich-Kwong equations are explicit in pressure and take a similar form, but with different algebraic structures that affect their accuracy for density and equilibrium calculations.

The Soave-Redlich-Kwong (SRK) equation is given by: $$P=\frac{RT}{V-b}-\frac{a\alpha (T)}{V(V+b)}$$ Here, $P$ is pressure, $R$ is the universal gas constant, $T$ is temperature, $V$ is molar volume, $b$ is a parameter representing molecular volume, and $a$ is a parameter representing intermolecular attraction forces. The function $\alpha (T)$ introduces temperature dependency to the attraction term.

The Peng-Robinson (PR) equation modifies the denominator of the attractive term: $$P=\frac{RT}{V-b}-\frac{a\alpha (T)}{V(V+b)+b(V-b)}$$ This adjustment was made to better represent liquid densities and vapor-liquid equilibria near the critical point. In both equations, the key to their improved performance over earlier models lies in the formulation of $\alpha (T)$, which incorporates the acentric factor.

Determining Parameters from Critical Properties and Acentric Factor

For a pure component, the parameters $a$ and $b$ are not arbitrary; they are determined from the substance's critical properties—critical temperature $T_c$ and critical pressure $P_c$—and its acentric factor $\omega$. This ensures the equation correctly reproduces the critical point and leverages known physical data.

The general procedure involves applying the critical point conditions: $(\partial P/\partial V{)}_T=0$ and $({\partial }^{2}P/\partial V^2{)}_T=0$ at $T=T_c$ and $P=P_c$. For both SRK and PR, this yields expressions for $a$ and $b$ at the critical temperature. For SRK: $$a_c=0.42747\frac{R^2{T}_c^2}{P_c}\text{and}b=0.08664\frac{R{T}_c}{P_c}$$ For PR: $$a_c=0.45724\frac{R^2{T}_c^2}{P_c}\text{and}b=0.07780\frac{R{T}_c}{P_c}$$ The temperature-dependent function $\alpha (T)$ is then defined as $\alpha (T)=[1+m(1-\sqrt{T_r}){]}^2$, where $T_r=T/T_c$ is the reduced temperature. The parameter $m$ is correlated with the acentric factor. For SRK: $m=0.480+1.574\omega -0.176{\omega }^2$. For PR: $m=0.37464+1.54226\omega -0.26992{\omega }^2$. Thus, by knowing $T_c$, $P_c$, and $\omega$, you can fully parameterize these equations for any pure fluid.

Mixing Rules and Fugacity Coefficients for Mixtures

In chemical engineering, you rarely deal with pure components; mixtures are the norm. To apply cubic EOS to a multicomponent system, mixing rules are used to compute the equation's parameters for the blend. The most common approach uses quadratic (van der Waals one-fluid) mixing rules for the attraction parameter $a$ and linear mixing for the co-volume $b$:

$$a=\sum _i\sum _j{y}_i{y}_j{a}_{ij}\text{and}b=\sum _i{y}_i{b}_i$$

Here, $y_i$ is the mole fraction of component $i$. The cross-parameter $a_{ij}$ is typically calculated using a geometric mean combined with a binary interaction parameter $k_{ij}$ to account for deviations from ideal mixing: $a_{ij}=\sqrt{a_i{a}_j}(1-k_{ij})$. For many hydrocarbon pairs, $k_{ij}$ is near zero, but for mixtures involving polar components or gases like $H_{2}S$, it must be fitted from experimental data.

The primary utility of EOS in separation process design is calculating phase equilibria, which requires the fugacity coefficient ${\hat{\varphi }}_i$. For a component in a mixture using a cubic EOS, the fugacity coefficient is derived from the thermodynamic relationship: $$\ln {\hat{\varphi }}_i=\frac{1}{RT}{\int }_V^{\infty }\left[{\left(\frac{\partial P}{\partial n_i}\right)}_{T,V,n_j}-\frac{RT}{V}\right]dV-\ln Z$$ where $Z=PV/RT$ is the compressibility factor. Performing this derivation for the PR or SRK equation yields an explicit, albeit algebraic, expression. For example, for the Peng-Robinson EOS: $$\ln {\hat{\varphi }}_i=\frac{b_i}{b}(Z-1)-\ln (Z-B)+\frac{A}{2\sqrt{2}B}\left(\frac{2{\sum }_j{y}_j{a}_{ij}}{a}-\frac{b_i}{b}\right)\ln \left(\frac{Z+(1+\sqrt{2})B}{Z+(1-\sqrt{2})B}\right)$$ where $A=aP/(RT{)}^2$ and $B=bP/RT$. You then solve for $Z$ from the cubic EOS and compute ${\hat{\varphi }}_i$ for each component in each phase, setting fugacities equal to solve for equilibrium.

Accuracy Comparison and Practical Applications

The choice between Peng-Robinson and Soave-Redlich-Kwong often hinges on the specific fluid type and property of interest. For hydrocarbon systems and natural gas processing, both equations are excellent for predicting vapor-liquid equilibrium (VLE) and saturation pressures. The Peng-Robinson equation generally provides more accurate liquid density predictions, which is crucial for sizing pumps and vessels. For light hydrocarbons and gases, SRK is often preferred for its accuracy in vapor phase properties and enthalpy departures.

However, for highly polar fluids (e.g., water, alcohols) or asymmetric mixtures, both models can struggle without tuned binary interaction parameters. In practice, PR is frequently the default in modern process simulators for general petroleum and chemical applications due to its all-around performance. A key advantage of these cubic EOS is their ability to describe the entire fluid region (vapor, liquid, supercritical) with a single set of parameters, making them computationally efficient for flash calculations and process simulation.

Common Pitfalls

1. Ignoring the Acentric Factor: Using default or zero acentric factor for components where it is significant leads to large errors in saturation pressure predictions, especially for heavier hydrocarbons. Always verify and use the correct $\omega$ from reliable databases.
2. Misapplying Mixing Rules: Assuming binary interaction parameters $k_{ij}$ are zero for all mixtures. For non-ideal systems like those containing carbon dioxide or hydrogen, failing to use appropriate $k_{ij}$ values (often found in literature or simulation databanks) can render phase equilibrium calculations inaccurate.
3. Incorrect Root Selection for Compressibility Factor: Solving the cubic EOS for $Z$ yields three roots in the two-phase region. The smallest positive root corresponds to the liquid phase, the largest to the vapor phase, and the middle is physically meaningless. Automatically choosing the wrong root can lead to incorrect phase identification and property calculations.
4. Overextending the Model: Cubic EOS like PR and SRK are not universally accurate. For polymers, ionic liquids, or strongly associating fluids, more complex equations (like SAFT) are required. Recognize the limitations and validate predictions against experimental data when possible.

Summary

• Cubic equations of state, particularly Peng-Robinson and Soave-Redlich-Kwong, are workhorse models for calculating thermodynamic properties and phase equilibria in chemical engineering by introducing temperature-dependent attraction terms.
• The parameters $a$ and $b$ are determined from a substance's critical temperature, critical pressure, and acentric factor, which allows the models to accurately represent real fluid behavior.
• For multicomponent systems, quadratic mixing rules with binary interaction parameters are used to extend the pure-component equations to mixtures.
• Calculating fugacity coefficients from these EOS enables the determination of vapor-liquid equilibrium, which is fundamental to designing distillation, absorption, and other separation processes.
• While both models excel with hydrocarbons, Peng-Robinson tends to better predict liquid densities, whereas SRK may be favored for vapor phases and light gases; accuracy for polar fluids requires careful parameterization.