Lee–Kesler method

The Lee–Kesler method allows the estimation of the saturated vapor pressure at a given temperature for all components for which the critical pressure Pc, the critical temperature Tc, and the acentric factor ω are known.

Equations

 * $$ \ln P_{\rm r} = f^{(0)} + \omega \cdot f^{(1)} $$


 * $$ f^{(0)}=5.92714 - \frac{6.09648}{T_{\rm r}} - 1.28862 \cdot \ln T_{\rm r} + 0.169347 \cdot T_{\rm r}^6 $$


 * $$ f^{(1)}=15.2518 - \frac{15.6875}{T_{\rm r}}-13.4721 \cdot \ln T_{\rm r} + 0.43577 \cdot T_{\rm r}^6 $$

with


 * $$P_{\rm r}=\frac{P}{P_{\rm c}}$$ (reduced pressure) and $$T_{\rm r}=\frac{T}{T_{\rm c}}$$ (reduced temperature).

Typical errors
The prediction error can be up to 10% for polar components and small pressures and the calculated pressure is typically too low. For pressures above 1 bar, that means, above the normal boiling point, the typical errors are below 2%.

Example calculation
For benzene with
 * Tc = 562.12 K
 * Pc = 4898 kPa
 * Tb = 353.15 K
 * ω = 0.2120

the following calculation for T=Tb results:


 * Tr = 353.15 / 562.12 = 0.628247
 * f(0) = -3.167428
 * f(1) = -3.429560
 * Pr = exp( f(0) + ω f(1) ) = 0.020354
 * P = Pr * Pc = 99.69 kPa

The correct result would be P = 101.325 kPa, the normal (atmospheric) pressure. The deviation is -1.63 kPa or -1.61 %.

It is important to use the same absolute units for T and Tc as well as for P and Pc. The unit system used (K or R for T) is irrelevant because of the usage of the reduced values Tr and Pr.