(169b) The Effect of Solute-Solute Interactions in the Presence of CO2 on the High Pressure Thermodynamic Behaviour of CO2 + n-Alkane + 1-Alcohol Systems

Detergent range alcohols are used in the
production of surfactants. These alcohols, with 10 to 16 carbon atoms, are
often produced by grafting a hydroxyl group onto an alkane molecule. However,
the alcohol conversion is often incomplete. Additionally, as the alkanes have a
molecular mass distribution, so to have the alcohols. Therefore the resultant
mixture contains both alkanes and alcohols, both with a molecular mass
distribution. A need therefore exists to separate these alkanes and alcohols.

Due to the molecular mass distribution of
the alkanes and alcohols, cross-over melting and boiling points prevent the
implementation of traditional processes. Advanced separation techniques are
thus required to achieve the separation. While extractive distillation has been
shown to be successful, high temperatures and low pressures are required and
the product may contain organic solvent residue. Supercritical CO₂
fractionation is a possible alternative separation technique that utilizes a
green solvent at mild operating temperatures.

Binary phase equilibria of supercritical CO₂
with n-alkanes and 1-alcohols shows that (1) for both n‑alkanes and
1-alcohols an increase in molecular mass leads to an increase in phase
transition pressure and (2) the alkanes in the range n-decane to n-hexadecane
are more soluble in supercritical CO₂ than alcohols in the range
1-decanol to 1-hexadecanol. From a solubility point separation may thus be
possible. However, the binary phase behaviour analysis does not consider
molecular interactions between the 1-alcohols and the n-alkanes and the
influence of these interactions may have on the phase behaviour.

Previous studies have shown that mixtures
of CO₂ + n-alkanes + 1-alcohols display interesting phase behaviour
such as a liquid-liquid-hole, especially at temperatures close to the critical
temperature of CO₂ [1]. However, to date a systematic study on the phase
behaviour of CO₂ + n-alkanes + 1-alcohols and the effect of n-alkane
– 1-alcohol interactions have not been conducted. Additionally, data in the
industrially relevant temperature range (308 to 359 K) are limited.

To investigate the effect of the n-alkane –
1-alcohol interaction on the thermodynamic behaviour of the CO₂ +
n-alkane + 1-alcohol system, high pressure bubble and dew point measurements
and high pressure phase vapour-liquid measurements were conducted for a
selection of CO₂ + n-alkanes + 1-alcohols. In general, to
investigate the pinch in the process, the lower alcohols (1-decanol and
1-dodecanol) are considered with higher n-alkanes.

High pressure bubble and dew point
measurements of the systems CO₂ + n-dodecane + 1-decanol [2], CO₂
+ n-tetradecane + 1-decanol [3] and CO₂ + n-tetradecane +1-dodecanol
[4] were conducted on a high pressure view cell. For all three of these systems
it was shown that at low 1-alcohol concentrations, co-solvency occurs, thus at
low 1-alcohol concentrations, the mixture of n-alkane + 1-alcohol is more
soluble in CO₂ than the pure n-alkane. For example Figure 4 shows
co-solvency in the system CO₂ + n‑tetradecane + 1-dodecanol.

Figure 1. Comparison of the phase
behaviour of the CO₂ + Mixture A (50% n-tetradecane + 50%
1-dodecanol) and CO₂ [4] + Mixture B (75% n-tetradecane + 25%
1-dodecanol) [4] systems with that of CO₂ + n-tetradecane [5] and CO₂
+ 1-dodecanol [5,6] at (a) 313 K and (b) 353 K

The effect of the co-solvency is also clear
when considering the phase equilibria measurements and the resultant Gibbs
phase diagram plots. The results show that for both the CO₂ +
n-dodecane + 1-decanol [2] and CO₂ + n-tetradecane + 1-decanol [3]
system close to the total solubility pressure, two regions of immiscibility
occur (See Figure 2). The split of the regions of immiscibility are believed to
be a result of the n-alkane – 1-alcohol molecular interaction.

Figure 2: Ternary Gibbs phase diagram
of the system CO₂ + n-tetradecane + 1-decanol at 328 k and (a) 8.0
and (b) 8.4 MPa [3]

Additionally when considering the relative
solubility of n-tetradecane to 1-decanol it is seen that at high 1-alcohol
concentrations the n-alkane will be enriched in the extract while at low
1-alcohol concentrations, the 1-alcohol will be enriched (See Figure 3). Similar
results are seen for other systems. A pinch point therefore exists, similar to
an azeotrope. Fortunately, as seen in Figure 3, the pinch is temperature and
pressure sensitive and the pinch point can be overcome using a two stage
separation process, similar to pressure swing azeotropic distillation.

Figure 3: Relative solubility of
n-tetradecane – 1-dodecanol separation using high pressure CO₂ as
separation medium as a function of pressure and compositions (g n-tetradecane
per g (n-tetradecane + 1-decanol)) at various temperatures [3]

The results from this investigation have
shown that binary data alone cannot provide sufficient information on the
thermodynamic behaviour of a CO₂ + n-alkane + 1-alcohol system to
allow for the design of a separation process. The molecular interactions
between the n-alkane and 1-alcohol are significant and result in a
temperature/pressure dependent azeotrope.

The current analysis has been conducted on
selected CO₂ + n-alkane + 1-alcohol systems. All these systems have
a varying degree of co-solvency. It therefore appears as if there is a range
where the n‑alkane + 1-alcohol solute-solute interactions results in
co-solvency. Measurement of further ternary data is recommended to clearly
define the combinations of n-alkanes and 1‑alcohols where co-solvency is
present and the conditions where co-solvency occurs.

Additionally, to date thermodynamic models
struggle to model these type of systems. Further work will thus also focus on
improvement of models to provide a better predict or correlate the data,
especially representation of the composition of the co-existing phases.

References

[1]   Patton C. L., Kisler S. H., Luks K. D., Multiphase Equilibrium
Behavior of a Mixture of Carbon Dioxide, 1-Decanol, and n-Tetradecane, in:
Supercritical Fluid Engineering Science, American Chemical Society, 1992: pp.
55–65.

[2]   F.C. v. N. Fourie, High pressure phase behaviour of detergent
range alcohols and alkanes, PhD Thesis in Chemical Engineering, Stellenbosch
University, 2018.

[3]   M. Ferreira, Phase Equilibria & Thermodynamic Modelling of
the Ternary System CO₂ + 1-Decanol + n-Tetradecane, PhD Thesis in
Chemical Engineering, Stellenbosch University, 2018.

[4]   S.P. Nortje, C.. Schwarz, Experimental Measurements and
Thermodynamic Modelling of the Phase Behaviour of  n-Tetradecane + 1-Dodecanol
in Supercritical Fluids, in: 15th European Meeting on Supercritical Fluids,
2016.

[5]   G.J.K. Bonthuys, Separation of 1-dodecanol and n-tetradecane
through supercritical extraction, Masters Thesis in Chemical Engineering,
Stellenbosch University, 2008.

[6]   A. Kordikowski, G.M. Schneider, Fluid phase equilibria of
binary and ternary mixtures of supercritical carbon dioxide with low-volatility
organic substances up to 100 MPa and 393 K: c, Fluid Phase Equilibria. 90
(1993) 149–162.