Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
RSC Advances
Paper
Our strategy has been to study the differences in perfor- pipes of the same alloy and 1/800 (3.18 mm) outer diameter. This
mance using several reactor geometries while keeping the rest conguration exploits the buoyancy of the hot stream to
of the reaction parameters as near constant as possible.
enhance the mixing. All the reactants should be efficiently
The equipment and the experimental procedure have already mixed in the middle section of the reactor (l1 ¼ 6.3 cm), aer
been described in detail.6,7 A simplied scheme is shown in which they ow upwards along the outer section of the reactor
Fig. 1. Reactions were carried out in a continuous fashion. An (l2 ¼ 8.5 cm) and along a 1/400 pipe (l3 ¼ 8.5 cm) until they meet
aqueous H2O2 solution was passed through a coiled pre-heater the NaOH quench.
at supercritical temperature long enough to achieve total
Residence time for this reactor cannot be calculated with
decomposition of the H2O2 solution to form a homogeneous high accuracy since the mixing area is located somewhat
mixture of O2 and supercritical water. Unless otherwise stated, imprecisely in the middle. Nevertheless, considering the
the concentration of H2O2 in the feed solution was 2% by w/w dimensions, the residence time should lie between 2.3 and 3.3 s
and the ow rate was 8 mL minꢀ1. The organic ow rate was at 380 C and between 7.4 and 11.9 s at 330 C, using a total
0.06 mL minꢀ1, that is 0.5% w/w in the reactor. A mixture of ambient ow rate of 12 mL minꢀ1 and the density of pure
CuBr2 and NH4Br in aqueous solution, 1.16 and 13.1 mM water.24 As an approximation, we take the average values (i.e. 2.8
ꢁ
ꢁ
respectively, was fed as catalyst solution at 4 mL minꢀ1, thus s at 380 C and 9.7 s at 330 C).
ꢁ
ꢁ
giving a catalyst concentration in the reactor, expressed as the
The second conguration (TB, shown Fig. 2b) is based on
concentration of each species,7 Cu/NH4/Br ¼ 0.39/4.42/5.2 mM. a tubular design and it is the one used in all of our previous
Downstream of the reactor, the liquor was quenched with work.3–7 All the reactants and solvent are mixed at a 1/400 mixing
a solution of 1 M NaOH, to neutralize the CO2 originating from cross at the top of the reactor. The catalyst and organic are
burn and converting it to carbonate which is thus retained in delivered from opposite sides via 1/1600 Hastelloy pipes of 0.57
solution for offline analysis. NaOH also prevents precipitation mm inner diameter. The reactive mixture then ows downwards
of TA and avoids consequent blockages. The NaOH solution through a tubular reactor, at the bottom of which it is rapidly
ow rate was 3.5 mL minꢀ1. All the chemicals were purchased cooled by the NaOH quench solution. The reactor is constructed
from Aldrich Ltd and used without further purication.
of Hastelloy C276 tubing, 1/400 outer diameter, and 4.6 mm
Analysis of the products was carried out by HPLC. A Waters inner diameter. The length of the reactor from the mixing to the
Xterra reverse phase C18 column, maintained at 37 ꢁC, was used quench point, l4, was 34 cm. Based on the density of pure
(ow rate 0.7 mL minꢀ1, run time 15 min; UV detection at 230 water,24 a total ow rate of 12 mL minꢀ1 will give calculated
2ꢀ
ꢁ
ꢁ
nm). The CO2 generated was quantied by measuring CO3
residence times of 5.8 s at 380 C and 19.2 s at 330 C.
The importance of the TB design is that it allows the mixing
efficiency to be varied without altering the other aspects of the
concentration by titration of the sample with 0.2 N HCl. Details
of these procedures have been given elsewhere.6,7
This study is based on the two basic reactor congurations reactor, by changing the distance by which the catalyst and pX
shown in Fig. 2.
feed pipes protrude into the cross-piece.
The Opposed Flow reactor, OF (shown in Fig. 2a) was
inspired by the nozzle-like reactor developed by Lester et al.19,23
The pX pipe is concentric with the catalyst pipe and both point
upwards. These upward owing streams meet the downward
owing stream of heated H2O + O2. All the outer pipes are built
using Hastelloy C276 tubing, 1/400 (6.35 mm) outer diameter,
and 4.6 mm inner diameter. The inner pipes are built using
Results and discussion
The oxidation of TA in acetic acid is known to follow a sequence
of partially oxidized intermediates:7 para-xylene (pX) / 4-
methyl-benzaldehyde (p-tolualdehyde, pTOL) / p-toluic acid
(PTA) / 4-carboxybenzaldehyde (4CBA) / TA. Benzoic acid
(BA) and CO2 result from both decarboxylation and combustion
(burn) mechanisms.6 Cu-based catalysts have previously been
shown to be particularly active.7 Unless otherwise stated,
a mixture of CuBr2 and NH4Br was used as catalyst under
unsaturated concentrations, i.e. at a concentration lower than
that needed to give the maximum yield; such so-called
“stressed” reaction conditions are more sensitive to the
different reaction paramenters.7 Yields of product, by-products
and intermediates were calculated as before.6,7
The opposed ow reactor, (OF)
Fig. 1 Simplified scheme of the continuous system used in this work.
O2 is generated from aqueous H2O2 by high temperature decompo- The rst study varied residence time (by varying ow rate) for
sition in a pre-heater, pH.2 Pure p-xylene, an aqueous solution of the
catalyst, and a mixture of O2 + H2O are pumped separately into the
continuous reactor by means of HPLC pumps. After the reactor, the
mixture is quenched with a cold solution of NaOH that neutralizes any
a given reactor conguration and constant feed composition.
The results are gathered in Table 1. At 330 ꢁC, (entry 1), one
observes a wide distribution of intermediates indicating that as
previously observed,7 the catalyst is not active enough at these
CO2 to form carbonate. The detailed geometry of the reactor itself is
changed between experiments.
concentrations to complete the reaction. By contrast entries 2–4
11290 | RSC Adv., 2016, 6, 11289–11294
This journal is © The Royal Society of Chemistry 2016