Comparison between the Performances of a Well-Stirred Slurry
Reactor and a Spray Loop Reactor for the Alkylation of p-Cresol
with Isobutene
Elio Santacesaria,* Martino Di Serio, Riccardo Tesser, and Francesco Cammarota
Universita ` di Napoli - Dipartimento di Chimica-Via Cinthia - Complesso Universitario di Monte S. Angelo,
80126 - Napoli - Italy
The kinetics of the alkylation of p-cresol with isobutene has been studied, first of all, in a well-
stirred slurry reactor. Preliminary runs of CO
2
absorption in water and of isobutene absorption
in p-cresol, in the same reactor, have been made for estimating the gas-liquid mass-transfer
coefficient in an independent way. The kinetic model, developed in a previous work, has been
improved and the related parameters have been determined by mathematical regression analysis
of many kinetic runs. Different kinetic runs have then been performed also by using a spray
tower loop reactor (STLR) in similar physical conditions. These runs have been simulated in a
satisfactory way by using the same kinetic model and parameters, changing only mass-transfer
parameters for taking into account the different fluid dynamic conditions. The performances
obtained by the two different reactors are in some conditions comparable. This work has
demonstrated, therefore, the possibility and the convenience of extending the use of STLR reactors
in performing gas-liquid-solid reactions.
Introduction
The behavior of multiphase reactors is often strongly
affected by mass-transfer limitations, and different
types of reactor configurations (bubble columns, jet loop
reactors, spray tower loop reactors, and mechanically
agitated reactors) are employed, depending on the
characteristics of the reactions. Spray tower loop reac-
tors (STLR) are largely employed for very fast reactions,
such as neutralization reactions or pollutant abatement
by absorption from gaseous mixtures. One of the main
applications for spray contactors, for example, is in gas
stream desulfurization.
1
It is important to point out that in STLR reactors the
liquid is the dispersed phase instead of the gaseous one,
as it occurs in the other mentioned reactors. The role
of the mass-transfer limitations in these conditions have
been poorly investigated in the literature.
2-7
In a previous work
6
we have shown that a STLR
reactor can be conveniently used for some gas-liquid
reactions provided that the solubility of the gaseous
reagent is high enough. In particular we have studied
the ethoxylation of different organic substrates,
8-11
and
we concluded that, for moderately fast reactions (with
Hatta no. = 1), the use of a STLR reactor can be more
convenient than one of the other reactors, such as, for
example bubble columns or well-stirred reactors. As a
matter of fact, we have found that the gas-liquid mass-
transfer rate in a STLR reactor is very high when a
suitable spray nozzle is used and the behavior of the
reactor can be interpreted with a mathematical model
that assumes the produced droplets are internally well-
mixed.
6
Moreover, we have evidence that, in some cases,
the spray tower loop reactors can give better perfor-
mances of the mechanically agitated reactors with the
advantage of the absence of any rotating device and with
the possibility of operating easily also at high pressure.
A further positive aspect of the STLR reactor is the
separation existing between the mass-transfer zone and
the reaction one, inside the reactor, in the presence of
a moderately fast gas-liquid reaction, and this is an
element of greater flexibility.
In the present work, we have studied the possibility
of extending the use of the STLR reactor also to gas-
liquid-solid reactions. No paper has been published
exploring this possibility. We will consider here and
discuss all the possible constraints in the use of a STLR
reactor in performing gas-liquid-solid reactions, by
examining also the possible advantages and drawbacks
of the proposed technique. To do this, we have chosen
for our investigation the alkylation of p-cresol with
isobutene as a test reaction. This reaction, catalyzed by
acid exchange resins, occurs in two consecutive steps,
12,13
the first giving 2-tert-butyl-p-cresol (M) and the second
2,6 di-tert-butyl-p-cresol (BHT). Isobutene can also give
place to oligomers as side products and, normally, only
dimerization (D) and trimerization (T) occur.
In the present work, the reaction has been performed
in both a well-mixed gas-liquid-solid reactor and in a
spray tower loop reactor. The kinetic model developed
in a previous work
12
has been improved in view of a
possible scale-up, and more reliable kinetic parameters
have been determined by mathematical regression
analysis on many more kinetic runs performed in the
slurry reactor. The same kinetic model has also been
successfully used for simulating the kinetic runs per-
formed in the STLR reactor. The performances of the
two reactors have then been compared with particular
attention devoted to the effect of mass-transfer limita-
tions. A preliminary fluid dynamic characterization has
been performed for both of the reactors by studying the
physical absorption of CO
2
in water.
Experimental Section
Apparatus, Methods, and Reagents. Two different
reactors have been used for studying the kinetics and
* To whom correspondence must be addressed. E-mail:
santacesaria@chemistry.unina.it.
9473 Ind. Eng. Chem. Res. 2005, 44, 9473-9481
10.1021/ie050221j CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/06/2005