Statistical associating fluid theory
(SAFT) equations of state
(EoSs) are powerful thermodynamic modeling tools that show promise
in application to a wide range of different properties and systems.
SAFT-γ Mie, the group-contribution variant of the state-of-the-art
SAFT-VR Mie, can describe new systems using transferable functional-group
parameters. There had been a void in the modeling of nonself-associating
dipolar species prior to this work, in which groups were parametrized
for 2-ketones, 3-ketones, and n-alkyl propanoates
(viz. CH2CO, CH3CO, and COOpr., respectively).
These components occur in a wide variety of industrial processes and
modeling them with SAFT-γ Mie presented the opportunity to evaluate
the model’s treatment of dipolar interactions without a fundamental
dipolar term in the EoS. Our new groups provide reliable binary mixture
phase-equilibrium, excess enthalpy, and speed of sound predictions
for all of the considered components, despite the fact that 2-ketone
pure-component predictions are slightly less accurate than what is
expected from such a complex SAFT model. The latter observation suggests
that very precise modeling of smaller, highly dipolar molecules is
challenging with a first-order group-contribution model, even in the
SAFT-VR Mie-based framework. Binary mixture VLE predictions of ketone
+ n-alkane and ketone +1-alkanol systems are in good
agreement with experimental data, suggesting that the pseudo-association
approach used to treat the dipolar interactions of ketones is adequate,
and that its nonrigorous nature does not inherently produce an erroneous
representation of the balance between different intermolecular interactions.
Two successful high-pressure binary system VLE predictions were also
generated, which may spur further research into using SAFT-γ
Mie and the new dipolar groups for modeling high-pressure systems.