Two-phase (bio)catalytic reactions in a table-top centrifugal contact separator

Article abstract of DOI:10.1002/anie.200705426

(Figure Presented) A new spin on catalysis: A table-top centrifugal contact separator allows for fast continuous two-phase reactions to be performed by intimately mixing two immiscible phases and then separating them. Such a device has been used to produce biodiesel from sunflower oil and MeOH/NaOMe. A lipase-catalyzed esterification of oleic acid with nBuOH (see picture) also proceeds with high conversion and can be run for up to 13 h.

Full text of DOI:10.1002/anie.200705426

Angewandte
Chemie
DOI: 10.1002/anie.200705426
Process Intensification
Two-Phase (Bio)Catalytic Reactions in a Table-Top Centrifugal
Contact Separator**
Gerard N. Kraai, Floris van Zwol, Boelo Schuur, Hero J. Heeres,* and Johannes G. de Vries*
Batch-wise production is the state-of-the art in the manufac-
ture of fine chemicals, as it allows the multiple use of reactors
for different processes. However, the use of batch reactors has
some serious drawbacks: For the production of larger
quantities, multiple batch runs have to be performed and
this often leads to batch to batch variation in the product
quality and performance. Furthermore, the productivity is
often lower than for dedicated continuous reactors, and fixed
costs are significant because it is labour intensive. Therefore,
on the order of seconds. In addition, these concepts are not
always easily scaled up to ton amounts.
We decided to focus our research activities in this field on
the use of a table-top-sized flow reactor for (bio-)catalytic
reactions. The device is a centrifugal contact separator (CCS)
that has been used for oil–water separation (for example, for
[
12]
cleaning up oil spills),
for the continuous extraction of
[13]
fermentation products, such as penicillin and phenylala-
nine, and in the atomic waste industry for the extraction
and purification of radioactive waste.
[14]
[1]
[15,16]
switching to continuous processes appears highly beneficial.
Figure 1 shows a
In addition, continuous production in small flow reactors is
very advantageous for reactions in which highly toxic and/or
explosive materials are used or produced.
Based on this analysis, many groups have started to work
on concepts of process intensification aimed at the develop-
ment of smaller reactors or the integration of the reactor with
schematic representation of a CCS. The device is in essence a
centrifuge. The immiscible liquid phases are introduced in the
small annular mixing zone between the outside of the rotor
and the inside of the outer housing. Here, very efficient and
fast mixing between the two phases occurs, which is highly
conducive to a two-phase catalytic reaction. The dispersion is
then sucked inside the centrifuge, where the two phases are
gradually but very efficiently separated whilst moving
upwards, after which they leave the device through separate
exits.
[
2,3]
the separation.
undoubtedly the use of microreactors.
workers have recently reported the use of a microstructured
The most visible activity in this field is
[4–6]
Poechlauer and co-
[
7]
reactor for the ton-scale execution of a Ritter reaction.
Wakami and Yoshida have reported the pilot-scale produc-
tion of a Grignard exchange reaction in a microreactor. The
use of enzymes as catalysts in microreactors has also been
examined. Ley and Baxendale published a series of papers
that describe a cascade of reactions in microflow reactors
where the reagents are present in an immobilized form.
The principle of cascade catalysis in a sequence of
To the best of our knowledge the use of CCSs as chemical
reactors has not been reported to date. A centrifuge has been
used to continuously remove the polymeric product (dextran)
[
8]
[9]
[17]
from the enzymatic conversion of sucrose in a batch mode.
It is potentially very attractive to use a CCS for biphasic
[
10]
[
11]
continuous flow reactors is highly interesting.
T o be
effective, high conversions are required in each reactor of
the sequence. However, not many reactions are fast enough to
be used in microreactors, where residence times are typically
[
*] Ir. G. N. Kraai, F. van Zwol, Ir. B. Schuur, Prof. Dr. Ir. H. J. Heeres
Department of Chemical Engineering, University of Groningen
Nijenborgh 4, 9747 AG Groningen (The Netherlands)
Fax: (+31)50-363-4174
E-mail: h.j.heeres@rug.nl
Prof. Dr. J. G. de Vries
DSM Pharmaceutical Products, Advanced Synthesis, Catalysis &
Development, P.O. Box 16, 6160 MD Geleen (The Netherlands)
and
Department of Organic and Molecular Inorganic Chemistry
Stratingh Institute, University of Groningen
Nijenborgh 4, 9747 AG Groningen (The Netherlands)
Fax: (+31)46-476-7604
E-mail: hans-jg.vries-de@dsm.com
[**] We acknowledge helpful discussions with Dr. Gerard Kwant (DSM)
and Prof. D. B. Janssen, Prof. B. L. Feringa, and Prof. A. J. Minnaard
(
University of Groningen). This project was funded by the Dutch
IBOS program, with contributions from DSM, Organon, the Dutch
Ministry of Economic Affairs, and the Dutch Science Foundation.
Figure 1. Schematic cross-section of a centrifugal contact separator
(Courtesy of CINC-Solutions, The Netherlands).
Angew. Chem. Int. Ed. 2008, 47, 3905 –3908
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3905

Communications
[18]
liquid–liquid (catalytic) reactions. In this case, the annular
zone acts as the reactor and the centrifuge as the liquid–liquid
separator. Two-phase catalysis in flow devices has been
[
19]
[20]
reported by de Bellefon et al.
and by Claus et al.,
however, without integrated phase separation. Ryu and co-
workers have reported a two-phase Heck reaction in a
microreactor in which the phase containing the palladium
catalyst was an ionic liquid. In this case the separation and
[21]
recycling of the catalyst was fully integrated.
Cascade
catalysis with different types of catalysts is in principle
possible by connecting a number of these devices, as shown in
Figure 2.
Figure 2. Cascade two-phase catalysis using CCSs in series.
Figure 3. Continuous conversion (in duplicate) of sunflower oil into
FAME in a CCS at 30 Hz.
We decided to test the efficiency of the CCS for the
[22]
continuous production of biodiesel from sunflower oil. The
reaction is a typical example of a catalytic liquid–liquid
reaction. It was performed on sunflower oil with a sixfold
molar excess of methanol at elevated temperatures (608C)
and using a basic catalyst (NaOMe, 1% w/w with respect to
sunflower oil). The CCS was equipped with a heating jacket to
ensure isothermal conditions. The sunflower oil was pre-
À1
heated to 608C and was pumped at 12.6 mLmin into one
entrance of the CCS. A solution of NaOMe in MeOH was
then introduced through the other entrance at a flow rate of
À1
3
.1 mLmin . After about 40 minutes, the system reached a
Figure 4. Liquid hold-up in the CCS at low- (left) and high-speed
(right); blue=heavy phase, yellow=light phase, green=mixed phase.
steady state and the fatty acid methyl esters (FAME)
containing some residual sunflower oil came out as the light
phase, whereas the heavy phase consisted of a solution of
glycerol in MeOH. Depending on the rate of the centrifuge, a
maximum yield of 96% of FAME was reached under these
catalysts to test in the CCS. As enzymes can be easily
damaged by shear forces, we used a CCS with a low-mix
bottom plate. This bottom plate is connected to a protective
cylinder around the centrifuge to avoid direct contact of the
entering liquids with the rotating centrifuge. As a model
reaction, the esterification of oleic acid with 1-butanol
catalyzed by a Rhizomucor miehei lipase was investigated
(Scheme 1).
[
22]
conditions (Figure 3).
The conversion reaches a maximum at 30 Hz. At a higher
rate of rotation the increased separatory power of the
centrifuge leads to a reduction in the volume of the mixed
phase in which the reaction takes place (Figure 4). The mixing
process becomes less efficient at reduced rotational speeds of
the centrifuge which results in larger average drop sizes in the
dispersed phase and thus to reduced rates of mass transfer and
conversion.
By using the established optimum conditions, biodiesel
À3
À1
was produced at a rate of 61 kg biodieselm min , which
À3
À1
compares well with the 42 kgm min reported for typical
batch processes. In addition, the current process is much
more efficient, since there is no distinct separation step, and
[
23]
[24]
cleaning of the reactor between batches can be omitted.
Next, the potential of the CCS to perform enzymatic
conversions was investigated. Most enzymes function opti-
mally in an aqueous environment, and as such are ideal
Scheme 1. Lipase-catalyzed esterification of oleic acid with 1-butanol.
3
906 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3905 –3908

Angewandte
Chemie
À1
The lipase-catalyzed reaction between oleic acid and
ethanol was already known, but we found that replacement of
(3.0 instead of 1.0 gL ) was performed (Figure 6). After
about 2 h, the conversion became reasonably steady—fluctu-
ating between 78 and 87%, with an average of 82% over the
experiment. Repeat runs showed good reproducibility.
[
25]
ethanol by 1-butanol led to much higher conversions. The
esterification of oleic acid with butanol using a crude extract
of penicillium coryophilum in a micellar system has also been
[
26]
described.
In batch mode, this reaction goes to full
conversion despite the large excess of water present. Presum-
ably, the reaction is driven by the lipophilicity of the reactants.
In a first series of experiments an organic phase consisting of a
À1
À1
mixture of oleic acid (0.6 molL ) and 1-butanol (0.9 molL )
in heptane was used. The aqueous phase consisted of a
À1
solution of R. miehei lipase (1 gL ) in a phosphate buffer at
pH 5.6. We first examined the effect of the flow rates of both
phases and the rotational speed of the centrifuge on the
conversion (Figure 5). Under these conditions, the highest
Figure 6. Lipase-catalyzed esterification of oleic acid with nBuOH in a
À1
CCS ([oleic acid]=0.6m, [nBuOH]=0.9m, [lipase]=3.0 gL
,
À1
Forg =Faq =6 mLmin , spinning rate=40 Hz, T=508C).
In the previous experiments, the enzyme solution was
used in a once-through mode. To boost the turnover number
of the enzyme, an experiment was performed in which the
enzymatic solution was continuously recycled in combination
with a partial recycling of the organic phase. With a 90%
recycling of the organic phase, a close to 80% conversion of
oleic acid into butyl oleate was achieved (Figure 7). The
reactor was run in this mode for a period of 13 h, which led to
a TON of 486 g butyl oleate per gram of enzyme. Although
there is some erosion of conversion over time, the stability of
the enzyme during this period can be considered remarkable
in view of the high speed of the centrifuge. The stability of the
Figure 5. Effect of flow rate (a) and rotational speed (b) on the
enzymatic esterification ([oleic acid]=0.6m, [BuOH]=0.9m, [lipa-
À1
se]=1 gL T=508C; a) rotational speed=40 Hz, Forg =F ; b)
aq
À1
Forg =F =10 mLmin ).
aq
steady-state conversion (70%) was found at a rotational
À1
speed of 40 Hz, and a flow rate of both phases of 6 mLmin .
The conversion shows a clear maximum with respect to the
flow rate of each phase. At lower flow rates, phase separation
in the CCS takes place more efficiently at the expense of the
mixed phase, and is comparable to the effect of high spinning
rates. At higher flow rates, the residence time in the CCS is
too short, and also leads to lower conversions. In this
À1
particular case, the optimum flow rate was 6 mLmin for
each phase. Similar to the biodiesel case, the rotational speed
of the centrifuge has a profound effect on the conversion of
oleic acid, and an optimum value was found to be 40 Hz.
Using the optimum settings determined above, a lipase-
catalyzed esterification reaction at a higher enzyme loading
Figure 7. Lipase-catalyzed esterification of oleic acid with nBuOH in a
CCS with full recycling of the water phase and 90% recycling of the
heptane phase.([oleic acid]=0.6m, [nBuOH]=0.9m, [Lipase] =
À1
À1
6.0 gL , Forg =F =6.2 mLmin , spinning rate=40 Hz, T=508C).
aq
Angew. Chem. Int. Ed. 2008, 47, 3905 –3908
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3907

Communications
enzyme remains a key issue for future developments. The
Bieler, P. Cretton, B. Zimmerman, Chem. Eng. Technol. 2005, 28,
3
1
2
18 – 323; e) L. Kiwi-Minsker, A. Renken, Catal. Today 2005,
10, 2 – 14; f) J. Kobayashi, Y. Mori, S. Kobayashi, Chem. Asian J.
006, 1, 22 – 35.
deactivation of the catalyst observed in Figure 7 may have a
number of different causes: The most likely hypothesis is that
an organic component which acts as an enzyme inhibitor
accumulates in the aqueous phase. This hypothesis is cur-
rently under investigation.
In conclusion, we have shown that it is possible and highly
advantageous to perform chemo- and biocatalytic conver-
sions continuously in a table-top centrifugal contactor sepa-
rator. Even in the current low-cost equipment, which can be
situated in a fume cupboard, it is already possible to produce
[
6] B. P. Mason, K. E. Price, J. L. Steinbacher, A. R. Bogdan, D. T.
McQuade, Chem. Rev. 2007, 107, 2300 – 2318.
7] a) http://isic2.epfl.ch/webdav/site/lcbp/shared/Renken/work-
shop/presentations/7-Poechlauer.pdf; b) P. Poechlauer, M. Kott-
haus, M. Vorbach, M. Zich, R. Marr (DSM), WO2006/125502,
[
2
006.
[
[
8] H. Wakami, J. Yoshida, Org. Process Res. Dev. 2005, 9, 787 – 791.
9] a) L. Z. Avila, G. M. Whitesides, J. Org. Chem. 1993, 58, 5508 –
5
4
512; b) M. Miyazaki, H. Maeda, Trends Biotechnol. 2006, 24,
63 – 470; c) T. Honda, M. Miyazaki, H. Nakamura, H. Maeda,
1
00 kg amounts of chemicals in a matter of days. With larger
volume CCSs commercially available, the way is opened for
the continuous production of fine chemicals on a ton scale
using two-phase catalysis.
Adv. Synth. Catal. 2006, 348, 2163 – 2171; d) M. S. Thomsen, P.
Pölt, B. Nidetzky, Chem. Commun. 2007, 2527 – 2529; e) T. C.
Logan, D. S. Clark, T. B. Stachowiak, F. Svec, J. M. J. FrØchet,
Anal. Chem. 2007, 79, 6592 – 6598.
[
10] a) S. Saaby, K. R. Knudsen, M. Ladlow, S. V. Ley, Chem.
Commun. 2005, 2909 – 2911; b) I. R. Baxendale, J. Deeley,
C. M. Griffiths-Jones, S. V. Ley, S. Saaby, G. K. Trammer,
Chem. Commun. 2006, 2566 – 2568; c) N. Nikbin, M. Ladlow,
S. V. Ley, Org. Process Res. Dev. 2007, 11, 458 – 462.
Experimental Section
Typical experimental procedure for the enzymatic esterification of
oleic acid with butanol: The experiments were performed in a CINC
[
27]
V-02 separator (also known as CS-50). Two Verder VL 500 control
peristaltic tube pumps equipped with a double pump head (3.2 ” 1.6 ”
[
11] a) H. R. Sahoo, J. G. Kralj, K. F. Jensen, Angew. Chem. 2007,
119, 5806 – 5810; Angew. Chem. Int. Ed. 2007, 46, 5704 – 5708;
8
R) were used to feed the CCS. In the case of the enzymatic reaction,
b) G. Jas, A. Kirschning, Chem. Eur. J. 2003, 9, 5708 – 5723.
12] A. Jha, The Guardian, July 31, 2003 (http://www.guardian.co.uk/
life/thisweek/story/0„1008930,00.html).
13] a) E. J. Vandamme, Biotechnologyof industrial antibiotics ,
Marcel Dekker, New York, 1984; b) Z. Likidis, K. Schꢀgerl,
Biotechnol. Bioeng. 1987, 30, 1032 – 1040.
the low-mix bottom plate was used. To operate the reactor at a
desired temperature, it was equipped with a jacket which was
connected to a temperature-controlled water bath that had an
accuracy of Æ 0.018C. The CCS was fed with pure heptane and pure
[
[
À1
water, both with a flow rate of 6 mLmin . The centrifuge was then
started (40 Hz, which corresponds to 2400 rpm) and the set-up was
allowed to equilibrate for a period of 1 h. At this point, the heptane
feed stream was replaced by the organic feed stream (oleic acid
[
14] N. Rꢀffer, U. Heidersdorf, I. Kretzers, G. A. Sprenger, L.
Raeven, R. Takors, Bioprocess Biosyst. Eng. 2004, 26, 239 – 248.
15] R. J. Taylor, I. May, A. L. Wallwork, I. S. Denniss, N. J. Hill, B. Y.
Galkin, B. Y. Zilberman, Y. S. Fedorov, J. Alloys Compd. 1998,
[
(0.6m) and 1-butanol (0.9m) in heptane). After equilibration for
1
0 min, the reaction in the CINC was started by replacing the water
271–273, 534 – 537.
stream with the aqueous feed stream (0.1m phosphate buffer, pH 5.6,
[
16] D. H. Meikrantz, L. L. Macaluso, H. W. Sams III, C. H. Chardin,
Jr., A. G. Federici (Coster Industries Nevada, Inc.),
US5,762,800, 1998.
À1
containing 1 gL of the lipase form Rhizomucor miehei). Samples
were taken at regular intervals and analyzed by GC.
[
[
[
17] a) S. J. Setford, Biotechnol. Tech. 1998, 12, 701 – 705; b) S. J.
Setford, J. Chem. Technol. Biotechnol. 1999, 74, 17 – 24.
18] J. G. de Vries, G. J. Kwant, H. J. Heeres (DSM IP Assets, bv),
WO 2007/031332.
19] C. de Bellefon, N. Tanchoux, S. Caravieilhes, P. Grenouillet, V.
Hessel, Angew. Chem. 2000, 112, 3584 – 3587; Angew. Chem. Int.
Ed. 2000, 39, 3442 – 3445.
Received: November 27, 2007
Revised: February 2, 2008
Published online: April 15, 2008
Keywords: biphasic catalysis · enzyme catalysis · esterification ·
.
fatty acids · sustainable chemistry
[
[
[
20] Y. ꢁnal, M. Lucas, P. Claus, Chem. Eng. Technol. 2005, 28, 972 –
9
78.
21] S. Liu, T. Fukuyama, M. Sato, I. Ryu, Org. Process Res. Dev.
004, 8, 477 – 481.
[
[
[
1] N. G. Anderson, Org. Process Res. Dev. 2001, 5, 613.
2
2] C. Ramshaw, Green Chem. 1999, 1, G15 – G17.
3] a) A. I. Stankiewicz, J. A. Moulijn, Chem. Eng. Prog. 2000,
22] a) F. Ma, M. A. Hanna, Bioresour. Technol. 1999, 70, 1 – 15;
b) J. M. Marchetti, V. U. Michel, A. F. Errazu, Renewable
Sustainable EnergyRev. 2007, 11, 1300 – 1311.
23] J. Van Gerpen, B. Shanks, R. Pruszko, D. Clements, G. Knothe,
Biodiesel Production Technology, NREL/SR-510-36244, 2004.
24] The pharmaceutical industry uses more solvents for reactor
cleaning than for the actual chemical production.
25] A. C. Oliveira, M. F. Rosa, M. R. Aires-Barros, J. M. S. Cabral, J.
Mol. Catal. B 2001, 11, 999 – 1005.
26] A. M. Baron, M. I. M. Sarquis, M. Baigori, D. A. Mitchell, N.
Krieger, J. Mol. Catal. B 2005, 34, 25 – 32.
96(1), 22 – 34; b) A. Stankiewic, J. A. Moulijn, Re-engineering
the chemical processing plant, Marcel Dekker, New York, 2004.
4] a) W. Ehrfeld, V. Hessel, H. Löwe, Microereactors: New
technologyfor modern chemistry , Wiley-VCH, Weinheim,
[
[
[
[
[
[
2000; b) V. Hessel, S. Hardt, H. Löwe, Chemical Micro Process
Engineering, Wiley-VCH, Weinheim, 2004; c) V. Hessel, H.
Löwe, Chem. Eng. Technol. 2005, 28, 267 – 284.
[
5] a) J.-C. Charpentier, Chem. Eng. Technol. 2005, 28, 255 – 258;
b) J. Yoshida, A. Nagaki, T. Iwasaki, S. Suga, Chem. Eng.
Technol. 2005, 28, 259 – 266; c) P. Watts, S. J. Haswell, Chem.
Eng. Technol. 2005, 28, 290 – 301; d) D. M. Roberge, L. Ducry, N.
27] See: www.auxill.nl/uk/cinc.separators.php.
3
908 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3905 –3908