Continuous organic synthesis in a spinning tube-in-tube reactor: TEMPO-catalyzed oxidation of alcohols by hypochlorite

Article abstract of DOI:10.1021/op800051t

Continuous production of aldehydes in high yields (≥90%) can be accomplished by feeding aqueous (sodium hypochlorite/sodium bicarbonate, pH 8.5) and organic (TEMPO/tetrabutylammonium bromide/primary alcohol/toluene or CH₂Cl₂) solutio

Full text of DOI:10.1021/op800051t

Organic Process Research & Development 2008, 12, 946–949
Continuous Organic Synthesis in a Spinning Tube-in-Tube Reactor: TEMPO-Catalyzed
Oxidation of Alcohols by Hypochlorite
Philip D. Hampton,* Matthew D. Whealon, Lisa M. Roberts, Andrew A. Yaeger, and Rick Boydson
Department of Chemistry, California State UniVersity Channel Islands, Camarillo, California 93012, U.S.A.
Abstract:
synthesis of HIV-1 protease inhibitor DMP 323.1 Perhaps the
most useful conditions for this conversion involve the 2,2,6,6-
tetramethyl-4-X-1-piperidinyloxy (4-X-TEMPO: X ) H, OH,
OCH3, NHC(dO)CH3)-catalyzed oxidation of alcohols by
sodium hypochlorite.1–4 Typical conditions involve the slow
addition of sodium hypochlorite buffered with sodium bicar-
bonate to a biphasic mixture of aqueous sodium or potassium
bromide and an organic phase consisting of the alcohol and
4-X-TEMPO (X ) H, OCH3) in toluene, toluene/ethyl acetate,
or CH2Cl2. Efficient temperature control and rapid mixing
(mechanical stirring at or above 1000 RPM) are both necessary
to avoid heat build-up resulting from this highly exothermic
reaction; slow addition of the hypochlorite is frequently used
to control the temperature in this reaction.1a,3c–f,4 Polymer-bound
TEMPO catalysts show comparable or higher reactivity in this
oxidation reaction and offer the advantage of facile catalyst
recovery.2c,5
As a result of the requirement of this reaction for efficient
mixing and heat removal, we became interested in examining
the potential of running this reaction in a rapid-mixing,
continuous-flow reactor, called a spinning tube-in-tube (STT)
reactor manufactured by Kreido Biofuels.6 This research
describes the first demonstration of the continuous production
of aldehydes in high yields using the TEMPO-catalyzed
oxidation of alcohols by hypochlorite in an STT reactor. While
this work was being conducted, Friz-Langhals reported that the
4-hydroxy-TEMPO-catalyzed oxidation of two alcohols, 2-bu-
toxyethanol and 2-hydroxyethyl 2-methylpropanoate, by sodium
hypochlorite could be performed continuously in a simple 3
mm diameter static mixer tube reactor resulting in the oxidation
of ꢀ-substituted alcohols to the corresponding aldehydes in
yields of 60-77%.4
Continuous production of aldehydes in high yields (g90%) can
be accomplished by feeding aqueous (sodium hypochlorite/sodium
bicarbonate, pH 8.5) and organic (TEMPO/tetrabutylammonium
bromide/primary alcohol/toluene or CH2Cl2) solutions through the
inlets of a spinning tube-in-tube reactor (STT reactor manufac-
tured by Kreido Biofuels [STT is a registered trademark of Kreido
Biofuels]) at rotor speeds of 4000-6000 RPM with residence times
as short as 1-2 min. This approach eliminates the need for slow
addition of the bleach reagent to control this exothermic reaction.
Introduction
The TEMPO-catalyzed oxidation of alcohols to aldehydes
and ketones is an important transformation for pharmaceutical
synthesis as a result of the selectivity of the reaction between
primary and secondary alcohols and the absence of malodorous
(Swern oxidation) or hazardous (chromium oxidation) side
products.1–5 An example of the application of this reaction in
pharmaceutical chemistry can be seen in the DuPont Merck
* Corresponding author. E-mail: Philip.Hampton@csuci.edu. Telephone: 805-
437-8869. Fax: 805-437-8895.
(1) (a) Pierce, M. E.; Harris, G. D.; Islam, Q.; Radesca, L. A.; Storace, L.;
Waltermire, R. E.; Wat, E.; Jadhav, P. K.; Emmett, G. C. J. Org. Chem.
1996, 61, 444. (b) Confalone, P. N.; Waltermire, R. E. In Process
Chemistry in the Pharmaceutical Industry; Gadamasetti, K. G., Ed.;
Marcell Dekker, Inc.: New York, NY, 1999; pp 205-208.
(2) For a review of this reaction, see: (a) de Nooy, A. E. J.; Besemer, A. C.;
van Bekkum, H. Synthesis 1996, 1153. (b) Adam, W.; Saha-Mo¨ller,
C. R.; Ganeshpure, P. A. Chem. ReV. 2001, 101, 3499. (c) Sheldon,
R. A.; Arends, I. W. C. E.; Brink, G.-J. TB.; Dijksman, A. Acc. Chem.
Res. 2002, 35, 774. (d) Sheldon, R. A.; Arends, I. W. C. E. AdV. Synth.
Catal. 2004, 346, 1051. (e) Sheldon, R. A.; Arends, I. W. C. E. J. Mol.
Catal. A: Chem. 2006, 251, 200.
(3) (a) Anelli, P. L.; Biffi, C.; Montanari, F.; Quici, S. J. Org. Chem. 1987,
52, 2559. (b) Anelli, P. L.; Banfi, S; Montanari, F.; Quici, S. J. Org.
Chem. 1989, 54, 2970. (c) Tong, G.; Perich, J. W.; Johns, R. B.
Tetrahedron Lett. 1990, 31, 3759. (d) Siedlecka, R.; Skarz˙ewski, J.;
Młochowski, J. Tetrahedron Lett. 1990, 31, 2177. (e) Leanna, M. R.;
Sowin, T. J.; Morton, H. E. Tetrahedron Lett. 1992, 33, 5029. (f)
Jurczak, J.; Gryko, D.; Kobrzycka, E.; Gruza, H.; Prokopowicz, P.
Tetrahedron 1998, 54, 6051. (g) Csuk, R.; Schmuck, K.; Scha¨fer, R.
Tetrahedron Lett. 2006, 47, 8769.
A picture of an STT reactor is shown in Figure 1 and a
schematic cross-section of an STT reactor is provided in Figure
2. The basic design of an STT reactor involves the introduction
of reactants into the gap (0.25-0.38 mm) between a rapidly
rotated (100-12000 RPM) solid or hollow rotor and a stationary
outer cylinder (referred to as the “stator”). Heat-exchangers
surround the stator and allow for efficient temperature control
of the contents of the reactor along the length of the stator.
With a rotor-stator gap of 0.25 mm and a rotor working length
of 10.1 cm, a Magellan STT reactor has a cavity volume
between the rotor and stator of 1.43 mL. Kreido Biofuels also
(4) Fritz-Langhals, E. Org. Process Res. DeV. 2005, 9, 577.
(5) (a) Dijskman, A.; Arends, I. W. C. E.; Sheldon, R. A. Chem. Commun.
2000, 271. (b) Dijskman, A.; Arends, I. W. C. E.; Sheldon, R. A. Synlett
2001, 102. (c) Dijskman, A.; Arends, I. W. C. E.; Sheldon, R. A. Special
Publ. Royal Soc. Chem. 2001, 266, 118. (d) Burkhardt, O.; Woeltinger,
J.; Karau, A.; Philippe, J.-L.; Henniges, H.; Bommarius, A.; Krimmer,
H.-P.; Drauz, K. U.S. Patent 6,451,943 B1, 2002. (e) Tanyeli, C.;
Gu¨mu¨s¸, A. Tetrahedron Lett. 2003, 44, 1639. (f) Ferreira, P.; Phillips,
E.; Rippon, D.; Tsang, S. C.; Hayes, W. J. Org. Chem. 2004, 69, 6851.
(g) Ferreira, P.; Hayes, W.; Phillips, E.; Rippon, D.; Tsang, S. C. Green
Chem. 2004, 6, 310. (h) Pozzi, G.; Cavazzini, M.; Quici, S.; Benaglia,
M.; Dell’ Anna, G. Org. Lett. 2004, 6, 441. (i) Gilhespy, M.; Lok, M.;
Baucherel, X. Catal. Today 2006, 117, 114. (j) Michaud, A.; Gingras,
G.; Morin, M.; Be´land, F.; Ciriminna, R.; Avnir, D.; Pagliaro, M. Org.
Process Res. DeV. 2007, 11, 766.
(6) STT, Magellan, and Innovator are registered trademarks of Kreido
Biofuels, formerly Holl Technologies Corporation and Kreido Labo-
ratories. (a) Holl, R.; Gulliver, E.; Hall, N.; Sojka, S. InnoV. Pharm.
Technol. 2003, 30, 116. (b) Cihonski, J. Pristine Processing 2004, 25.
(c) Holl, R. A.; McGrevy, A. N. U.S. Patent 6,471,392, 2002. (d) Holl,
R. A. U.S. Patent 6,752,529, 2004.
946
•
Vol. 12, No. 5, 2008 / Organic Process Research & Development
10.1021/op800051t CCC: $40.75
2008 American Chemical Society
Published on Web 08/07/2008

Scheme 1. Reaction conditions for the TEMPO-catalyzed
oxidation of alcohols by hypochlorite in an STT reactor
Figure 1. Picture of a Magellan STT reactor.
Figure 2. Cross-section of a Magellan STT reactor.
manufactures the much larger Innovator 200 type STT reactor
(reactor volume of 38.6 mL) and the STT 30G. They report
that the shear scales from smaller to larger STT reactors.6
quantitative conversion was observed for benzyl alcohol (1a),
2-phenylethanol (1c), and 1-octanol (1e) with short residence
times of 86-172 s (0.250-0.500 mL/min flow of the organic
and aqueous phases) and at a rotor speed of 4000-6000 RPM.8
Yields were lower with shorter residence times and with lower
rotor speeds (less than 4000 RPM). Longer residence times
resulted in somewhat lower yields, possibly due to oxidation
of the aldehyde to a carboxylic acid. There was no evidence
for the formation of esters under these conditions.4 Yields were
lower at rotor speeds above 6000 RPM or below 4000 RPM
and were somewhat greater in toluene compared with CH2Cl2.
(Table 1, entries 5-8 vs entries 1-4). An orange color of Br2
was observed in the organic phase in all of these experiments.
Diminished yields were obtained when Aliquat 336 (with a
chloride counterion) was used in place of Bu4NBr (Table 1,
entry 19). Omission of the quaternary ammonium salt and/or
TEMPO catalyst resulted in very low yields (Table 1, entries
20 and 21). The oxidation of benzyl alcohol was examined at
a higher concentration of bleach (13%) and in the absence of
an organic solvent (Table 1, entry 10). The yield of isolated
benzaldehyde was high (91-95%) under these concentrated
Results and Discussion
The conditions for the TEMPO-catalyzed oxidation of
alcohols by hypochlorite are shown in Scheme 1. Organic (0.2
M alcohol substrate, 2 mM TEMPO, and 10 mM Bu4NBr in
toluene or dichloromethane) and aqueous solutions (3-6%
commercial bleach diluted 1:1 with saturated sodium bicarbon-
ate) were introduced through PEEK tubing to the two inlets of
a Magellan STT reactor possessing a titanium rotor and stator
that were cooled to 0 °C.7 Syringe or HPLC pumps were used
to control the rates of flow of the two solutions into the reactor.
Samples were collected as a function of rotor speed and flow
rates. The organic phases in these samples were separated from
the aqueous phases, dried over anhydrous sodium sulfate, and
analyzed by GC/MS. An internal standard was used to quantify
the yields of aldehydes and unreacted alcohol substrate in the
reaction samples. Isolated yields were obtained by omitting the
internal standard and collecting a defined volume of organic
phase under a set of reaction conditions.
Table 1 provides aldehyde/ketone yields for the oxidation
of alcohols 1a-f by TEMPO/sodium hypochlorite. Essentially
(7) In preliminary experiments, attempts to use a Magellan STT reactor
with a stainless steel rotor and stator led to the formation of a thin
layer of rust on the surface of the rotor and the yield of aldehyde
decreased with time, possibly due to deactivation of the TEMPO
catalyst.
(8) The residence time for reactants in the reactor depends on the total
flow of reagents and the cavity volume of the reactor, i.e. a total flow
of 1 mL/min (0.5 mL/min on each of two inlet flows) yields a residence
time of 1.43 minutes for a Magellan reactor with a 1.43 mL volume.
Vol. 12, No. 5, 2008 / Organic Process Research & Development
•
947

Table 1. TEMPO-catalyzed oxidation of alcohols in an STT reactor as a function of alcohol structure, temperature, flow rates,
and rotor speed.
temp
(°C)
flow rate
organic (mL/min)
residence
time (s)
rotor speed
(RPM)
aldehyde/ketone
yieldb
entry
substratea
solvent
1
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1c
1c
1c
1c
1c
1c
1c
1ce
1cf
1cg
1c
1c
1d
1d
1d
1d
1e
1f
0
0
toluene
toluene
toluene
toluene
DCM
0.25
0.25
0.50
0.50
0.25
0.25
0.50
0.50
1.00
0.50
0.50
0.25
0.25
0.50
0.50
1.00
1.00
0.50
0.50
0.50
0.50
0.50
0.50
0.25
0.25
0.50
0.50
0.50
0.50
172
172
86
4000
6000
4000
6000
4000
6000
4000
6000
6000
4000
4000
4000
6000
4000
6000
4000
6000
6000
4000
4000
4000
4000
6000
4000
6000
4000
6000
4000
4000
91-93
95-96
90-91
98-99
86-90
81-86
86-90
81-86
89c
2
3
0
4
0
86
5
0
172
172
86
6
0
DCM
7
0
DCM
8
0
DCM
86
9
0
DCM
44
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
0
none
27
91-95d
90-92c
91-93
94-96
90-91
97-99
93-94
92-94
91c
47-49
14-22
11-12
65-70
69-70
75-81
75-89
49-56
71-76
94-96
70-82c
0
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
86
0
172
172
86
0
0
0
86
0
43
0
43
0
86
0
86
0
86
0
86
25
25
0
86
86
172
172
86
0
0
0
86
0
86
0
86
a An organic solution consisting of the alcohol substrate (0.2 M), chlorobenzene or dodecane (0.1 M), Bu4NBr (10 mM), and TEMPO (2 mM) dissolved in the indicated
organic solvent, and an aqueous solution consisting of a 1:1 mixture of saturated sodium bicarbonate and 3-6% bleach, were fed through the inlets of a Magellan STT
reactor at the indicated temperature, flow rate, and rotor RPM. b Unless otherwise noted, yields are based on GC/MS analysis with an internal standard (chlorobenzene or
dodecane). c Isolated yield after separation and drying of organic phase followed by chromatography on silica gel. d Reaction run without an organic solvent using 13%
bleach, as described in the Experimental Section. Yields were determined using chlorobenzene as an internal standard. e Aliquat 336 (chloride counterion) was used as the
phase-transfer agent (5%). f Quaternary ammonium salt phase-transfer agent was omitted from this reaction. g TEMPO and quaternary ammonium salt phase-transfer agent
were omitted from this reaction.
conditions, and there was no increase in the temperature of the
reactor output compared to the stator temperature setting (0 °C).
The yield of acetophenone from sec-phenethyl alcohol
increased with increasing residence time but was still not
complete even with a residence time of over 400 s (Table 1,
entries 24-27). The slower reactivity of sec-phenethyl alcohol
is consistent with literature reports of the slower reactivity of
secondary alcohols in the TEMPO-catalyzed oxidation reaction.3
Aldehydes were isolated in excellent yield from these
reactions in the absence of internal standard and after column
chromatography. Benzaldehyde (2a) could be obtained in 89%
isolated yield from benzyl alcohol (1a) with a residence time
of only 44 s in dichloromethane (Table 1, entry 9). Anisaldehyde
(2b) could be obtained in 90-92% yield in toluene with a
residence time of 86 s (Table 1, entry 11). Phenylacetaldehyde
(2c) was obtained from 2-phenylethanol (1c) in 91% isolated
yield with a residence time of 86 s in toluene (Table 1, entry
18). By comparison, a literature report indicated only an 87%
yield of phenylacetaldehyde from 2-phenylethanol when sodium
hypochlorite was added slowly over 45 min and with stirring
for 80 min after complete addition.3d Only 3 mmole of
phenylacetaldehyde was produced in the 125 min reaction time.
This same quantity could be generated using a Magellan STT
reactor in less than 2 min. Lower yields (70-82%, unoptimized)
were obtained of the BOC-protected aldehyde 2f; the same
aldehyde was prepared in a batch process in 90% yield (Table
1, entry 29).1b
Conclusion
These results show that when the TEMPO-catalyzed oxida-
tion of alcohols by hypochlorite is performed in an STT reactor,
there is an increased yield of aldehyde product, an increased
reaction rate, and an ability to produce aldehyde products
continuously. In principle, the scale of this reaction could be
increased by running the reaction in an Innovator 200 type STT
reactor which has a 25-fold larger reactor volume than the
Magellan STT reactor used in this study. Kreido has constructed
a much larger reactor for the production of biodiesel. We
hypothesize that the rapid mixing of reactants and efficient heat
transfer are responsible for the high yield and rapid reaction
observed in this research, and we postulate that highly exo-
thermic and mixing-limited (i.e., solid-forming) reactions may
be improved using an STT reactor. We are currently examining
whether the STT reactor can improve these and other pharma-
ceutically relevant reactions.
Experimental Section
Reagents were purchased from Fisher Scientific or Sigma-
Aldrich and used without additional purification. GC/MS
analysis was performed on a Shimadzu QP-5000 equipped with
948
•
Vol. 12, No. 5, 2008 / Organic Process Research & Development

an AOC-20i autosampler. Solutions of the alcohol (0.2 M),
TEMPO (2 mM), quaternary ammonium salt (10 mM), and
internal standard (chlorobenzene or dodecane 0.1 M) were
prepared fresh for each experiment by weighing the appropriate
quantity of the component into a volumetric flask and bringing
the volume to the mark with the solvent (toluene or CH2Cl2)
used in the experiment. Bleach solutions were similarly prepared
fresh for each experiment by diluting 6% commercial bleach
(approximately 0.88 M) with saturated sodium bicarbonate in
a volumetric flask. Isolated yields were obtained using similar
solutions but without the internal standard.
Solutions were introduced into a Magellan STT reactor using
Shimadzu LC10-ATVp HPLC pumps or a Harvard Apparatus
Pump 22 infusion/withdraw dual syringe pump.6 In all cases,
PEEK tubing was used to connect the pumps to the Magellan
STT reactor. When syringe pumps were used, PEEK plastic
tubing was connected to a luer lock adapter using HPLC and/
or Swagelok fittings. The temperature of the stator and the STT
bearings were controlled using Julabo F33 and F12 heater/
chillers containing propylene glycol as the coolant. In some
experiments, the solutions were precooled using simple heat
exchangers (not shown in Figures 1 or 2) which were set to
the same temperature as the heat exchangers on the stator; the
presence or absence of the heat exchangers did not impact the
yield. Whenever the flow rate or rotor speed was changed in
an experiment, the reactor was flushed with a total of at least
five reactor volumes (7-15 mL) to ensure equilibration of the
reactor and its contents.
duplicate on the GC/MS. Duplicate runs of each substrate were
performed with different reagent solutions. Standard procedures
were used to obtain response factors between the authentic
distilled aldehyde or ketone product, the alcohol substrate, and
chlorobenzene or dodecane (internal standard).
Isolated yields were obtained by collecting the mixture of
organic and aqueous phases in a separatory funnel. The contents
of the separatory funnel were periodically separated. After
approximately 30 mL of product stream was collected, the
magnesium sulfate was filtered, and 25 mL of the filtrate was
collected in a volumetric flask. After the solvent was removed
in Vacuo, the crude product was purified by column chroma-
tography on silica gel using petroleum ether/ethyl acetate as
the eluent.
In an attempt to determine whether this reaction could be
scaled to higher concentrations of alcohols and bleach, a reaction
was run without the use of an organic solvent and using 13%
commercial bleach. A mixture of benzyl alcohol (7.2 M),
chlorobenzene (1.1 M), TEMPO (0.02 M), and tetrabutylam-
monium bromide (0.067 M) was used as the organic phase,
and commercial 13% sodium hypochlorite (saturated with solid
sodium bicarbonate and filtered) was used as the aqueous phase.
A flow rate of 2.5 mL/min for the bleach solution and 0.5 mL/
min for the organic phase was used in this experiment. The
yield of benzaldehyde was determined using an internal
standard.
Acknowledgment
We acknowledge the loan of a Magellan STT reactor by
Kreido Biofuels and the California State University Program
for Education and Research in Biotechnology (CSUPERB) for
a grant in support of this research.
Mixtures of organic and aqueous phases flowing out of the
STT reactor were collected in a small vial and the organic phase
was quickly removed and dried over anhydrous sodium sulfate.
The filtered solutions were loaded into labeled 2 mL GC/MS
sample vials and diluted to approximately 2% of their original
concentration. For each set of reactor conditions, a minimum
of two samples were collected, and these samples were run in
Received for review September 22, 2005.
OP800051T
Vol. 12, No. 5, 2008 / Organic Process Research & Development
•
949