Equilibrium Reactions in sc HFC32
J. Phys. Chem. B, Vol. 108, No. 15, 2004 4923
methanol around the phthalic anhydride molecules exceeding
the average bulk methanol concentration. Using estimated local
densities and local compositions, it was shown that the observed
increase in rate constant could be accounted for.
The esterification of acetic acid and ethanol in compressed
CO2 was studied at 333 K over a range of pressures10 and found
to reach equilibrium in 3 h under supercritical conditions and
in 6 h under subcritical conditions. Results showed that the
apparent equilibrium constant (KX) increased as the pressure
increased in the gas phase and reached a maximum in the critical
region where the system becomes one phase. After the critical
point was passed, KX decreased as the pressure increased. The
observed increase in KX at low pressures was explained, with
respect to the degree of clustering/local density enhancement
at these pressures. At high pressures, where clustering is
insignificant, KX for the reaction was similar to that of the
reaction in the absence of CO2.
In the present study, we report the first equilibrium reactions
in supercritical difluoromethane (sc HFC32) (critical temperature
of Tc ) 78.1 °C; critical pressure of pc ) 57.8 bar). This solvent
has previously been shown to be useful, because of its high
dielectric constant11 and large polarizability change with pres-
sure.12,13 A recent study has also shown that HFC 32 is a strong
hydrogen-bond donor close to the critical pressure14 and has
demonstrated high solubility for polar aromatic solutes.15
Figure 1. Comparison of measured and calculated conversion for the
esterification reaction.
capacitance of the solution. For this to occur, the products also
must remain in solution as the reaction proceeds.
With these limitations in mind, the esterification reagents used
were benzoic acid, 1-butanol, and p-Tos acid as the catalyst.
Benzoic acid was observed to have a high solubility in sc HFC
32 (at 363 K and 100 bar, the solubility was 0.542 mol dm-3),
which is considerably in excess of the concentrations used in
these experiments (0.04 mol dm-3). The 1-butanol and p-Tos
acid were readily soluble in the solvents under the reaction
conditions.
The dielectrometry technique was used to monitor the solution
concentration of reagents in liquid dichloroethane (DCE),
because of its similar dielectric properties to HFC 32 (for DCE,
ꢀ ) 10.37 at 278 K; for HFC 32, ꢀ ) 8.99 at 363 K and 200
bar). The results were compared to those obtained using GC-
MS. The reaction was followed over a 2-h period, using
dielectrometry and GC-MS. Dielectrometry measurements were
measured in situ, and for the GC-MS measurements, 2-mL
aliquots of the reaction solution were taken. The actual reaction
conversion was obtained from the GC-MS results at specific
time intervals. The conversion value from the dielectrometry
measurements was calculated from knowledge of the final
conversion of the reaction, from GC-MS, and the capacitance
value at each time interval. The measured (GC-MS) and
calculated (dielectrometry) conversions for this esterification
reaction are shown in Figure 1. This figure shows that the
dielectometry readings correlate well with those from GC-MS,
signifying that the dielectrometry technique is suitable for
following the progress of this reaction.
Experimental Procedure
The high-pressure apparatus used in this study was the same
as that described previously.15 HFC 32 (Ineos Fluor, 99.99%)
and CO2 (BOC, 99.9%) were used as received. The esterification
of benzoic acid (Scientific and Chemical Supplies, Ltd., 98%)
with 1-butanol (Fisher, 98%), using p-toluenesulfonic (p-Tos)
acid as the catalyst, was performed using reagent concentrations
of 0.04 mol dm-3. The acid-catalyzed aldol condensation of
cyclohexanone (Aldrich, 98%) was performed using p-Tos acid
as the catalyst. All experiments were conducted at 363 K and
in the pressure range of 60-220 bar. The reactants were placed
into the doser and the acid catalyst was placed in the reaction
vessel. This was then heated to the desired temperature and
subsequently pressurized, forcing the reactants from the doser
into the reaction vessel. The total time taken for the vessel to
reach the reaction pressure was <1 min. The system was left
to react for the desired time and where applicable capacitance
measurements were taken. The products were trapped by
depressurization into a larger volume autoclave, which also
stopped the reaction and the products were analyzed by gas
chromatography-mass spectroscopy (GC-MS) (Perkin-
Elmer).
Results and Discussion
The esterification reaction was then performed in sc HFC 32
as the reaction solvent. It is assumed that this esterification
reaction follows the mechanism shown in Scheme 1. Using an
optical cell, it was visually observed that the reaction was
homogeneous under all conditions studied in this work. The
dielectrometry technique was used to follow the progress of
the benzoic acid esterification reaction as a function of time,
and the results are shown in Figure 2 at selected pressures. The
uncertainty of the conversion is no greater than (2%. Figure 2
shows that this dielectrometry technique is useful for following
equilibrium reactions, because the point at which the reaction
reaches equilibrium can be easily observed. This figure shows
that the reaction has reached equilibrium in 2 h or less, under
the conditions studied. The reaction was performed for up to 6
h without any appreciable change in yield. Therefore, in the
Esterification Reaction. It has previously been shown that
a dielectrometry technique can be used to determine the
solubility of solutes in supercritical fluids.15 Therefore, it is
logical that the same technique can be used to follow the changes
in solution concentration during a reaction, because it is a rapid,
in situ method for measuring capacitance changes of reaction
systems. Dielectrometry is not affected by solution concentra-
tion, high pressure, and turbidity restraints that are present with
spectroscopic techniques. However, note that there are some
limitations: (i) it can only be used with reagents that are readily
soluble, because accurate capacitance measurements can only
be obtained in homogeneous solutions, and (ii) a large difference
in dielectric constant between the reactants and products is
necessary, so that product formation produces a change in the