Equipo 4 PDF
Equipo 4 PDF
Equipo 4 PDF
Membrane reactors have found utility in a broad range of applications including biochemical,
chemical, environmental, and petrochemical systems. The variety of membrane separation
processes, the novel characteristics of membrane structures, and the geometrical advantages
offered by the membrane modules have been employed to enhance and assist reaction schemes
to attain higher performance levels compared to conventional approaches. In these, membranes
perform a wide variety of functions, often more than one function in a given context. An
understanding of these various membrane functions will be quite useful in future development
and commercialization of membrane reactors. This overview develops a functional perspective
for membranes in a variety of reaction processes. Various functions of the membranes in a reactor
can be categorized according to the essential role of the membranes. They can be employed to
introduce/separate/purify reactant(s) and products, to provide the surface for reactions, to provide
a structure for the reaction medium, or to retain specific catalysts. Within these broad contexts,
the membranes can be catalytic/noncatalytic, polymeric/inorganic, and ionic/nonionic and have
different physical/chemical structures and geometries. The functions of the membrane in a
reaction can be enhanced or increased also by the use of multiple membrane-based schemes.
This overview develops a perspective of each membrane function in a reactor to facilitate a better
appreciation of their role in the improvement of overall process performance.
to be de facto equivalent to that with a membrane inside membrane reactor whereby the product H2 is withdrawn
a reactor. Therefore, such recycle membrane reactors through the palladium membrane to shift the reaction
will also be included in the following treatment: pri-
mary emphasis will however be on systems where a C2H5OH T CH3CHO + H2 (3a)
membrane (or two) is physically located in the reactor.
One must recognize the major advantages of these to the right showed that the deleterious effect of the side
different arrangements: reaction
(1) The mixing conditions and the flow velocities (and
therefore the extent of consequent concentration polar- C2H5OH + CH3CHO T CH3COOC2H5 + H2 (3b)
ization in membrane devices involving liquid-phase
systems) can be maintained at different levels in the can be drastically reduced, provided the reaction (3a)
reactor and the membrane separator if recycle mem- is shifted to the right via the Cu/SiO2-aq catalyst and
brane reactors are employed; conditions can be opti- H2 removal by the membrane.
mized for each. The reactor may require long residence (c) In consecutive catalytic reactions,
times whereas the membrane device may need a short AfB (4a)
residence time.
(2) Building a reactor with a membrane in it or using BfC (4b)
a membrane device as the reactor can sometimes be very
demanding on the membrane, especially for higher where B is the desired intermediate product, if the rate
temperature systems. The recycle membrane reactor constant for reaction 4b is significantly larger than the
allows the reactor and the membrane unit to operate rate constant for reaction 4a, it is difficult to achieve a
at two different temperatures by using heat exchangers high selectivity to B using a conventional packed bed,
in between. plug flow reactor. By using an inert sweep gas on the
(3) Recycle membrane reactors allow use of existing outside of a permeable tube having the catalysts and
equipment, namely, a separate reactor and a separate the reaction taking place inside the tube, the intermedi-
membrane device. ate product B may be selectively removed from the
(4) For fast reactions, the membrane in a reactor is reaction zone, leading to increased selectivity.14 Removal
likely to be a more desirable configuration. of the intermediate products (methanol/formaldehyde)
2.1. Separation of Products from the Reaction via a membrane in the partial oxidation of methane is
Mixture. Separation of products from the reaction an example; this strategy will prevent further oxidation
mixture is one of the most common functions of a of these products to CO and CO2.
membrane in a reactor. The separation may be purifica- (d) In fermentation processes, one of the products may
tion, enrichment, or concentration. Consider the follow- be inhibitory to the fermentation process. Removal of
ing elementary reversible reaction (see Figure 1): the product from the fermentation broth via a mem-
brane can substantially reduce product inhibition and
A+BTC+D (1) increase volumetric productivity of the fermentor.15,16
Further, one can use higher concentrations of the
where D is a product needed to be removed via the substrate in the feed (e.g., glucose for ethanol fermenta-
membrane to the permeate side. The separation process tion) since the product is being removed as it is being
employed may produce a permeate side stream where formed.17
the mole fraction of D is much higher than that in the The separation of a reaction product(s) (C or D or
reactor side. In the case of species D being H2 and a both) can be implemented using a variety of membrane
palladium membrane, pure H2 is obtained in the perme- processes. The nature of the membrane process is
ate side. The H2 partial pressure on the permeate side obviously influenced by the phase of the reaction
will be lower than the partial pressure of H2 on the feed medium exposed to the membrane and the desired
side. If species C also permeates to some extent through phase of the permeated product stream. Examples of
the membrane, the permeate stream is enriched in D such processes will be provided under two categories,
vis-à-vis the product species C: permeation of D leads namely, (1) liquid reaction medium/liquid feed phase
to partial purification of the product C in the reactor and (2) gaseous reactions/gaseous feed phase.
outlet stream. 2.1.1. Separation from a Liquid Reaction Mix-
Removal of D via the separation function of the ture. We will briefly mention and/or illustrate the use
membrane has the following effects on reaction (1) and of the following membrane separation processes for
the reactor performance: removing products from the liquid reaction medium:
(a) The equilibrium condition indicated in the revers- reverse osmosis, nanofiltration, ultrafiltration, pervapo-
ible reaction (1) is shifted to the right, i.e., leading to ration, gas membranes, electrodialysis, and liquid mem-
higher equilibrium conversion of A and B to C and D. branes.
(b) If there is an undesirable side reaction as shown 2.1.1.1. Reverse Osmosis. Vasudevan et al.18 have
below, described a membrane sandwich reactor in which the
Saccharomyces cerevisae (ATCC 4126) cells were ef-
B+DTE (2) fectively placed between an ultrafiltration (UF) mem-
brane and a reverse osmosis (RO) membrane; the
taking place in the reactor (see Figure 1), the separation reactor was fed with a solution of glucose at a high
of product D from the reaction mixture reduces the loss pressure from the UF membrane side and the product
of reactant B to the side reaction, increasing the solution was forced out through the RO membrane. The
selectivity of conversion to product C (or D) (modeled product solution concentration progressively increased
by Whu et al.12 for nanofiltration-aided liquid-phase in ethanol; the glucose in the feed solution was ef-
organic synthesis). An experimental study by Raich and fectively rejected by the RO membrane. The reactor
Foley13 of ethanol dehydrogenation in a palladium structure shown in Figure 3 has a microfiltration
3718 Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999
Figure 8. Apparatus for the resolution of (R,S)-Phe-O-iPr-HCl with subtilisin, using a hollow fiber SLM module (reprinted with
permission from Ricks et al.26 Copyright 1992, American Chemical Society).
brane, respectively, more water is dissociated. Thus, 2.1.2.1. Liquid Membrane. One of the earliest
separation of products, H+ and OH-, through the studies by Ollis et al.27 involved acetaldehyde synthesis
membranes is essential to continue the water-splitting in a multiphase catalytic liquid membrane-based reac-
reaction. tor. The feed gaseous mixture of O2 and C2H4 as the
2.1.1.7. Liquid Membranes. Enantioselective hy- inner gas phase surrounded by an aqueous catalytic
drolysis of (R,S)-phenylalanine isopropyl esters by the liquid membrane layer was prepared by dispersing the
enzyme subtilisin Carlsberg in solution in a reactor was gas as bubbles into an aqueous liquid membrane
shown by Ricks et al.26 to selectively convert the (S)- reservoir containing the palladium-based catalyst PdCl2
ester to (S)-Phe-COO- while leaving the (R)-ester and the oxidizing agent CuCl2. The individual gas
essentially unchanged. Part of the enzymatic solution bubbles surrounded by the aqueous liquid membrane
was circulated on the shell side of a hollow fiber module layer then spontaneously rose through a solvent layer
in which a liquid membrane of N,N-diethyldodecana- kept above the aqueous catalyst reservoir. Any reaction
mide in dodecane is immobilized as a supported liquid product was extracted into the outer solvent layer. The
membrane in the pores of hydrophobic Celgard polypro- overall reaction and the continuous extraction of the
pylene hollow fibers. The (R)-ester is permeable through product acetaldehyde into a solvent phase (e.g., ethyl
this liquid membrane into an acidic strip solution where acetate, n-butanol, etc.) can be indicated by
it gets protonated and cannot partition back into the
feed phase. The (S)-Phe-COO- is essentially imperme- O2 (w) + C2H4 (w) f CH3CHO (w) f
able through the liquid membrane and is recovered from
CH3CHO (s) (9)
the feed solution. A part of the feed solution is recircu-
lated through the reactor and the membrane separator
in the recycle mode. If the (R)-ester accumulates in the The product acetaldehyde is recovered from the solvent
reaction media, it can inhibit the rate of conversion of in a separate flash drum, while the solvent phase is
the (S)-ester, which can also permeate through the recycled back to the reactor to form the outer continuous
membrane. The retention time of any ester fed to the solvent phase in the double-emulsion liquid membrane
reaction vessel is adjusted such that hydrolysis is employed in the reactor. The liquid membrane in this
essentially completed (12 min) before the solution is case allows not only product separation but also the
introduced in the membrane module in the recirculation segregation of the soluble catalyst in the aqueous
mode (Figure 8). This is an example of an unconverted membrane phase (function 2.6) in addition to being the
reactant being removed from the reaction mass as if it actual site where the reaction is taking place (function
were an inhibitory product. Further, a reaction (in this 2.8). The product separation here does not lead to a pure
case protonation) is carried out in the permeate phase product for which an additional step (flash) is necessary.
(Figure 1), 2.1.2.2. Knudsen Diffusion. Since the mid 1980s, a
large number of studies have been carried out using
D+FfG (8) reactants in the gas phase and an inorganic membrane
through which one or more of the gaseous products
to maintain a large driving force for the permeation of (usually H2) is withdrawn. The membranes used often
D. were microporous/mesoporous, e.g., γ-alumina, Vycor
2.1.2. Separation from a Gaseous Reaction Me- glass, etc., with or without a catalyst deposited in the
dium. The membrane processes employed include those pores. Tubular noncatalytic membranes were also used
using liquid membranes, Knudsen diffusion, gas per- packed with catalyst particles.
meation, molecular sieving, etc. We will briefly identify Sun and Khang28 have compared the performances
some typical examples to illustrate product removal by of a catalytic membrane reactor, an inert membrane
membranes in each case. reactor, and an ordinary reactor without any membrane
Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999 3721
Figure 10. Two reactions coupled through a gas membrane; membrane-assisted synthesis (reprinted from Van Eikeren et al.44 with
permission).
reactor on the left-hand side. The effect of this separa- water (having high pH 9-11) are partitioned through
tion on the reaction system is generally quite different the silicone membrane into the biological reaction
from that of a reaction product from the reaction medium. Reactors operated over 3000 h show very high
mixture. The purification may lead to pure A being reductions of the pollutants without any form of pre-
introduced into the reactor; a direct effect of this is treatment, pH adjustment, or dilution of the wastewa-
prevention of dilution of the reaction mixture. It can also ter. The membrane essentially isolates the bioreactor
lead to rejection of a class of compounds by the mem- environment from the vagaries of the industrial waste-
brane while reactant species (one or a class) may be water properties as the pollutants get extracted out into
introduced by the membrane preferentially into the the bioreaction medium for degradation by the ap-
reactor from the feed stream; the species rejected can propriate microorganisms. Successful pilot plant studies
inhibit the reaction. An additional possibility involves have been conducted using this technique.
simultaneous operation of two different reactions on two 2.2.3. Purify Organic Compound from a Synthe-
sides of the membrane wherein the products of one sis Medium. To prepare a concentrated aqueous solu-
reaction feeds the other and vice versa; the latter could tion of diltiazem malate by reacting diltiazem (sparingly
be in a coupled mode as well (to be explained later). soluble in water) with malic acid, a liquid membrane
2.2.1. Pure O2 from Air. Mixed conducting dense was utilized to recover diltiazem from an aqueous
ceramic membranes (see sections 2.3 and 2.9 as well) reaction mixture containing diltiazem, NaCl, NaHCO3,
allow O2 transport from air to a lower O2 partial etc. A contained liquid membrane of decanol was
pressure side without allowing N2 to be transported utilized by Basu and Sirkar43 to extract diltiazem into
through the membrane. When such a membrane is used the reaction zone where it reacted with L-malic acid in
with air on one side and CH4 on the other side, partial water to form diltiazem malate (in solution). When a
oxidation of CH4 to syngas can be carried out at a high very high concentration of L-malic acid is maintained,
temperature (∼800 °C) without contaminating the reac- the aqueous concentration of diltiazem malate achieved
tion gas mixture with N2. Further, the need for an O2 could be 3 orders of magnitude higher than the very low
plant is eliminated,40 improving the economics consider- concentration of diltiazem in the feed solution. The
ably. When a similar dense ceramic membrane in a membrane not only facilitated production of a highly
solid-electrolyte-cell reactor (see section 2.9) is used, concentrated and purified solution of diltiazem malate
HCN is produced in a tubular reactor fed with a NH3 + but also avoided two steps of solvent extraction and back
CH4 mixture as O2 permeates through the membrane41 extraction which was very problematic due to the orders
as an anion O2- into the reaction zone; in this case air of magnitude difference in the two aqueous phase flow
is fed on the outside of the tubular reactor to produce rates, namely, the very high flow rate of a very dilute
oxygen throughout the reactor length (function 2.3). Had feed solution and the very concentrated reactor effluent
air been used directly, the auxiliary byproduct of H2 + having a low flow rate.
CO, a high-quality fuel stream, would have been diluted 2.2.4. Coupling of Two Chemical Reactions. In
with N2. enzymatic methods for the production of pure enanti-
2.2.2. Purify Organic Pollutants from Wastewa- omers from achiral precursors using, say, dehydroge-
ter for Biodegradation. Point-source industrial waste- nase enzymes, specialized and costly reagents, e.g.,
waters containing a variety of priority pollutants were nicotinamide cofactors (NAD+ or NADH), are required.
often considered recalcitrant from a biodegradation A process for efficient and continuous regeneration of
perspective. These industrial wastewaters are fre- these nicotinamide cofactors by a chemical reaction
quently released from organic synthesis and contain isolated from the main reaction by a membrane parti-
high salts, extreme pHs, and residual catalysts, all of tion is illustrated in Figure 10.44 The membrane em-
which are either singly or jointly very harmful to the ployed is a gas membrane (see 2.1.1.5) utilizing a
growth of microbial cultures used for biodegradation. microporous hydrophobic polypropylene membrane whose
Brookes and Livingston42 have employed a silicone pores (∼0.03 µm) are gas-filled. In the main reaction
capillary membrane-based device to extract organic compartment on the left side, D-lactic acid is produced
priority pollutants from these demanding wastewaters. in an aqueous solution by catalyzing the reduction of
The biological reaction medium is circulated between a pyruvic acid using D-lactic acid dehydrogenase. The
bioreactor and one side of this membrane device. On NAD+ produced by this reaction reacts with ethanol to
the other side of the silicone capillary membranes flows regenerate NADH required for the main reaction. In an
the wastewater. Priority organic pollutants, e.g., aniline, adjacent compartment, sodium borohydride reduction
4-chloroaniline, 3,4-dichloroaniline, etc., from the waste- of acetaldehyde leads to ethanol which, being volatile,
Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999 3723
enters the main reaction chamber whereas acetaldehyde to study the oxidative dehydrogenation of C2H6 to C2H4
produced in the regeneration of NADH in the main at 600 °C. Air was introduced via permeation through-
reaction chamber enters the adjacent chamber (through out the length of the reactor from the outside of the tube.
the membrane). Thus, the membrane allows separation Controlling the ratio of C2H6 to O2 was found to be
of the volatile species from the rest of the nonvolatile crucial to selectivity (with respect to CO2, CO, etc.) and
reaction mixtures in both compartments and feeds each conversion. At low-to-moderate C2H6 to O2 feed ratios
volatile species at a controlled rate into the adjacent (<1.5), the C2H4 yield in the membrane reactor exceeded
reaction chamber (function 2.3). that in a cofeed plug flow reactor (PFR) operated under
A weaker coupling between two reactions on two sides the same conditions by a factor of 3. At higher feed ratios
of a membrane (see Figure 1 for A + B T C + D and D (g2.5) the PFR performed almost as well as the mem-
+ F f G where the membrane controls the addition of brane reactor. In the membrane reactor, the reactant
reactant D to the permeate side reaction) was explored feed ratio (C2H6:O2) near the inlet to the reactor was
numerically by Itoh and Govind.45 On the basis of earlier quite high; as a result, the selectivity was very high.
proposals by Gryaznov and Itoh, they modeled the Continuous but slow addition of O2 further down the
catalytic dehydrogenation of 1-butene to butadiene in reactor ensured high C2H6 conversion as well, in the
the main packed bed reactor. The permeate side was membrane reactor.
being continuously fed with O2 in air which oxidizes the Lu et al.48 have carried out a modeling study of the
permeated H2 to produce water and liberate heat by an performance of a tubular reactor for oxidative coupling
exothermic reaction. This reaction is surface-catalyzed of methane to yield C2 hydrocarbons. Methane was fed
by the palladium membrane used to remove hydrogen at the reactor inlet along with inerts. O2 was introduced
from the main reactor. The heat so generated is con- into the reactor at different feed points along the reactor
ducted through the palladium membrane to the main length; in another configuration O2 was introduced
endothermic dehydrogenation reaction. The membrane continuously along the length via an O2-permeable
has many functions: separate product of main reaction, membrane acting as the reactor wall. In terms of
H2 (function 2.1); provide purified reactant (H2) to maximum C2 hydrocarbon yield, the membrane reactor
second reaction (function 2.2); add reactant H2 to second performed the best, the distributed feed reactors came
reaction in a controlled way (function 2.3); supply heat next (the larger the number of feed points, the better
to main endothermic reaction (function 2.10) and thereby the performance), and the cofeed reactor (CH4 and O2
control both reactions; and catalyze the second reaction together) performed the worst.
(function 2.7). To produce synthesis gas from CH4 via partial oxida-
2.3. Controlled Addition of a Reactant or Two tion, Balachandran et al.40 employed CH4 feed through
Reactants. Control of the reaction pathway is a major a membrane tube packed with a Rh-based reforming
concern in reaction engineering. Partial oxidation reac- catalyst. Air was supplied on the outside of the dense
tions of hydrocarbons are especially relevant here. In ceramic tube made of a La-Sr-Fe-Co-Zr-O system;
particular cases, possibilities of thermal runaway and the mixed conducting ceramic membrane tube allowed
catalyst poisoning do exist. In biodegradation processes pure O2 to be introduced along the reactor length at 850
for toxic organics, microorganism growth may be af- °C. CH4 conversion efficiency was >98%, CO selectivity
fected by inhibition from the toxic organics unless their was 90%, and H2 produced was twice that of CO.
concentrations are controlled. In an aerobic wastewater Zaspalis et al.49,50 have experimentally studied the
treatment process, high O2 utilization with minimum dehydrogenation of methanol in a microporous γ-Al2O3
waste to the atmosphere requires controlled but efficient membrane with methanol fed from one side and O2 from
introduction of O2 to the system. In processes using the other (Figure 1, reactants A and B introduced into
reactants having limited half-lives, e.g., ozonation of the pore from opposite sides). This arrangement avoided
wastewater or for water purification, efficient and direct contact between the two streams (i.e., an alcohol
localized introduction of O3 at a controlled rate can lead and an oxidant), thereby minimizing undesirable gas-
to higher O3 utilization. Using a membrane to introduce phase reactions. Further, the inner surface areas of the
a reactant or two in a controlled fashion in the reactor membranes were used. Zaspalis et al.50 used silver
can facilitate achievement of the desired reaction condi- particles deposited on the γ-Al2O3 membrane pore
tions. A number of examples provided below will il- surfaces; CH3OH was fed from one side and O2 from
lustrate the role of a membrane in such processes. the other so that activation of the Ag catalyst occurred
2.3.1. Gas-Phase Reactions. Gas-phase reactions simultaneously with the methanol conversion to form-
where reactants are introduced into the reaction zone aldehyde. This opposed flow mode of feeding two reac-
by membranes at a controlled rate(s) (see Figure 1, tants must be carried out in a controlled manner so as
purify reactant A from species F before addition) can not to decrease the selectivity for the desired product.
involve three types of membranes: (1) inert porous 2.3.2. Gas-Liquid Reactions. Catalytic reactions
membranes which provide no selectivity; (2) microporous between gas and liquid phases pumped concurrently
membranes with some selectivity; (3) nonporous/dense down a bed of catalyst particles in conventional reactors
membranes having much higher selectivity. The first encounter mass-transport limitations due to intrapar-
two types of membranes may employ catalysts in the ticle mass-transfer resistance, liquid film resistance,
pores; the nonporous membrane can be inherently liquid maldistribution, channeling, etc., apart from hot
catalytic. All three types of membranes have been spots and undesirable side reactions. To overcome these
studied with O2 as one of the reactants introduced in a problems, Cini and Harold51 have studied hydrogenation
controlled manner for partial oxidation or oxidative of R-methylstyrene (diluted in mesitylene) to cumene
dehydrogenation reactions. in a porous (6 µm) tubular γ-Al2O3 membrane impreg-
Tonkovich et al.46,47 employed 50 Å γ-Al2O3 (effective nated with a Pd catalyst on the pore surface area. H2
layer) or 200 Å R-Al2O3 membrane tubes packed with a was supplied on one side of the porous membrane and
magnesium oxide catalyst doped with samarium oxide the liquid reactant flowed on the other side of the
3724 Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999
Figure 12. Bifunctional nature of hydrophobic membranes. Shown above is the end view of the hollow fiber membranes with biocatalyst
particles surrounding them. (a) The membrane can be used to supply gas throughout the reactor volume while at the same time removing
carbon dioxide and hydrogen. (b) When a solvent is passed through the membrane lumen under correct pressure conditions, solvent
extraction can provide integrated product recovery while removing carbon dioxide (reprinted with permission from Frank and Sirkar.61
Copyright 1986, John Wiley & Sons).
water-filled substructure bounded by the membrane and L-aspartate-β-decarboxylase in the second stage,
skin impermeable to the enzymes. If the enzyme activity
is reduced substantially, the enzyme can be flushed out L-aspartate-β-decarboxylase
L-aspartic acid 98
easily and fresh enzyme loaded in the absence of the L-alanine + CO2 (13b)
organic phase. The organic flow is then restarted.
Stereoselective enzymatic hydrolysis of the undesired
Some enzymatic synthesis reactions are carried out
isomer from a racemic mixture of glycidic esters and its
in hollow fiber ultrafiltration membrane devices in what
removal in an aqueous buffer is carried out continu- is called the “perfusion reactor” mode. The enzymes or
ously. Here, the membrane provides reversible im- the whole cells are packed on the shell side with the
mobilization of the hydrolytic enzymes in the pores of shell side ports being closed off; substrate-containing
the hydrophilic asymmetric hollow fiber, immobilization feed is pumped through the tube side. The substrate
of the aqueous-organic interface on the fiber outside diffuses through the membrane pores to the shell side
diameter, reaction product extraction, and a very large and reacts with the enzymes/whole cells, and the
aqueous-organic interfacial area. smaller molecular weight products diffuse back to the
2.5. Segregation of the Catalyst (and Cofactor) tube side and are carried away. The enzymes or whole
in a Reactor. A membrane incorporated in a catalytic cells may also be kept in the tube side with both ends
closed off: the substrate-containing solution will then
reaction system can perform, among others, a number
flow into and out of the shell side. This latter mode is
of functions related to the catalyst. If the catalyst is
not commonly used. A review of membrane bioreactors
mobile in the reaction fluid, the membrane can prevent wherein the membranes segregate enzymes/whole cells
its escape from the system. If the catalyst is to be is available in Cheryan and Mehaia.6
immobilized with easy access to the reactants and Some enzyme-based reactions, however, require low-
convenient exit for the products, a porous/microporous molecular-weight coenzymes or cofactors in addition to
membrane structure may have the catalyst immobilized the main enzyme to carry out the overall enzymatic
on/within its structure (function 2.6). Alternately, the reaction. Typical examples of such coenzymes are nico-
membrane material itself may act as the catalyst tinamide adenine dinucleotide (NAD+), the reduced
(function 2.7). We focus here on cases where the catalyst form of NAD+, viz., NADH, NADPH (the reduced form
is mobile in the reaction fluid. Examples of such of NAD phosphate), etc. Figure 10 shows one such
catalysts are enzymes (and cofactors where applicable), reaction where the D-lactate dehydrogenase enzyme
whole cells, and homogeneous catalysts (in organic needs NADH for the conversion of pyruvate to D-lactate.
synthesis). The segregation of particulate heterogeneous In the process, the oxidized form of NADH, viz., NAD+,
catalysts by filters is not under consideration. is produced. Unless this NADH is regenerated from
NAD+, one has to continuously supply NADH from
The production of organic compounds by synthesis in external sources. NADH is costly ($1000/mol). A number
organic solvents or an aqueous-organic biphasic reac- of strategies have been explored to solve this problem
tion medium is very common. Many use homogeneous (see Cheryan and Mehaia6).
catalysts whose molecular weights are considerable The strategy shown in Figure 10 is an important one,
(e.g., in the range of, say, 300-800). Very few mem- viz., an additional enzymatic reaction employing a
branes are capable of retaining such species while regenerating enzyme, in this case an alcohol dehydro-
allowing the organic solvent to pass at appreciable rates. genase, to regenerate the NADH from NAD+. However,
Nanofiltration membranes, just becoming available, this requires a cosubstrate (e.g., ethanol) and yields a
have the necessary solvent resistance and rejection coproduct (e.g., acetaldehyde). To design a continuous
behavior in a few cases. Whu et al.12 have identified process to retain the enzymes as well as the cofactors
in the system by a membrane as the substrates come
some of these capabilities including retaining the ho-
into the reaction chamber and the product and coprod-
mogeneous catalysts while passing the organic solvents.
uct leave the reaction chamber, one needs a very specific
More extensive use of such nanofiltration membranes membrane. Figure 10 provides a special example for
will allow their use in organic synthesis for, among volatile cosubstrate and coproduct. We consider here a
others, retaining the homogeneous catalysts. separate situation, but one more commonly encountered;
The use of membranes to segregate enzymes used as viz., enzymes have molecular weights > 40000 Da and
catalysts for biosynthesis or biocatalysis is practiced in the substrates are 5-12 carbon sugars; note that the
a wide scale. Primarily, ultrafiltration membranes hav- cofactor molecular weights are around 700 Da. Thus, if
ing molecular weight cutoffs in the range of 5000- we use too tight of a membrane, the substrate introduc-
100000 are used to retain the enzymes (molecular tion into the reaction zone will encounter considerable
weight range 10000-100000 Da) in the CSTR reactor resistance, although NAD+ and NADH will be retained
by the membrane.
as smaller products are removed with water through
To solve the problem, Nidetzky et al.69,70 have selected
the membrane (Figure 2). Originally suggested by a charged nanofiltration (NF) membrane having -ve
Michaels67 and implemented as early as 1970,19 this charges on the surface (as shown in Figure 14) that
technique is used commercially for the production of preferably retains the cofactors almost completely with-
amino acids. Jandel et al.68 have illustrated continuous out binding the enzymes or cofactors. However, the NF
production of L-alanine from fumaric acid in a two-stage membrane had a size exclusion (∼1 kD) slightly higher
membrane reactor using the enzyme aspartase in the than those molecules having the molecular mass of the
first stage, cofactor; this ensured higher substrate fluxes. The
charge of the NF membrane (-ve) in this case allows for
aspartase cofactor retention since both NAD and NADH carry
fumaric acid + NH3 {\} L-aspartic acid (13a) negative net charge at pH values greater than 3. This
Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999 3727
Figure 18. Schematic diagrams for four configurations of a reactor where two membranes are separated by the reaction medium.
the two sides of the membrane. As long as the two feed 3. Functions of Multiple Membranes in a
phases are immiscible with the reaction medium, reac- Reactor
tants can partition into the reaction medium and react
and then the products can partition back into the Although more than two different types of membranes
flowing phases on opposite sides of the membrane. can be accommodated in a reactor, this section will
consider primarily the functions of only two membranes
Unfortunately, such a configuration, usually termed as
(different types or similar) in a reactor. There are cases
the supported liquid membrane (SLM), has limited
where two similar membranes carry out somewhat
stability93 because of a variety of reasons including a different functions or the same function vis-à-vis dif-
finite solubility of the reaction medium in the two ferent reactants/products. All such functional variations
different reactant-containing phases. Therefore, as dis- may be considered under two broad categories:
cussed in the next section, Guha and Sirkar,94 Chen et
al.,95 and Guha et al.11,96 have employed two different (1) Multiple membranes physically separated by the
hollow fiber membranes for bringing two reactant- reaction medium
containing streams into the membrane reactor while the (2) Multiple membranes existing as a laminate
reactants partitioned into the liquid reaction medium
contained (confined) in the shell side between the two We have already seen how many functions can be
sets of hollow fibers. implemented through a single membrane. In a practical
situation, only a few such functions (of all possible ones
Kim and Datta97 have used a porous support disk to appearing in Figure 1) are implemented via a given
immobilize a homogeneous catalyst in a high-boiling membrane. Obviously, when there is another mem-
organic solvent. However, they did not completely wet brane, the number of possible combinations increase
the pores so that the gas space was left in the pore. The greatly. Totally new functions may also become possible,
reaction was hydroformylation of ethylene to give pro- e.g., containing a reaction medium, separation of two
pionaldehyde. With an appropriate support membrane, product species, enrichment of product, and intensifica-
the capabilities of the reaction medium within the tion of a reaction. The particularity of a given chemical
membrane could be substantially enhanced. The only reaction will, however, limit the number of desired
issue is the lifetime of the reaction medium. functions. Thus, instead of considering all possible
3732 Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999
combinations of functions possible in a two-membrane PFC medium. When one considers hydrophobic pollut-
reactor, we consider only particular cases that have been ants which are concentrated in the reaction medium by
studied experimentally and/or analyzed theoretically. extraction (section 2.2), one has now considerable
Under appropriate conditions, we briefly identify reac- process intensification. The thickness of the contained
tors having three or four membranes. First, we review PFC liquid layer is also important. By providing suf-
those systems where the reaction medium exists be- ficient thickness, one prevents O3 loss into water or the
tween two or more membranes. loss of a volatile organic pollutant into the gas phase.
3.1. Two or More Membranes Physically Sepa- It would be difficult to achieve it by immobilizing the
rated by the Reaction Medium. Of the many com- reaction medium in the pores of a single membrane,
binations possible, we will consider four types of systems since the required membrane thickness cannot be made
as illustrated in Figure 18. to order. Meanwhile, an extremely efficient pollutant
3.1.1. Homogeneous Catalytic Oxidation of Eth- destruction environment is created in the reaction
ylene. As shown in Figure 18a, the membrane reactor medium.
has two sets of microporous hydrophobic hollow fibers, This basic reactor configuration has been adopted by
one bringing in C2H4 (species A) while the other brings Shanbhag et al.99 to destroy volatile organic compounds
in O2 (species B). The catalytic reaction medium present (VOCs) from a gas phase via ozonation in a PFC liquid
on the shell side (in the manner of a hollow fiber- contained between three membranes in a reactor. One
contained liquid membrane of Majumdar et al.98) is an set of silicone capillaries in the reactor bring in a
aqueous solution of palladium chloride, cupric chloride, gaseous mixture of O3 in O2 which permeate into the
and cupric acetate. Chen et al.95 have studied the PFC phase on the shell side. Another set of silicone
oxidation of C2H4 to CH3CHO in such a system; the capillaries bring in the VOCs in air; the VOCs permeate
product CH3CHO can be removed through both gas through the silicone membrane and are dissolved in the
streams or through one of the gas streams. These PFC phase where they react with O3 and are destroyed.
investigators have shown that controlled and separate The nonvolatile reaction products are extracted into
introduction of C2H4 and O2 (function 2.3) through two water which flows through the bore of a third set of
separate membranes led to much better conversion porous Teflon capillaries in the reactor shell. The
compared to that achievable in a single hollow fiber volatile reaction products are removable through each
membrane-based reactor where both C2H4 and O2 were membrane. Use of silicone capillaries dramatically
simultaneously fed to the reactor in one gas stream. reduces the vapor-phase loss of PFC.
This two-membrane configuration may be replaced by 3.1.3. Efficient Mixing Membrane Reactor. In
a single porous membrane immobilizing the aqueous some biochemical processes, two reactants present in
reaction medium in the membrane pores; C2H4 and O2 two separate aqueous (say) solutions are to be mixed
may be separately introduced on two sides of the together. Direct mixing leads to further dilution of each
membrane. One has to be careful that the reaction reactant by the solvent present in the other incoming
medium properties remain unchanged with time. stream. However, if the scheme of Figure 18c suggested
3.1.2. Multiphase Ozonolysis of Organic Pollut- by Stanley and Quinn100 is adopted, then each mem-
ants. Figure 18b has two sets of hollow fibers: one set brane is used to introduce its own reactant, A and B,
of nonporous silicone capillaries through which gas B respectively, into the reaction liquid between the two
(in this case O3 in O2) is introduced into an organic inert membranes. Thus, the extent of dilution of species A
perfluorocarbon (PFC) reaction medium on the shell side as well as B is considerably reduced. One has to make
and another set of porous Teflon capillaries through sure, however, that the membrane for introduction of
which wastewater containing organic pollutants, e.g., A rejects species B and vice versa. The product P may
species A, flow. Pollutants are partitioned into the be permeable through both membranes. Here, each
contained liquid reaction medium at the aqueous- membrane carries out its own function of controlled
organic interface. The organic reaction medium concen- introduction of one reactant and rejection of the other.
trates the organic pollutants, especially the more hy- However, a combination of the two membranes achieves
drophobic ones, by extraction and allows O3 to react with a novel and important function, among others, i.e.,
these pollutants without the O3 being consumed by prevention of dilution of the two reactants.
nontargeted chemicals present in wastewater. 3.1.4. Partial Oxidation Process for Methane to
When first presented by Guha and Sirkar,94 the C2 Compounds. Oxidative coupling of methane in
system used two sets of microporous hydrophobic hollow membrane reactors is of considerable interest. Lu et al.48
fiber membranes instead of nonporous silicone capil- have made a numerical study for a cross-flow reactor
laries and porous Teflon capillaries. One could use just with a distributed feed of O2 and product removal. They
one set of microporous fibers and immobilize the reac- have modeled a number of reactor configurations. Their
tion medium in the pores. Both would lead to a high results show that distributed feed O2 could give rise to
loss, primarily by vaporization and also by solubilization much higher C2 yields than the cofeed reactor as long
of the perfluorocarbon (PFC) liquid which has a low but as the ratio of catalyst loading to initial methane flow
finite vapor pressure. The silicone membrane reduced rate was sufficiently high. On one hand, their calcula-
the vaporization-based loss of PFC liquid drastically. tions have further shown that although reactors with
Guha et al.11,96 have studied the system of Figure 18b optimally distributed O2 feed give much higher yield
using silicone capillaries and have demonstrated its than evenly distributed ones, the improvement is not
high efficiency in destroying most pollutants. The reac- significant. On the other hand, if a two-membrane
tion products partition back into the water phase and reactor is used (Figure 18d) where one membrane is
the gas phase on the two sides. This system avoids direct used for the O2 feed and the other for C2 product
ozonation of pollutants in wastewater where most of the removal, higher C2 yields could be obtained than that
O3 is wasted. Further, O3 has a very low solubility in in a single membrane reactor with distributed O2 feed.
water, whereas it has 14 times higher solubility in the However, if the membrane for C2 product removal is also
Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999 3733
by the nonpolar liquid membrane and therefore ac- Kim and Datta97 have suggested using a porous support
cumulates in the sweep aqueous liquid flowing on the disk to load the catalyst by evaporation of a volatile
other side. If the sweep liquid flow rate is much lower solvent in the catalyst solution impregnated into the
than that of the feed, it is possible to increase the pores. During the reaction, an essentially nonvolatile
concentration of P in the sweep stream to a very high solvent, dioctylphthalate, is used. To protect this solu-
value compared to the feed concentration of the reactant tion from being forced out, a second nonselective mem-
R. Thus, the membrane laminate allows simultaneous brane is used (Figure 19c) whereas the first membrane
separation of the product from the feed stream and its must be freely permeable to all the reactants but not
considerable enrichment in the product stream. Any the product. With a low flow rate of the sweep stream
inert in the feed, if it partitions through the liquid in contact with the second membrane, the product
membrane, will be present in the product stream at a concentration can be raised to a high level compared to
concentration lower than that in the feed. Matson and the feed concentration level (as in Figure 19a,b). Un-
Quinn101 have experimentally demonstrated this con- fortunately the membrane employed as the first mem-
cept by resolving a racemic aqueous mixture of a D,L- brane was a ultrafiltration-type membrane of 50 Å pore
ester where the L-ester was converted to the L-acid size; therefore, it had only a Knudsen diffusion-type
enzymatically in the catalytic membrane containing the selectivity (1.44) between ethylene and propionaldehyde.
enzyme chymotrypsin. The product stream was highly If a reaction
enriched in the L-acid.
3.2.2. Separation of Multiple Products. The con- AfBTC (26)
cept may be illustrated by considering the schematic of
Figure 19b. Reactant A in the feed solution is highly is desired to be carried out where C is the desired
permeable though the permselective membrane exposed product, then the schematic of Figure 19c can be
to the feed solution. On the other side of this membrane, employed, except that membrane 2 must also reject the
reactant A enters a catalytic membrane in an aqueous intermediate product B and be freely permeable to
environment and undergoes the following reaction: product C (but not to reactant A). This has been termed
as the “trapped intermediates” membrane bioreactor
AfB+C (24) since the intermediate B is trapped between the mem-
brane 1 and 2 and its increasing concentration forces
It so happens that product C is also very permeable the reaction in the direction of C.101 Contrast this
through the permselective membrane but product B is scheme with those of reactions 4a and 4b where the
rejected by this membrane. On one hand, if now the intermediate B is the desired product:14 the membrane
sweep solution on the other side of the catalytic mem- plays quite a different role for species B in these two
brane has a flow rate much lower than that of the feed, cases.
product B concentration will be very high in this sweep
stream. On the other hand, the concentration of C will 4. Concluding Remarks
be essentially the same in both solutions; however, most
of C will appear in the feed stream since its flow rate is A membrane in a reactor is capable of performing a
much higher. Thus, products B and C are effectively wide variety of useful functions. These capabilities can
separated; further, product B concentration is also much result in the performance levels of the membrane
higher than that of A. reactor and the process to be considerably higher than
those of a conventional reactor under similar operating
Matson and Quinn101 have demonstrated the utility
conditions. Membrane functions that are particularly
of this concept by carrying out the hydrolysis of N-
useful and unique in this regard are separation of
acetyl-L-tyrosine ethyl ester into the corresponding acid
product(s) from the reaction mixture, separation of a
(product B) and ethanol (product C); the catalytic
reactant from a mixed feed for introduction to the
membrane employs chymotrypsin covalently bound to
reactor, controlled addition of a reactant or two reac-
a Millipore filter to catalyze this hydrolysis. The perm-
tants, and a solid-electrolyte membrane acting as sup-
selective membrane was an immobilized liquid mem-
port for the electrodes, allowing the conduction of ions
brane of decanol through which alcohol, product C, as
and having reactions on its surfaces. When compared
well as the reactant ester are freely permeable.
with conventional reactors, there are additional unique
The bipolar ion-exchange membrane described in
functions, e.g., nondispersive phase contacting, segrega-
section 2.1.1.6 may also be considered as a membrane
tion of a catalyst (and cofactor) in a reactor, catalyst
laminate where a cation-exchange membrane and an
immobilization, etc., which impart considerable opera-
anion-exchange membrane have been joined together,
tional and economic advantages to devices in reaction
even though there is some space between them where
processes employing membrane(s). Simultaneous achieve-
water comes in from both sides. Further, under the
ment of a number of such functions by a membrane in
applied potential difference between the two electrodes
a reactor has made membrane reactors quite attractive.
(Figure 7), water dissociates in this space and two
Even when the membrane is located outside the reactor
products of the reaction, H+ and OH-, are effectively
and connected to it in a recycle mode, such coupled
separated by the two ion-exchange membranes.
reaction-separation processes provide considerable im-
3.2.3. Supported Liquid-Phase Catalytic Mem- provements in the overall reaction performance.
brane Reactor-Separator for Homogeneous Ca- Many investigations on membrane reactors have been
talysis. If instead of a covalently bound enzyme catalyst conducted in biochemical processing, petrochemical
in the membrane, a homogeneous catalyst is necessary, applications, and environmental pollution control. A few
say, for the hydroformylation of ethylene to form pro- processes employing polymeric membranes have been
pionaldehyde by the reaction commercialized. Utilizing more than one membrane in
a reactor imparts additional functional capabilities. For
H2 + CO + CH2dCH2 f CH3CH2CHO (25) effective large-scale utilization of the diverse functional
Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999 3735
capabilities of a membrane in a reactor, considerable (17) Fournier, R. L. Mathematical Model of Microporous Hol-
research and development on membrane lifetime, avail- low-Fiber Membrane Extractive Fermentor. Biotechnol. Bioeng.
able space time, module design, membrane fouling, 1988, 31, 235.
(18) Vasudevan, M.; Matsuura, T.; Chotani, G. K.; Vieth, W.
membrane poisoning, and membrane cost are essential. R. Membrane Transport and Biocatalytic Reaction in an Im-
This is especially true for inorganic membranes and mobilized Yeast Membrane Reactor. Ann. N.Y. Acad. Sci. 1987,
higher temperature applications where the hurdles for 506, 345.
industrial use are considerable. (19) Ghose, T. K.; Kostick, J. T. A Model for Continuous
Enzymatic Saccharification of Cellulose with Simultaneous Re-
moval of Glucose Syrup. Biotechnol. Bioeng. 1970, 12, 921.
Acknowledgment (20) Closset, G. P.; Shah, Y. T.; Cobb, J. T. Analysis of
Membrane Reactor Performance for Hydrolysis of Starch by
This article is dedicated to George Keller for his many Glucoamylase. Biotechnol. Bioeng. 1973, 15, 441.
contributions. Along with his colleague M. Bhasin at (21) David, M.-O.; Nguyen, T. Q.; Neel, J. Pervaporation-
Union Carbide, George Keller carried out a pioneering Esterification Coupling: Part II. Modelling of the Influence of
study on oxidative coupling of methane to C2 hydrocar- Different Operating Parameters. Trans Inst. Chem. Eng. 1991,
bons. Many current and large efforts are focused on 69(A), 441.
achieving improved oxidative coupling of methane via (22) Strathmann, H.; Gudernatsch, W. Continuous Removal of
Ethanol from Bioreactor by Pervaporation. In Extractive Biocon-
membrane reactors. The senior author wishes to ac-
versions; Mattiasson, B., Holst, O., Eds.; Bioprocess Technology;
knowledge invaluable discussions with Ka Ng, Paul Marcel Dekker: New York, 1991; Vol. 11.
Bryan, D. Bhattacharyya, T. Tsotsis, Steve Matson, W. (23) Sirkar, K. K. Other New Membrane Processes. In Mem-
S. Winston Ho, Arvind Varma, and Richard Noble on brane Handbook; Ho, W. S. W., Sirkar, K. K., Eds.; Chapman and
various aspects of a membrane in a reactor. P.V.S. Hall: New York, 1992.
would like to thank Asim K. Guha, Sudipto Majumdar, (24) Twardowski, Z.; McGilvey, J. D. Membrane Pervaporation
and Zhi-Fa Yang for sharing their insights and thoughts Process. U.S. Patent 4,683,039, July 18, 1987.
(25) Strathmann, H. Ion-Exchange Membranes. In Membrane
on membranes and their applications. Handbook; Ho, W. S. W., Sirkar, K. K., Eds.; Chapman and Hall:
New York, 1992.
Literature Cited (26) Ricks, E. E.; Estrada-Valdes, M. C.; McLean, T. L.;
Iacobucci, G. A. Highly Enantioselective Hydrolysis of (R,S)-
(1) Hsieh, H. P. Inorganic Membranes for Separation and Phenylalanine Isopropyl Ester by Subtilisin Carlsberg. Continuous
Reaction; Elsevier Science: Amsterdam, 1996. Synthesis of (S)-Phenylalanine in a Hollow Fiber/Liquid Mem-
(2) Falconer, J. L.; Noble, R. D.; Sperry, D. P. Catalytic brane Reactor. Biotechnol. Prog. 1992, 8, 197.
Membrane Reactors. In Membrane Separations Technology. Prin- (27) Ollis, D. F.; Thompson, J. B.; Wolynic, E. T. Catalytic
ciples and Applications; Noble, R. D., Stern, S. A., Eds.; Elsevier Liquid Membrane Reactor: I. Concept and Preliminary Experi-
Science: Amsterdam, 1995. ments in Acetaldehyde Synthesis. AIChE J. 1972, 18 (2), 457.
(3) Govind, R., Itoh, N., Eds. Membrane Reactor Technology; (28) Sun, Y.-M.; Khang, S.-J. Catalytic Membrane for Simul-
AIChE Symposium Series 268; AIChE: New York, 1989; Vol. 85. taneous Chemical Reaction and Separation Applied to a Dehy-
(4) Lòpez, J. L.; Matson, S. L.; Stanley, T. J.; Quinn, J. A. drogenation Reaction. Ind. Eng. Chem. Res. 1988, 27, 1136.
Liquid-Liquid Extractive Membrane Reactors. In Extractive (29) Ziaka, Z.; Minet, R. G.; Tsotsis, T. T. Propane Dehydro-
Bioconversions; Mattiasson, B., Holst, O., Eds.; Bioprocess Tech- genation in a Packed-Bed Membrane Reactor. AIChE J. 1993, 39
nology; Marcel Dekker: New York, 1991; Vol. 11. (3), 526.
(5) Matson, S. L.; Quinn, J. A. Membrane Reactors. In Mem- (30) Wu, J. C. S.; Liu, P. K. T. Mathematical Analysis on
brane Handbook; Ho, W. S. W., Sirkar, K. K., Eds.; Chapman and Catalytic Dehydrogenation of Ethylbenzene Using Ceramic Mem-
Hall: New York, 1992. branes. Ind. Eng. Chem. Res. 1992, 31, 322.
(6) Cheryan, M.; Mehaia, M. A. Membrane Bioreactors. In (31) Mohan, K.; Govind, R. Analysis of a Cocurrent Membrane
Membrane Separations in Biotechnology; McGregor, W. C., Ed.; Reactor. AIChE J. 1986, 32, 2083.
Bioprocess Technology; Marcel Dekker: New York, 1986; Vol. 1. (32) Collins, J. P.; Schwartz, R. W.; Sehgal, R.; Ward, T. L.;
(7) Tsotsis, T. T.; Champagnie, A. M.; Minet, R. G.; Liu, P. K. Brinker, C. J.; Hagen, G. P.; Udovich, C. A. Catalytic Dehydro-
T. Catalytic Membrane Reactors. In Computer-Aided Design of genation of Propane in Hydrogen Permselective Membrane Reac-
Catalysts; Becker, E. R., Pereira, E. R., Eds.; Chemical Industries; tors. Ind. Eng. Chem. Res. 1996, 35, 4398.
Marcel Dekker: New York, 1993; Vol. 51. (33) Ioannides, T.; Gavalas, G. R. Catalytic Isobutane Dehy-
(8) Armor, J. N. Applications of Catalytic Inorganic Membrane drogenation in a Dense Silica Membrane Reactor. J. Membr. Sci.
Reactors to Refinery Products. J. Membr. Sci. 1998, 147, 217. 1993, 77, 207.
(34) Collins, J. P.; Way, J. D. Catalytic Decomposition of
(9) Saracco, G.; Versteeg, G. F.; van Swaaij, W. P. M. Current
Ammonia in a Membrane Reactor. J. Membr. Sci. 1994, 96, 259.
Hurdles to the Success of High Temperature Membrane Reactors.
J. Membr. Sci. 1994, 95, 105. (35) Uemiya, S.; Sato, N.; Ando, H.; Kikuchi, E. The Water Gas
Shift Reaction Assisted by a Palladium Membrane Reactor. Ind.
(10) Ho, W. S. W., Sirkar, K. K., Eds. Membrane Handbook;
Eng. Chem. Res. 1991, 30, 585.
Chapman and Hall: New York, 1992.
(36) Edlund, D. J.; Venkatraman, V. K. A Catalytic Membrane
(11) Guha, A. K.; Shanbhag, P. V.; Sirkar, K. K.; Vaccari, D. Reactor for Facilitating the Water-Gas Shift Reaction at High
A.; Trivedi, D. H. Multiphase Ozonolysis of Organics in Waste- Temperature; Technical Report DOE/METC-94/1008; Proceedings
water by a Novel Membrane Reactor. AIChE J. 1995, 41, 1998. of the Coal-Fired Power Systems 94, Advances in IGCC and PFBC
(12) Whu, J.; Baltzis, B. C.; Sirkar, K. K. Modeling of Nano- Review Meeting; U.S. Department of Energy: Washington, DC,
filtration-Aided Organic Synthesis. J. Membr. Sci. 1999, in press. 1994; Vol. 2, p 709-B.
(13) Raich, B. A.; Foley, H. C. Ethanol Dehydrogenation with (37) Willms, R. S.; Willhelm, R.; Okuno, K. Performance of
a Palladium Membrane Reactor: An Alternative to Wacker Palladium Membrane Reactor Using a Ni Catalyst for Fusion Fuel
Chemistry. Ind. Eng. Chem. Res. 1998, 37, 3888. Impurities. Fusion Eng. Des. 1995, 28, 397.
(14) Agarwalla, S.; Lund, C. R. F. Use of a Membrane Reactor (38) Dixon, A. G.; Moser, W. R.; Ma, Y. H. Waste Reduction
to Improve Selectivity to Intermediate Products in Consecutive and Recovery using O2-Permeable Membrane Reactors. Ind. Eng.
Catalytic Reactions. J. Membr. Sci. 1992, 70, 129. Chem. Res. 1994, 33, 3015.
(15) Frank, G. T.; Sirkar, K. K. Alcohol Production by Yeast (39) Rezac, M. E.; Koros, W. J.; Miller, S. J. Membrane-Assisted
Fermentation and Membrane Extraction. Biotechnol. Bioeng. Dehydrogenation of n-Butane: Influence of Membrane Properties
Symp. Ser. 1985, 15, 621. on System Performance. J. Membr. Sci. 1994, 93, 193.
(16) Kang, W.; Shukla, R.; Sirkar, K. K. Ethanol Production in (40) Balachandran, U.; Dusek, J. T.; Maiya, P. S.; Mieville, R.
a Microporous Hollow-Fiber Based Extractive Fermentor with L.; Kleefisch, M. S.; Udovich, C. A. Ceramic Membrane Reactor
Immobilized Yeast. Biotechnol. Bioeng. 1990, 34, 826. for Converting Methane to Syngas. Catal. Today 1997, 36, 265.
3736 Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999
(41) McKenna, E. A.; Othoneos, A.; Kiratzis, N.; Stoukides, M. (65) Stanley, T. J.; Quinn, J. A. Phase Transfer Catalysis in a
Synthesis of HCN in a Solid-Electrolyte-cell Reactor. Ind. Eng. Membrane Reactor. Chem. Eng. Sci. 1987, 42, 2313.
Chem. Res. 1993, 32, 1904. (66) Loṕez, J. L.; Matson, S. L. A Multi-phase/Extractive
(42) Brookes, P. R.; Livingston, A. G. Biotreatment of a Point- Enzyme Membrane Reactor for Production of Diltiazem Chiral
Source Industrial Wastewater Arising in 3,4-Dichloroaniline Intermediate. J. Membr. Sci. 1997, 125, 189.
Manufacture Using an Extractive Membrane Bioreactor. Biotech- (67) Michaels, A. S. Ultrafiltration. In Progress in Separation
nol. Prog. 1994, 10, 65. and Purification; Perry, E. S., Ed.; Wiley-Interscience: New York,
(43) Basu, R.; Sirkar, K. K. Pharmaceutical Product Recovery 1968; Vol. 1.
Using a Hollow Fiber Contained Liquid Membrane: A Case Study. (68) Jandel, A. S.; Hustedt, H.; Wandrey, C. Continuous
J. Membr. Sci. 1992, 75, 131. Production of L-Alanine from Fumarate in a Two-Stage Membrane
(44) van Eikeren, P.; Brose, D. J.; Muchmore, D. C.; West, J. Reactor. Eur. J. Appl. Microbiol. Biotechnol. 1982, 15, 59.
B. Membrane-Assisted Synthesis of Chiral Drugs and Fine (69) Nidetzky, B.; Haltrich, D.; Kulbe, K. D. Carry Out Coen-
Chemicals. Ann. N.Y. Acad. Sci. 1990, 613, 796. zyme Conversions Economically. CHEMTECH 1996, 1, 31.
(45) Itoh, N.; Govind, R. Combined Oxidation and Dehydroge- (70) Nidetzky, B.; Neuhauser, W.; Haltrich, D.; Kulbe, K. D.
nation in a Palladium Membrane Reactor. Ind. Eng. Chem. Res. Continuous Enzymatic Production of Xylitol with Simultaneous
1989, 28, 1554. Coenzyme Regeneration in a Charged Membrane Reactor. Bio-
(46) Tonkovich, A. L. Y.; Secker, R. B.; Reed, E. L.; Roberts, G. technol. Bioeng. 1996, 52, 387.
L.; Cox, J. L. Membrane Reactor/Separator: A Design for Bimo- (71) Wichman, R.; Wandrey, C.; Buckmann, A. F.; Kula, M. R.
lecular Reactant Addition. Sep. Sci. Technol. 1995, 30 (7-9), 1609. Continuous Enzymatic Transformation in an Enzyme Membrane
(47) Tonkovich, A. L. Y.; Jimenez, D. M.; Zilka, J. L.; Roberts, Reactor with Simultaneous NAD(H) Regeneration. Biotechnol.
G. L. Inorganic Membrane Reactors for the Oxidative Coupling of Bioeng. 1981, 23, 2789.
Methane. Chem. Eng. Sci. 1996, 51 (11), 3051. (72) Kragl, U.; Dreisbach, C. Continuous Asymmetric Synthesis
(48) Lu, Y.; Dixon, A. G.; Moser, W. R.; Ma, Y. H. Analysis and in a Membrane Reactor. Angew. Chem. 1996, 35, 642.
Optimization of Cross-Flow Reactors for Oxidative Coupling of (73) Jones, C. K. S.; Yang, R. Y. K.; White, E. T. A Novel
Methane. Ind. Eng. Chem. Res. 1997, 36, 559. Hollow-Fiber Reactor with Reversible Immobilization of Lactase.
(49) Zaspalis, V. T.; van Praag, W.; Keizer, K.; van Ommen, J. AIChE J. 1988, 34, 293.
G.; Ross, J. R. H.; Burggraaf, A. J. Reactor Studies Using Alumina (74) Drioli, E.; Scardi, V. Ultrafiltration Processing with En-
Separation Membranes for the Dehydrogenation of Methanol and zyme Gel Composite Membranes. J. Membr. Sci. 1976, 1, 273.
n-Butane. Appl. Catal. 1991, 74, 223. (75) Matson, S. L.; Quinn, J. A. Product Separation and
(50) Zaspalis, V. T.; van Praag, W.; Keizer, K.; van Ommen, J. Enrichment in a Membrane Reactor. Presented at the AIChE 72nd
G.; Ross, J. R. H.; Burggraaf, A. J. Reactions of Methanol over Annual Meeting, San Francisco, CA, 1979.
Alumina Catalytically Active Membranes Modified by Silver. Appl. (76) Butterfield, D. A.; Subramaniam, R.; Bhattacharyya, D.;
Catal. 1991, 74, 235. Vishwanath, S.; Huang, W.; Bachas, L. Biofunctional Mem-
(51) Cini, P.; Harold, M. P. Experimental Study of the Tubular branes: Molecular Recognition Chemistry and the Active Site
Multiphase Catalyst. AIChE J. 1991, 37, 997. Structure of Enzymes Site-Specifically Immobilized onto Polymeric
(52) Torres, M.; Sanchez, J.; Dalmon, J. A.; Bernauer, B.; Lieto, Supports: An Electron Paramagnetic Study. Poster 33 presented
J. Modeling and Simulation of a Three-Phase Catalytic Membrane at the 9th Annual Meeting of the North American Membrane
Reactor for Nitrobenzene Hydrogenation. Ind. Eng. Chem. Res. Society, Baltimore, MD, May 31-June 4, 1997.
1994, 33, 2421. (77) Kang, W. K.; Shukla, R.; Sirkar, K. K. Novel Membrane-
Based Immobilization Technique for Bioreactors. Ann. N.Y. Acad.
(53) Brindle, K.; Stephenson, T.; Semmens, M. J. Nitrification
Sci. 1990, 589, 192.
and Oxygen Utilization in a Membrane Aeration Bioreactor. J.
Membr. Sci. 1998, 144, 197. (78) Sloot, H. J.; Smolders, C. A.; van Swaaij, W. P. M.;
Versteeg, G. F. High-Temperature Membrane Reactor for Catalytic
(54) Parvatiyar, M. G.; Govind, R.; Bishop, D. F. Biodegradation
Gas-Solid Reactions. AIChE J. 1992, 38 (6), 887.
of Toluene in a Membrane Biofilter. J. Membr. Sci. 1996, 119, 17.
(79) Szenger, J.; Yeung, K. L.; Varma, A. Effect of Catalyst
(55) Shanbhag, P. V.; Guha, A. K.; Sirkar, K. K. Single-Phase
Distribution in a Membrane Reactor: Experiments and Model.
Membrane Ozonation of Hazardous Organic Compounds in Aque-
AIChE J. 1997, 43 (8), 2059.
ous Streams. J. Hazard Mater. 1995, 41, 95.
(80) Fritsch, D. Reactions with Catalytically Active Polymer
(56) Shanbhag, P. V.; Guha, A. K.; Sirkar, K. K. Membrane- Membranes at Low Temperature. Presented at the 1996 14th
Based Ozonation of Organic Compounds. Ind. Eng. Chem. Res. Annual Membrane Technology/Separations Planning Conference,
1998, 37, 4388. Business Communications Co., Newton, MA, Oct 28-30, 1996.
(57) Lee, J.-Y.; Choi, Y.-B.; Kim, H.-S. Simultaneous Biodeg- (81) Bellobono, I. R.; Barni, B.; Gianturco, F. Pre-industrial
radation of Toluene and p-Xylene in a Novel Bioreactor: Experi- Experience in Advanced Oxidation and Integral Photodegradation
mental Results and Mathematical Analysis. Biotechnol. Prog. of Organics in Potable Waters and Wastewaters by PHOTOPERM
1993, 9, 46. Membranes Immobilizing Titanium Dioxide and Promoting Pho-
(58) Bar, R.; Gainer, J. L. Acid Fermentation in Water Organic tocatalysts. J. Membr. Sci. 1996, 102, 139.
Solvent, Two Liquid Phase Systems. Biotechnol. Prog. 1987, 3, 109. (82) Chemseddine, B.; Audinos, R. Use of Ion-Exchange Mem-
(59) Kiani, A.; Bhave, R. R.; Sirkar, K. K. Solvent Extraction branes in a Reactor for Esterification of Oleic Acid and Methanol
with Immobilized Interfaces in a Microporous Hydrophobic Mem- at Room Temperature. J. Membr. Sci. 1996, 115, 77.
brane. J. Membr. Sci. 1984, 20, 125. (83) Bagnell, L.; Cavell, K.; Hodges, A. M.; Mau, A. W.-H.; Seen,
(60) Prasad, R.; Sirkar, K. K. Membrane-Based Solvent Extrac- A. J. The Use of Catalytically Active Pervaporation Membranes
tion. In Membrane Handbook; Ho, W. S. W., Sirkar, K. K., Eds.; in Esterification Reactions To Simultaneously Increase Product
Chapman and Hall: New York, 1992. Yield, Membrane Permselectivity and Flux. J. Membr. Sci. 1993,
(61) Frank, G. T.; Sirkar, K. K. An Integrated Bioreactor- 85, 291.
Separator. In-situ Recovery of Fermenatation Products by a Novel (84) Dai, X.-P.; Kovvali, A. S.; Sirkar, K. K. Pore Flow through
Dispersion-Free Solvent Extraction Technique. Biotechnol. Bioeng. Reactor (PFTR). In preparation.
Symp. Ser. 1986, 17, 303. (85) Pina, M. P.; Irusta, S.; Menendez, M.; Santamaria, J.
(62) Shukla, R.; Kang, W. K.; Sirkar, K. K. Acetone-Butanol- Combustion of Volatile Organic Compounds over Platinum-Based
Ethanol (ABE) Production in a Novel Hollow Fiber Fermentor- Catalytic Membranes. Ind. Eng. Chem. Res. 1997, 36 (11), 4557.
Extractor. Biotechnol. Bioeng. 1989, 34, 1158. (86) Pina, M. P.; Menendez, M.; Santamaria, J. The Knudsen-
(63) Hoq, M. M.; Yamane, T.; Shimizu, S.; Funada, T.; Ishida, Diffusion Catalytic Membrane Reactor: An Efficient Contactor for
S. Continuous Hydrolysis of Olive Oil by Lipase in a Microporous the Combustion of Volatile Organic Compounds. Appl. Catal. 1996,
Hydrophobic Membrane Bioreactor. J. Am. Oil Chem. Soc. 1985, 11, L19.
62 (6), 1016. (87) Wu, S. Q.; Bouchard, C.; Kaliaguire, S. Zeolite Containing
(64) Molinari, R.; Santoro, M. E.; Drioli, E. Study and Com- Catalytic Membranes as Interphase Contactors. Presented at the
parison of Two Enzyme Membrane Reactors for Fatty Acids and 8th Annual Meeting of the North American Membrane Society,
Glycerol Production. Ind. Eng. Chem. Res. 1994, 33, 2591. Ottawa, Canada, May 22, 1996.
Ind. Eng. Chem. Res., Vol. 38, No. 10, 1999 3737
(88) Fuller, T. F.; Newman, J. Water and Thermal Management (96) Guha, A. K.; Shanbhag, P. V.; Sirkar, K. K.; Yun, C. H.;
in Solid-Polymer-Electrolyte Fuel Cells. J. Electrochem. Soc. 1993, Trivedi, D. H.; Vaccari, D. A. Novel Membrane-Based Separation
140 (5), 1218. and Oxidation Technologies. Waste Manage. 1993, 13, 395.
(89) Leddy, J.; Vanderborgh, N. E. Modeling of the Electrode/ (97) Kim, J. S.; Datta, R. Supported Liquid-Phase Catalytic
Ionomer Interface. Diaphragms, Separators and Ion-Exchange Membrane Reactor-Separator for Homogeneous Catalysis. AIChE
Membranes; Van Zee, J. W., White, R. E., Kinoshita, K., Burney, J. 1991, 37 (11), 1657.
H. S., Eds.; The Electrochemical Society Proceedings Series; The (98) Majumdar, S.; Guha, A. K.; Sirkar, K. K. A New Liquid
Electrochemical Society: Pennington, NJ, 1968; pp 15-26. Membrane Technique for Gas Separation. AIChE J. 1988, 34,
(90) Marnellos, G.; Stoukides, M. Ammonia Synthesis at At- 1135.
mospheric Pressure. Science 1998, 282, 98.
(99) Shanbhag, P. V.; Guha, A. K.; Sirkar, K. K. Membrane-
(91) Schwartz, M. Personal communications, Eltron Research
Based Integrated Absorption-Oxidation Reactor for Destroying
Inc. Oct 22, 1996.
VOCs in Air. Environ. Sci. Technol. 1996, 30, 3435.
(92) Adris, A.-E. M.; Grace, J. R. Characteristics of Fluidized-
Bed Membrane Reactors: Scale-up and Practical Issues. Ind. Eng. (100) Stanley, T. J.; Quinn, J. A. Efficient Mixing in Membrane
Chem. Res. 1997, 36 (11), 4549. Bioreactors: Minimizing Dilution Effects. J. Membr. Sci. 1987,
(93) Kemperman, A. J. B.; Bargeman, D.; Boomgaard, T. V. D.; 30, 243.
Strathmann, H. Stability of Supported Liquid Membranes. Sep. (101) Matson, S. L.; Quinn, J. A. Membrane Reactors in
Sci. Technol. 1996, 31 (20), 2733. Bioprocessing. Ann. N.Y. Acad. Sci. 1986, 469, 152.
(94) Guha, A. K.; Sirkar, K. K. A Novel Separation-Reaction-
Separation System. Presented at the AIChE Annual Meeting, Received for review January 29, 1999
Chicago, IL, Nov 13, 1990; paper 58g. Revised manuscript received July 14, 1999
(95) Chen, S.; Fan, H.; Kao, Y.-K. A Membrane Reactor with Accepted July 19, 1999
Two Dispersion-Free Interfaces for Homogeneous Catalytic Reac-
tions. Chem. Eng. J. 1992, 49, 35. IE990069J