Highly efficient continuous flow reactions using singlet oxygen as a "Green" reagent

Article abstract of DOI:10.1021/ol2017643

Described is a new method for the efficient in situ production of singlet oxygen in a simple continuous flow photochemical reactor. The extremely large interfacial area generated by running the biphasic mixture in a narrow channel at a high flow rate ensures high throughput as well as fast and efficient oxidation of various alkenes, 1,3-dienes, and thioethers on a preparative scale.

Full text of DOI:10.1021/ol2017643

ORGANIC
LETTERS
2011
Vol. 13, No. 19
5008–5011
Highly Efficient Continuous Flow
Reactions Using Singlet Oxygen as a
“Green” Reagent
ꢀ
Franc-ois Levesque and Peter H. Seeberger*
Max Planck Institute of Colloids and Interfaces, Department of Biomolecular Systems,
€
Am Mu€hlenberg 1, 14476 Potsdam, Germany, and Freie Universitat Berlin,
Institute for Chemistry and Biochemistry, Arnimallee 22, 14195 Berlin, Germany
peter.seeberger@mpikg.mpg.de
Received July 1, 2011
ABSTRACT
Described is a new method for the efficient in situ production of singlet oxygen in a simple continuous flow photochemical reactor. The extremely
large interfacial area generated by running the biphasic mixture in a narrow channel at a high flow rate ensures high throughput as well as fast and
efficient oxidation of various alkenes, 1,3-dienes, and thioethers on a preparative scale.
Molecular oxygen is an attractive reagent due to its
availability, low cost, and negligible environmental im-
pact. Singlet oxygen1 (1O2) is formed via dye-sensitized
photoexcitation of triplet oxygen (3O2) and facilitates
heteroatom oxidations, ene reactions, and [4 þ 2] and
[2 þ 2] cycloadditions.2 Consequently, 1O2 has been used
for the synthesis of natural products3 and fragrances.4
Widespread use of 1O2 in conventional batch systems has
been prevented by the need for specialized equipment to
produce the reagent, technical challenges associated with
scaling-up photochemical reactions,5 and the low rate of
mass transfer of oxygen gas into the solution.
surface-to-volume ratio that ensures efficient irradiation,7
and enable precise control over the reaction time to mini-
mize unwanted side reactions due to secondary photoche-
mical reactions.8 In addition, continuous flow reactors
improve safety as reactive intermediates9 are quenched10
or further transformed11 immediately after production.
Photochemical flow reactors have been explored12 for the
generation and use of 1O2. Although complete conversion
was achieved in a short residence time, the process suffered
from very low productivity,12a only 0.18 μmol/min,13
rendering the system inapplicable to use on a preparative
Continuous flow reactors allow for easy scale-up as
no change in reactor size is required,6 provide a large
(8) (a) For an example of a photochemically promoted hydroperoxide
€
Liebigs Ann. Chem. 1958, 618, 202. (b) For an example of photochemi-
cally promoted endoperoxide rearrangement, see: Maheshwari, K. K.;
De Mayo, P.; Wiegand, D. Can. J. Chem. 1970, 48, 3265.
(9) (a) Noller, D. C.; Mazurowski, S. J.; Linden, G. F.; De Leeuw,
F. J. G.; Mageli, O. L. Ind. Eng. Chem. 1964, 56 (12), 18. (b) Armitage,
J. B.; Strauss, H. W. Ind. Eng. Chem. 1964, 56 (12), 28.
rearrangement, see: Schenck, G. O.; Neumuller, O.-A.; Eisfeld, W. Justus
(1) Schenck, G. O. Angew. Chem. 1952, 64, 12.
(2) Schweitzer, C.; Schmidt, R. Chem. Rev. 2003, 103, 1685.
(3) Hoffmann, N. Chem. Rev. 2008, 108, 1052.
(4) Ohloff, G. Pure Appl. Chem. 1975, 43, 481.
ꢀ
(5) (a) Andre, J.-C.; Viriot, M.-L.; Villermaux, J. Pure Appl. Chem.
(10) For an example of endoperoxide quenching in a continuous flow
€
1986, 58, 907. (b) Pfoertner, K. H. J. Photochem. Photobiol. A: Chem.
1990, 51, 81.
system, see: Jahnisch, K.; Dingerdissen, U. Chem. Eng. Technol. 2005,
28, 426.
(6) (a) Thayer, A. M. Chem. Eng. News 2005, 83, 43. (b) O’Brien,
A. G.; Levesque, F.; Seeberger, P. H. Chem. Commun. 2010, 47, 2688.
(11) For examples of uses of peroxides, see: (a) Adam, W.; Braun,
M. A.; Griesbeck, A.; Lucchini, V.; Staab, E.; Will, B. J. Am. Chem. Soc.
1989, 111, 203. (b) Staben, S. T.; Linghu, X.; Toste, F. D. J. Am. Chem.
Soc. 2006, 128, 12658.
ꢀ
€
(7) Coyle, E. E.; Oelgemoller, M. Photochem. Photobiol. Sci. 2008, 7,
1313.
r
10.1021/ol2017643
Published on Web 08/31/2011
2011 American Chemical Society

scale. Other groups have tackled the problem of low O2(g)
mass transfer using supercritical carbon dioxide as a
solvent, though this mandates a highly specialized reaction
setup.14
Table 1. Influence of the Flow Rate on the 1O2 Oxidation of
Citronellola
Here, we report a simple continuous flow system that
renders reactions involving 1O2 practical for synthetic
chemists. Efficient mass transfer and sufficient irradiation
enable the straightforward production and use of 1O2 as a
reagent in several transformations such as oxidations and
cycloadditions.
Efficient oxidations are dependent on the solution con-
1
centration of O2, which in turn is proportional to the
flow rates
solution concentration of 3O2. Therefore, the productivity
of the oxidation depends on the rate of mass transfer
(d[3O2(sol)]/dt) of 3O2(g) into the solution. Based on Fick’s
law (eq 1) the rate of mass transfer is determined by the
liquid film transfer coefficient (KL), the specific surface
area of the solution (a), and the oxygen deficit within the
(μL/min)
concn of
solution
conversion
(%)b
productivity
entry
1 (M)
of 1
O2
(μmol/min)
1
2
3
0.10
0.10
0.10
26
13
130
65
31%
55%
0.81
0.72
0.65
6.5
33
g95%
solution ([3O2(sol) sat
]
ꢀ [3O2(sol)]).
a Reactions were performed in MeOH under the irradiation of a
green LED lamp (1 W), with rose bengal (5.1 mM) as the sensitizer.
b Determined by 1H NMR
d[3O2(sol)]=dt ¼ KLa([3O2(sol)
]
ꢀ [3O2(sol)]) ð1Þ
sat
When biphasic gasꢀliquid reactions are conducted at
high flow rates, the specific surface areas in continuous
flow reactors (up to 25300 m2/m3) can greatly exceed those
attained in conventional batch reactors (up to 2000 m2/
To increase material throughput, a system based on the
design of Booker-Milburn20 was constructed. Fluorinated
ethylene propylene (FEP) tubing was wrapped around a
Schenk photochemical reactor containing a 450 W med-
ium pressure mercury lamp that was cooled to 25 °C.
Substrate and oxygen were added via gastight syringes and
mixed using a polytetrafluoroethylene (PTFE) T-mixer.19
To overcome the photobleaching of rose bengal with this
more powerful lamp, tetraphenylporphyrin (TPP) was
instead used as a sensitizer.
First, the influence of oxygen stoichiometry on conver-
sion and productivity was investigated using a 10:1 v/v
ratio of oxygen gas to citronellol solution.21 Quantitative
conversion was achieved with as little as 1.6 equiv of
oxygen (Table 2, entry 3). A flow rate increase from 1.0
to 2.5 mL/min led to a 23% average rise of conversion
(Table 2, entries 2ꢀ5) as expected from Fick’s law.
mꢀ3 15 due to flow pattern effects.16 To date, synthetic
)
organic chemists have not taken full advantage of varia-
tions in the flow patterns of biphasic reactions.17
Initial reactions were conducted in a 78 μL silicon-glass
microreactor18 irradiated with an LED lamp19 to form 1O2
for thewell-documented photo-oxidation of citronellol (1),
a key intermediate en route to the valuable rose oxide fra-
grance3 (Table 1). Quantitative conversion was achieved
within a 2 min residence time using the photosensitizer
rose bengal (Table 1). The productivity of this system
(0.65 μmol/min) was limited by the small reactor volume
and the low lamp power that limited flow rates.
(12) (a) Wooton, R. C. R.; Fortt, R.; de Mello, A. J. Org. Process Res.
€
Dev. 2002, 6, 187. (b) Meyer, S.; Tietze, D.; Rau, S.; Schager, B.; Kreisel,
€
G. J. Photochem. Photobiol. A: Chem. 2007, 186, 248. (c) Jahnish, K.;
Dingerdissen, U. Chem. Eng. Technol. 2005, 28, 426.
(13) Productivity is defined as the amount of material, in mmol,
generated per minute. It is calculated by multiplying the flow rate by the
concentration of the substrate and the conversion.
(14) Bourne, R. A.; Han, X.; Poliakoff, M.; George, M. W. Angew.
Chem., Int. Ed. 2009, 48, 5322.
Table 2. Influence of Oxygen Stoichiometry and Flow Rate on
the 1O2 Oxidation of Citronellola
€
(15) (a) Ehrfeld, W.; Hessel, V.; Lowe, H. Microreactors: New
Technology for Modern Chemistry; Wiley-VCH: Weinheim, 2000; pp
230. (b) Luis, S. V., Garcia-Verdugo, E., Eds. Chemical Reactions and
Processes under Flow Conditions; RSC Publishing: Cambridge, 2010; p
202. (c) Wirth, T., Ed. Microreactors in Organic Synthesis and Catalysis;
Wiley-VCH: Weinheim, 2008; p 297. (d) Wegner, J.; Ceylan, S.; Kirschning,
A. Chem. Commun. 2011, 47, 4583.
flow rates
(mL/min)
concn of solution
equiv conversion productivity
entry 1 (M)
of 1
O2 of O2
(%)b
(μmol/min)
(16) (a) Yue, J.; Luo, L.; Gonthier, Y.; Chen., G.; Yuan, Q. Chem.
Eng. Sci. 2008, 63, 4189. (b) Niu, H.; Pan, L.; Su, H.; Wang, S. Ind. Eng.
Chem. Res. 2009, 48, 1621.
1
2
0.10
0.25
0.25
0.50
0.50
0.10
0.23
0.09
0.23
0.09
0.23
0.23
2.27 4.0
0.91 1.6
2.27 1.6
0.91 0.8
2.27 0.8
2.27 0.8
g95%
78%
23.0
17.6
57.5
25.7
92.0
15.2
3
g95%
57%
(17) (a) Kobayashi, J.; Mori, Y.; Okamoto, K.; Akiyama, R.; Ueno,
€
M.; Kitamori, T.; Kobayashi, S. Science 2004, 304, 1305. (b) Jahnisch,
4
€
K.; Baerns, M.; Hessel, V.; Ehrfeld, W.; Haverkamp, V.; Lowe, H.;
5
6c
80%
Wille, Ch.; Guver, A. J. Fluor. Chem. 2000, 105, 117.
66%
(18) (a) Ratner, D. M.; Murphy, E. R.; Jhunjhunwala, M.; Snyder,
D. A.; Jensen, K. F.; Seeberger, P. H. Chem. Commun. 2005, 578. (b)
Geyer, K.; Seeberger, P. H. Helv. Chim. Acta 2007, 90, 395. (c) Carrel,
a Reactions were performed in a 5 mL reactor in CHCl3 with TPP
(0.25 mM) as sensitizer. b Determined by 1H NMR. c Air was used
instead of pure oxygen.
ꢀ
F. R.; Geyer, K.; Codee, J. D. C.; Seeberger, P. H. Org. Lett. 2007, 9,
2285.
(19) See Supporting Information for description of the system.
Org. Lett., Vol. 13, No. 19, 2011
5009

The greater mass transfer allowed for a reduction in the
total amount of oxygen needed without affecting conver-
sion. The precise control of oxygen reduces the risk of
flammability associated with unreacted oxygen22 and con-
stitutes a major advantage compared to traditional batch
Schenck photoreactors where oxygen is constantly
bubbled through the solution. Good conversion was even
possible when air was used in place of pure oxygen
(Table 2, entry 6).
However, at concentrations above 2.0 mM conversion
decreased due to quenching of 1O2 by excess TPP (Figure 1).26
To further increase the productivity,13 much higher flow
rates were required. Use of an HPLC pump for substrate
delivery and connection of a mass flow controller at the gas
cylinder ensured accurate and consistent oxygen delivery
and enabled even greater mass transfer.19 In addition, the
total reactor volume was increased from 5 to 14 mL to
allow for faster flow rates while maintaining sufficient
irradiation. Conducting the reaction at a solution flow
rate of 5 mL/min and an O2 flow rate of 10 mL/min
resulted in up to 85% conversion (Table 3). Based on
Karayiannis’ flow pattern map,23 these conditions gener-
ate a slug flow pattern characterized by the formation of
large bubbles surrounded by a thin liquid film,24 thereby
increasing both the specific surface area and mass transfer.
Figure 1. Influence of TPP concentrations on conversion.
Table 4. Oxidation of Various Substrates by 1O2
a
Table 3. Oxidation of Citronellol by 1O2 under Pressurea
flow rates
(mL/min)
back
concn of solution
pressure conversion productivity
entry 1 (M)
of 1
O2
(bar)
(%)b
(mmol/min)
1
2
3
0.25
0.25
0.25
5
5
3
10
11
12
ꢀ
85%
g95%
81%
1.06
1.25
0.61
6.9
6.9
a Reactions were performed in a 14 mL reactor in CHCl3 with TPP
(0.25 mM) as sensitizer. b Determined by 1H NMR.
Considering Henry’s law, pressurizing a reactor should
increase the concentration of oxygen at saturation
([3O2(sol) sat
] , eq 1) and consequently improve mass transfer.
Indeed, pressurizing the system by connecting a 6.9 bar
back-pressure regulator25 resulted in quantitative conver-
sion (Table 3, entry 2). Attempts to increase the flow rate
beyond 5 mL/min yielded lower conversion, due to insuffi-
cient irradiation times.
The oxidation of citronellol (1) was shown to be depen-
dent on the sensitizer concentration. Increasing the TPP
concentration in general led to improved conversion.
(20) Hook, B. D. A.; Dohle, W.; Hirst, P. R.; Pickworth, M.; Berry,
M. B.; Booker-Milburn, K. I. J. Org. Chem. 2005, 70, 7558.
(21) The limitations of the setup would not allow a higher proportion
of oxygen.
(22) Mashuga, C. V.; Crowl, D. A. Process Saf. Prog. 1997, 17, 176.
(23) Chen, L.; Tian, Y. S.; Karayiannis, T. G. Int. J. Heat Mass
Transfer 2006, 49, 4220.
(24) For a schematic representation of the flow pattern, see Support-
ing Information.
(25) Use of a back-pressure regulator was incompatible with the
system using a syringe pump, because we needed to refill the syringes in
the middle of the run, leading to depressurization.
a Reactions were performed in a 14 mL reactor in CHCl3 with TPP
(2.0 mM) as sensitizer. Solution and oxygen were delivered at a flow rate
of 5 and 12 mL/min, respectively. Residence time: 0.8 min. b Determined
1
by H NMR. c The hydroperoxides formed were reduced with sodium
sulfite to the corresponding alcohols. d Solution and oxygen were
delivered at a flow rate of 1 and 10 mL/min, respectively. Residence
time: 1.3 min. e The hydroperoxide formed was reduced with triphenyl-
phosphine to the corresponding alcohol. f The product of photooxyda-
tion was treated with pyridine in a mixture of THF, acetone, and H2O to
give ketoacid 14.
5010
Org. Lett., Vol. 13, No. 19, 2011

Under optimal continuous flow conditions (Table 4,
entry 1), 27 mL of citronellol were oxidized in 1 h,
demonstrating the stability and scalabiltiy of the process.
Optimization of the reaction conditions allowed an un-
precendented productivity of 2.5 mmol/min, an improve-
ment by a factor of 3800 compared to the productivity
obtained with the silicon-glass microreactor (Table 1, entry 3).
The optimized reaction setup was readily applied to
other reactions using 1O2 (Table 4).27 Oxidation of
R-pinene (9) generated 10, a useful precursor to a dipho-
sphine chiral ligand.28 [4 þ 2] Cycloaddition to form
endoperoxide 12 (Table 4, entry 4) was successful. Endo-
peroxides are key functional groups in antimalarial
compounds29 and a precursor to 1,2,3,4-tetraols,30 a motif
present in several natural products.
In conclusion, we have developed a simple and inexpen-
sive continuous flow reactor system that enables synthetic
organic chemists to generate and use singlet oxygen for a
variety of transformations with minimal environmental
impact. The reactions produce up to 2.5 mmol of product
per min and benefitted from an increase in specific surface
area that allowed for faster flow rates. Continuous flow
processes hold great potential for the acceleration of
biphasic reactions involving oxygen and other gases.
Acknowledgment. The authors gratefully thank the
Max Planck Society for generous funding. F.L. is the
recipient of a postdoctoral fellowship from Fonds
ꢀ ꢀ
Quebecois de la recherche sur la nature et les technologies
(FQRNT).
The oxidation of furan 13 provided access to 14, a key
substrate to produce drug substances31 and DielsꢀAlder
reactions.32 The photo-oxygenation of furans to generate
butenolides is a key step in numerous natural product
syntheses.2 Finally, the setup was used to oxidize sulfide 15
to a mixture of the corresponding sulfoxide 16 and sulfone 17.
Supporting Information Available. Schematic diagrams
of reactor setup and flow patterns, experimental proce-
dures, spectroscopic data of compounds. This material is
available free of charge via the Internet at http://pubs.acs.
org.
(26) (a) Gollnick, K.; Griesbeck, A. Tetrahedron 1984, 17, 3235. (b)
Schmidt, R.; Brauer, H.-D. J. Am. Chem. Soc. 1987, 109, 6976. (c)
Tanielian, C.; Heinrich, G.; Entezami, A. J. Chem. Soc., Chem. Com-
mun. 1988, 1197.
(27) Products were purified by distillation under reduced pressure or
by adding water to precipitate TPP followed by filtration; no column
chromatography was used. See Supporting Information.
(28) Gavryushin, A.; Polborn, K.; Knochel, P. Tetrahedron: Asym-
metry 2004, 15, 2279.
(30) Valente, P.; Avery, T. D.; Taylor, D. K.; Tiekink, E. R. T. J. Org.
Chem. 2009, 74, 274.
(31) (a) Altman, M. D.; Ali, A.; Kumar Reddy, G. S. K.; Nalam,
M. N. L.; Anjum, S. G.; Cao, H.; Chellappan, S.; Kairys, V.; Fernandes,
M. X.; Gilson, M. K.; Schiffer, C. A.; Rana, T. M.; Tidor, B. J. Am.
Chem. Soc. 2008, 130, 6099. (b) Lohray, B. B.; Lohray, V. B.; Srivastava,
B. K.; Gupta, S.; Solanki, M.; Pandya, P.; Kapadnis, P. Bioorg. Med.
ꢁ
ꢀ
Chem. Lett. 2006, 16, 1557. (c) Etrych, T.; Sırova, M.; Starovoytova, L.;
´
ꢁ
ꢀ
Rı
´
hova, B.; Ulbrich, K. Mol. Pharmaceutics 2010, 7, 1015.
(32) (a) Murray, W. V.; Mishra, P. K. P.; Turchi, I. J.; Sawicka, D.;
Maden, A.; Sun, S. Tetrahedron 2003, 59, 8955. (b) Paulvannan, K. J.
Org. Chem. 2004, 69, 1207.
(29) Dechy-Cabaret, O.; Benoit-Vical, F.; Loup, C.; Robert, A.;
ꢀ
Gornitzka, H.; Bonhoure, A.; Vial, H.; Magnaval, J.-F.; Seguela, J.-P.;
ꢀ
Meunier, B. Chem.;Eur. J. 2004, 10, 1625.
Org. Lett., Vol. 13, No. 19, 2011
5011