Continuous flow synthesis of ketones from carbon dioxide and organolithium or grignard reagents

Article abstract of DOI:10.1002/anie.201405014

We describe an efficient continuous flow synthesis of ketones from CO ₂ and organolithium or Grignard reagents that exhibits significant advantages over conventional batch conditions in suppressing undesired symmetric ketone and tertiary alcohol byproducts. We observed an unprecedented solvent-dependence of the organolithium reactivity, the key factor in governing selectivity during the flow process. A facile, telescoped three-step-one-flow process for the preparation of ketones in a modular fashion through the in-line generation of organometallic reagents is also established.

Full text of DOI:10.1002/anie.201405014

Angewandte
Chemie
DOI: 10.1002/anie.201405014
Synthetic Methodology
Continuous Flow Synthesis of Ketones from Carbon Dioxide and
Organolithium or Grignard Reagents**
Jie Wu, Xiaoqing Yang, Zhi He, Xianwen Mao, T. Alan Hatton,* and Timothy F. Jamison*
Abstract: We describe an efficient continuous flow synthesis of
ketones from CO2 and organolithium or Grignard reagents
that exhibits significant advantages over conventional batch
conditions in suppressing undesired symmetric ketone and
tertiary alcohol byproducts. We observed an unprecedented
solvent-dependence of the organolithium reactivity, the key
factor in governing selectivity during the flow process. A facile,
telescoped three-step–one-flow process for the preparation of
ketones in a modular fashion through the in-line generation of
organometallic reagents is also established.
reagents based on solvent selection was also observed in this
study. Notably, the stoichiometry of CO2 gas (1 equiv or
2.5 equiv, depending on the reactivity of the organolithium
reagent) can be easily controlled by a mass flow controller
(MFC) in the continuous system. Moreover, Grignard
reagents are also well-suited to our flow system, and are
complementary to the lithium reagents in that they broaden
the substrate scope of these reactions, which represents the
first example of the direct synthesis of ketones through
a Grignard/CO2/organolithium sequence. In addition, a facile,
modular, telescoped three-step–one-flow preparation of
ketones has been realized through the incorporation of the
in-line generation of functionalized organolithium or organo-
magnesium species.
K
etones are fundamental structural motifs in the synthesis of
pharmaceuticals, natural products, functional materials, and
agrochemicals.[1] As such, the efficient preparation of ketones
is of great significance to chemical, medicinal, and material
science. The most common methods of synthesizing ketones
are Friedel–Crafts acylation and oxidation of secondary
alcohols. Reactions of carboxylic acid derivatives, such as
nitriles, Weinreb amides, anhydrides, and acid chlorides with
metallic reagents have also emerged as important alternatives
for the synthesis of the corresponding ketones.[2] In this
context, the direct conversion of a carboxylic acid into
a ketone by using at least two equivalents of an organolithium
reagent has long been studied and is still receiving significant
attention.[3] An even more straightforward and convenient
strategy was reported by the Breitmaier group: the one-pot
synthesis of ketones from two organolithium reagents and
carbon dioxide.[4]
The reported one-pot batch process involves four steps
(Scheme 1): 1) carboxylation of organolithium R1Li with
CO2; 2) removal of excess CO2; 3) introduction of R2Li to
Herein, we report the development of an efficient process
to synthesize ketones from organolithium reagents and CO2
in continuous flow systems. Through this approach, a wide
range of substituted ketones can be easily obtained and the
formation of symmetric ketone or tertiary alcohol byproducts
is almost completely suppressed. An important and unprece-
dented trend pertaining to the reactivity of organolithium
Scheme 1. Ketone synthesis from CO2 and organolithium reagents.
afford a dilithium intermediate, which is believed to be stable
during the reaction process,[3c] and 4) quenching to accom-
plish the ketone product. This method is attractive since
ketones with almost any substitution pattern can be accessed
directly by the selection of the two organolithium compounds.
Another intriguing characteristic of this reaction scheme is
the utilization of CO2 as a renewable, nonflammable, non-
toxic, and cheap C1 feedstock, which satisfies the criteria of
“green” chemistry.[5] However, these reported batch condi-
tions suffer from several drawbacks. Firstly, carboxylation of
organolithium reagents with CO2 is frequently accompanied
by the generation of symmetric ketone byproducts. High
concentrations of CO2 (achieved by the slow addition of
organolithium reagents to solid dry ice) and low temperatures
are thus required.[6] Secondly, the necessary removal of excess
CO2 compromises utility and throughput, particularly on
large scale.[4] Thirdly, a long reaction time is generally needed
[*] Dr. J. Wu, Dr. Z. He, Prof. Dr. T. F. Jamison
Department of Chemistry, Massachusetts Institute of Technology
77 Massachusetts Ave., Cambridge, MA 02139 (USA)
E-mail: tfj@mit.edu
Dr. X. Yang, Dr. X. Mao, Prof. Dr. T. A. Hatton
Department of Chemical Engineering
Massachusetts Institute of Technology
77 Massachusetts Ave., Cambridge, MA 02139 (USA)
E-mail: tahatton@mit.edu
[**] Support for this work was provided by the Novartis-MIT Center for
Continuous Manufacturing. We are grateful to MIT Skoltech
program (postdoctoral fellowship to J.W.) and to Dr. Andrea Adamo
for the gift of the back-pressure regulator.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201405014.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
for complete consumption of the intermediate carboxylic
lithium salt.[3c] Additionally, a troublesome side reaction is the
simultaneous formation of tertiary alcohols due to the
relatively slow hydrolysis of unreacted organolithium
reagents.[3,4]
a simple solvent change from Et2O to THF, the symmetric
ketone 3 was almost completely suppressed when only
a stoichiometric amount of CO2 was used (entry 6). It is
notable that the observed lower reactivity of the organo-
lithium compound in THF compared to Et2O was unexpected
based on literature reports that the reaction of organolithium
reagents with 1,1-diphenylethylene was found to be substan-
tially slower in Et2O than in THF.[14]
Carboxylation of a variety of organolithium reagents has
been examined using stoichiometric CO2 in the flow system;
the results are summarized in the Supporting Information.
Depending on the reactivity of organolithium reagents
towards CO2, two complementary flow systems (A and B)
were developed to achieve ketone products (Table 2). Con-
tinuous-flow system A was suited for reactions using stoi-
chiometric amounts of CO2 and consisted of two pumps, an
MFC, two reactors made of PFA tubing, and a quenching bath
with strong magnetic stirring. If excess (approximately
2.5 equiv) CO2 was used to prevent symmetric ketone
Continuous flow synthesis has opened up new strategies
and tactics for synthetic organic chemistry.[7,8] Given that
continuous flow systems are generally more efficient than
conventional batch conditions for biphasic gas/liquid reac-
tions,[8,9] we envisioned that the flow synthesis of ketones
would be a means to overcome all the drawbacks associated
with the analogous batch synthesis (Scheme 1): 1) The excep-
tionally high interfacial contact between gas and liquid phases
should enhance the selectivity and reactivity of carboxylation.
2) The excess CO2, which has negative effects on the down-
stream reaction, can be easily removed by integration of
a mini-vacuum degasser. This step may even be avoidable by
using stoichiometric amounts of CO2, which is generally
challenging in batch reaction. 3) Enhanced mass and heat
transfer may increase the rate of dilithium intermediate
formation. 4) The dropwise addition of the final reaction
mixture into quenching reagents should help to prevent the
formation of tertiary alcohol byproducts.[10]
Table 2: Complementary flow systems for ketone synthesis from orga-
nolithiums and CO2.[a]
We initially tested the reaction of (trifluoromethyl)phe-
nyllithium with CO2 in the flow system to allow rapid
optimization of carboxylation conditions.[11] Due to the
instability of organolithium reagents, peristaltic pumps or
Asia pumps were used and the lithium species were stored at
low temperature (08C or À208C) during the flow process.[12]
A low concentrated organolithium–THF solution (0.1m) was
used to preclude blocking of the reactor by precipitation of
lithium salts. CO2 gas was metered into the system through an
MFC,[13] mixed with the lithium solution in a polyetherether-
ketone (PEEK) Y-mixer, and reacted downstream in a per-
fluoroalkoxyalkane (PFA) tubing reactor coil (0.4 mL, inner
diameter (ID) 0.04’’). As shown in Table 1, the flow system led
to improvements in the selectivity for acid product 2 (entries 4
and 5 versus entries 1–3). Even at ambient temperature with
one equiv CO2, the acid was formed in 19% yield after
quenching with a 1m HCl solution (entry 4). Gratifyingly, with
Table 1: Optimization of organolithium carboxylation.
Entry System CO2
T
Yield of
Yield of
Yield of
[8C]
2 [%]
3 [%]
4 [%]
1[a]
2[a]
3[a]
4[b]
5[b]
batch
batch
batch
flow
massive dry ice
bubbled gas
bubbled gas
ca. 1 equiv
ca. 3 equiv
ca. 1 equiv
0
0
64
0
43
19
66
96
30
89
50
70
33
4
6
11
7
10
0
À78
rt
flow
rt
rt
6[b,c] flow
0
[a] Yield of the isolated product. [b] 1.15 Equiv of CO2 was used. [c] The
yield was calculated from the crude 1H NMR spectrum of the unpurified
product mixture with trichloroethylene as the external standard.
Productivity: (0.04 mmolminÀ1)ꢀyield [%]. [d] Yield of the isolated
product based on 5 h continuous collection of products.
[a] Data from Ref. [6b]. [b] Reaction conducted in 0.1 M Et2O. Yield
based on the crude 1H NMR spectrum with trichloroethylene as the
external standard. [c] Utilizing THF as the solvent.
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
byproduct formation in the carboxylation step, a mini vacuum
degassing chamber (670 mL) was integrated into the flow
system before the addition of the second organolithium
reagent (Table 2, system B). Back-pressure regulation was not
needed since all reactions were conducted under ambient
pressure.
utilization of expensive 13CO2 or 14CO2 gas in the preparation
of isotope-labeled ketones, which are used as tracers for
modeling a variety of chemical and biochemical systems.[17]
However, some limitations on using organolithium
reagents in the flow systems were observed. Reactive
secondary lithium compounds were prone to react with
THF instead of CO2 or the carboxylic lithium salt intermedi-
ate.[18] Some heteroaromatic lithium species such as the 2-
pyridinyllithium decomposed quickly at room temperature
and blocked the flow system. We surmised that use of the
more stable Grignard reagents may solve these problems. As
the most available and popular reagents in organometallics,
Grignard reagents provide a wider substrate scope than
organolithium reagents and their use should extend the range
of ketone products beyond those accessible with organo-
lithium compounds.
The carboxylation of Grignard reagents with CO2, a classic
textbook transformation, has been used in flow systems for
the synthesis of carboxylic acids,[9c] avoiding the problematic
formation of symmetric ketone byproducts which was
observed with organolithium reagents.[19] However, the reac-
tion between Grignard reagents and carboxylic acids or
carboxylic lithium salts cannot be used directly for the
synthesis of ketones in batch conditions since low yields and
the simultaneous formation of alcohols are observed.[3c,20]
This undesired process likely occurs as a result of the
bimetallic intermediate, R1R2C(OMgX)(OM), being more
prone than the corresponding dilithium intermediate to form
the free ketone prior to hydrolysis.[3c]
Relying upon the greater kinetic control of the flow
system, we envisioned that the generated unstable bimetallic
intermediate would be quenched efficiently and faster than
the conversion to byproducts would occur. We were pleased
to discover that the flow procedure can be successfully
applied to synthesize diaryl (25), aryl–alkyl (5, 23, 24, 29),
vinyl–alkyl (26), and alkyl–alkyl (15, 27, 28) ketones in good
yields (Table 3). In all cases, the alcohol byproducts were
observed in less than 10%, probably due to the mild
conditions (room temperature) and short reaction time
(10 min). Although the reaction rate of lithium addition to
carboxyl magnesium was generally lower than that of
carboxyl lithium intermediates, the rate was significantly
enhanced by using only Et2O as the solvent.[21] More
importantly, ketones with secondary alkyl substituents (27–
29) can be produced effectively by this approach, which
complements the scheme based solely on organolithium
reagents. To the best of our knowledge, our flow synthesis
presents the first example of a direct synthesis of ketones
exhibiting a general substrate scope and good selectivity that
involves Grignard reagents without any additives.[22] We also
tested the Grignard reagent/CO2/Grignard reagent sequence
in the flow system (30). It required a higher temperature
(308C) and longer reaction time (30 min) to achieve a good
conversion (70% for the second step), however, the alcohol
byproduct was significantly increased (approximately 20%).
Having demonstrated the efficient generation of ketones
from organolithium or organomagnesium reagents, we real-
ized that the introduction of another flow component for the
on-demand, in-line generation of organolithium or organo-
PhLi (as R1Li) and nBuLi (as R2Li) were used initially to
optimize the solvent selection, flow rate, and quenching
method under continuous flow conditions.[15] THF was
selected as the solvent for R1Li to avoid symmetric ketone
byproducts, and Et2O was used as the solvent for R2Li to
increase the reaction rate. It was gratifying that full con-
version of the carboxylithium intermediate could be achieved
at room temperature within 10 min. The optimized exper-
imental procedure is summarized as follows: a peristaltic
pump was used to feed the PhLi–THF solution (0.1m, stored
at 08C) at 400 mLminÀ1, while the CO2 stream (ca. 2.5 sccm,
ca. 3 equiv) was metered into the system using an MFC. The
gas and liquid streams met at a Y-mixer, and the reaction
occurred at room temperature in the PFA sample loop
(0.4 mL, ID 0.04’’). Next, excess CO2 gas was separated from
the slug-flow regime through a mini vacuum degasser
(containing a thin porous membrane that selectively released
the gas under vacuum). The organic solution was then mixed
with the nBuLi–Et2O solution (0.3m, stored at rt,
200 mLminÀ1 flow rate) in a T-mixer and reacted in the
second PFA sample loop (6 mL, ID 0.062’’). Finally, the
product stream was dripped into a strongly stirred 1m HCl or
NH4Cl aqueous solution kept at 08C. If carboxylation was
achieved in high selectivity with stoichiometric CO2 (ca.
0.9 sccm), the vacuum degasser was not needed.
With the optimal conditions established, we examined the
generality of the flow synthesis of ketones using organo-
lithium reagents with a range of substitution patterns
(Table 2). Compared to the long reaction times required
under batch conditions, all flow reactions were accomplished
at ambient temperature with a residence time of less than
1 min in the first reactor and 10 min in the second. It was even
more intriguing that generally no symmetric ketone or
alcohol byproduct was observed, and in most cases, more
than 70% yield was isolated. Both electron-rich and electron-
deficient aromatic lithium compounds could be employed
efficiently in our protocol (5–9). Some heteroaromatic lithium
reagents appeared to be good coupling partners as well (10–
13). Primary aliphatic lithium species generally reacted
effectively in the flow system, no matter whether they were
introduced in the first or the second step (14–16, 18–21).
Functionalities such as alkene (19), alkyne (20), and silane
(21) were all tolerated. Secondary lithium reagents such as the
cyclopropyl lithium (22) were also suitable substrates. Nota-
bly, most of these ketone products are difficult to access
through Friedel–Crafts acylation[16] and these representative
examples illustrate the broad applicability of flow systems for
the synthesis of substituted ketones. To further demonstrate
the robustness of our systems, we were able to run experi-
ments for 5 h without any interruption to collect products 5
(1.71 g, 88% yield) and 9 (2.26 g, 82% yield). Importantly,
due to the significantly reduced amount of CO2 compared to
batch conditions, our flow method would be ideal for the
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Table 3: Flow synthesis of ketones starting with Grignard reagents.[a]
In summary, two complementary continuous flow systems
(using either stoichiometric or excess CO2) have been
developed for an efficient synthesis of ketones with a variety
of substitution patterns through an organolithium/CO2/orga-
nolithium sequence at ambient temperature and pressure.
The merits of flow systems over batch conditions provide
a much more convenient protocol with almost complete
suppression of the formation of undesired symmetric ketones
or tertiary alcohol byproducts. The observation that in this
reaction organolithium reagents were more reactive in Et2O
than in THF was unprecedented as far as we are aware, and
the first synthesis of ketones from the Grignard/CO2/organo-
lithium sequence was also developed in this study. By
integrating an in-line generation of organolithium or organo-
magnesium species, a telescoped one-flow process capable of
preparing ketones in a modular fashion from simple precur-
sors has thus been established.
[a] Productivity: (0.04 mmolminÀ1)ꢀyield [%]. [b] THF was used as
solvent for Grignard reagents. [c] Et2O was used as solvent for Grignard
reagents. [d] Ultrasonic irradiation was used to prevent clogging of
reactor. [e] Product volatility diminished the yield. [f] The TolMgBr/CO2/
PhMgBr sequence was used, see the Supporting Information.
Received: May 6, 2014
Published online: && &&, &&&&
Keywords: CO2 conversion · gas/liquid continuous flow ·
.
Grignard reactions · ketones · organic lithium reactivity
magnesium intermediates before the carboxylation with CO2
would provide a more efficient overall process to prepare
ketone products.[23,24] We thus examined the organometallic
generation/carboxylation/lithium addition sequence in an
integrated flow system. As shown in Table 4, the continuous
flow system was effective for exothermic reactions such as
lithium–halogen exchange, lithiation, and magnesium–halo-
gen exchange due to fast mixing and efficient heat transfer.
Thus this three-step continuous process proved to be a more
convenient and effective protocol for the smooth generation
of ketones.
[1] For selected examples, see: a) H. Surgurg, J. Panten, Common
Fragrance and Flavor Materials, 5edth edWiley-VCH, Wein-
heim, 2006; b) M. Van de Putte, T. Roskams, J. R. Vandenheede,
P. Agostinis, P. A. M. deWitte, Br. J. Cancer 2005, 92, 1406; c) Y.
Nishizuka, Nature 1988, 334, 661; d) A. M. Prince, D. Pascual, D.
Meruelo, L. Liebes, Y. Mazur, E. Dubovi, M. Mandel, G. Lavie,
Photochem. Photobiol. 2000, 71, 188.
[2] a) R. C. Larock, Comprehensive Organic Transformations,
VCH, New York, 1989, p. 685; b) J. March, Advanced Organic
Chemistry, 3rd ed., Wiley, New York, 1985, pp. 433 – 435 and 824 –
827.
[3] For selected examples, see: a) H. Gilman, P. R. van Ess, J. Am.
Chem. Soc. 1933, 55, 1258; b) K. Mislow, J. Brenner, J. Am.
Chem. Soc. 1953, 75, 2318; c) R. Levine, M. J. Karten, W. M.
Kadunce, J. Org. Chem. 1975, 40, 1770; d) W. D. Lubell, H.
Rapoport, J. Am. Chem. Soc. 1988, 110, 7447; e) S. Daum, F.
Erdmann, G. Fischer, B. F. Lacroix, A. Hessamian-Alinejad, S.
Houben, W. Frank, M. Braun, Angew. Chem. 2006, 118, 7615;
Angew. Chem. Int. Ed. 2006, 45, 7454; f) D. R. Genna, G. H.
Posner, Org. Lett. 2011, 13, 5358.
Table 4: Integrated three-step continuous-flow synthesis of ketones.
[4] G. Zadel, E. Breitmaier, Angew. Chem. 1992, 104, 1070; Angew.
Chem. Int. Ed. Engl. 1992, 31, 1035.
[5] For recent reviews on the conversion of CO2 to useful products,
see: a) T. Sakakura, J. Choi, H. Yasuda, Chem. Rev. 2007, 107,
2365; b) M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann,
F. E. Kꢀhn, Angew. Chem. 2011, 123, 8662; Angew. Chem. Int.
Ed. 2011, 50, 8510; c) I. Omae, Coord. Chem. Rev. 2012, 256,
1384.
Entry
R1X
Organometallic
compound
CO2
T
t
Yield
(equiv) [8C] [min] [%][a]
[6] Selected examples: a) Ref. [3a]; b) E. J. Soloski, C. Tamborski, J.
Organomet. Chem. 1978, 157, 373; c) M. A. Peterson, J. R.
Mitchell, J. Org. Chem. 1997, 62, 8237; d) J. C. Anderson, S.
Broughton, Synthesis 2001, 2379.
[7] Recent reviews on continuous-flow chemistry: a) J. C. Pastre,
D. L. Browne, S. V. Ley, Chem. Soc. Rev. 2013, 42, 8849; b) J.
Yoshida, Y. Takahashi, A. Nagaki, Chem. Commun. 2013, 49,
9896; c) D. T. McQuade, P. H. Seeberger, J. Org. Chem. 2013, 78,
6384; d) N. G. Anderson, Org. Process Res. Dev. 2012, 16, 852;
e) R. L. Hartman, J. P. McMullen, K. F. Jensen, Angew. Chem.
2011, 123, 7642; Angew. Chem. Int. Ed. 2011, 50, 7502; f) D.
1
2
3
nBuLi (1.05 equiv)
nBuLi (1.05 equiv)
1.05
1.05
1.05
0
1
84
79
72
rt
rt
15
15
iPrMgCl·LiCl
(1.05 equiv)
[a] Yield of the isolated product. Productivity:
(0.04 mmolminÀ1)ꢀyield [%].
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Webb, T. F. Jamison, Chem. Sci. 2010, 1, 675; g) T. Razzaq, C. O.
Kappe, Chem. Asian J. 2010, 5, 1274.
[17] For selected recent examples of the utilization of 13C-labeled
ketones, see: a) M. A. Evans, J. P. Morken, J. Am. Chem. Soc.
2002, 124, 9020; b) Y. Tsuji, M. M. Oteva, T. L. Amyes, J. P.
Richard, Org. Lett. 2004, 6, 3633; c) O. Khdour, E. Skibo, Org.
Biomol. Chem. 2009, 7, 2140; d) P. Lehwald, M. Richter, C.
Rçhr, H. Liu, M. Mꢀller, Angew. Chem. 2010, 122, 2439; Angew.
Chem. Int. Ed. 2010, 49, 2389.
[8] Recent reports from our laboratory: a) Z. He, T. F. Jamison,
Angew. Chem. 2014, 126, 3421; Angew. Chem. Int. Ed. 2014, 53,
3353; b) J. Wu, J. A. Kozak, F. Simeon, T. A. Hatton, T. F.
Jamison, Chem. Sci. 2014, 5, 1227; c) J. A. Kozak, J. Wu, X. Su, F.
Simeon, T. A. Hatton, T. F. Jamison, J. Am. Chem. Soc. 2013,
135, 18497; d) D. R. Snead, T. F. Jamison, Chem. Sci. 2013, 4,
2822; e) Y. Zhang, M. L. Blackman, A. B. Leduc, T. F. Jamison,
Angew. Chem. 2013, 125, 4345; Angew. Chem. Int. Ed. 2013, 52,
4251.
[18] For half-lives of organolithium reagents in common ethereal
solvents, see: P. Stanetty, M. D. Mihovilovic, J. Org. Chem. 1997,
62, 1514; For more details of observed results, see the Supporting
Information.
[9] For selected recent examples of gas/liquid continuous flow, see:
a) J. Kobayashi, Y. Mori, K. Okamoto, R. Akiama, M. Ueno, T.
Kitamori, S. Kobayashi, Science 2004, 304, 1305; b) M. T.
Rahman, T. Fukuyama, N. Kamata, M. Sato, I. Ryu, Chem.
Commun. 2006, 2236; c) A. Polyzos, M. OꢁBrien, T. P. Petersen,
I. R. Baxendale, S. V. Ley, Angew. Chem. 2011, 123, 1222;
Angew. Chem. Int. Ed. 2011, 50, 1190; d) F. Lꢂvesque, P. H.
Seeberger, Org. Lett. 2011, 13, 5008; e) B. Pieber, S. T. Martinez,
D. Cantillo, C. O. Kappe, Angew. Chem. 2013, 125, 10431;
Angew. Chem. Int. Ed. 2013, 52, 10241; f) D. B. Ushakov, K.
Gilmore, D. Kopetzki, D. T. McQuade, P. H. Seeberger, Angew.
Chem. 2014, 126, 568; Angew. Chem. Int. Ed. 2014, 53, 557.
[10] House and Bare have shown that quenching the batch reaction
by removing aliquots from reaction mixtures and adding them
dropwise to ice and dilute hydrochloric acid with vigorous
stirring can effectively avoid alcohol byproduct formation, see:
H. O. House, T. M. Bare, J. Org. Chem. 1968, 33, 943.
[11] For recent reports on lithiation–carboxylation with CO2 in flow,
see: a) B. Pieber, T. Glasnov, C. O. Kappe, RSC Adv. 2014, 4,
13430; b) A. Nagaki, Y. Takahashi, J.-i. Yoshida, Chem. Eur. J.
2014, DOI: 10.1002/chem.201402520.
[12] For the generation of organolithium compounds, organolithium
stability test, and storage, see the Supporting Information.
[13] The mass flow controller was calibrated before using.
[14] R. Waack, M. A. Doran, J. Organomet. Chem. 1971, 29, 329.
[15] For condition optimization, see the Supporting Information.
[16] L. Kꢀrti, B. Czakꢃ, Strategic Applications of Named Reactions in
Organic Synthesis, Elsevier, Amsterdam, 2005, p. 176.
[19] V. Grignard, Hebd. Seances Acad. Sci. 1900, 130, 1322.
[20] For the synthesis of ketones from Grignard reagents and
carboxylic acids in the presence of additives, see: a) V. Fianda-
nese, G. Marchese, L. Ronzini, Tetrahedron Lett. 1983, 24, 3677;
b) T. Fujisawa, T. Mori, T. Sato, Tetrahedron Lett. 1982, 23, 5059.
[21] Et2O can be used as the solvent for the carboxylation with
Grignard reagents without formation of symmetric ketone
byproducts.
[22] a) For a report on the generation of ketones from a-amino
carboxylithium and Grignard reagents, see: C. G. Knudsen, H.
Rapoport, J. Org. Chem. 1983, 48, 2260; b) For a report on the
generation of ketones from carboxylic acids using Grignard
reagents with the aid of N-protecting groups, see: P. Zhang, E. A.
Terefenko, J. Slavin, Tetrahedron Lett. 2001, 42, 2097.
[23] For selected examples involving continuous-flow generation of
organolithium reagents, see: a) A. Nagaki, H. Kim, J. Yoshida,
Angew. Chem. 2008, 120, 7951; Angew. Chem. Int. Ed. 2008, 47,
7833; b) H. Kim, A. Nagaki, J. Yoshida, Nat. Commun. 2011, 2,
264; c) A. Nagaki, Y. Uesugi, H. Kim, J. Yoshida, Chem. Asian J.
2013, 8, 705; d) P. R. D. Murray, D. L. Browne, J. C. Pastre, C.
Butters, D. Guthrie, S. V. Ley, Org. Process Res. Dev. 2013, 15,
1192.
[24] For examples involving the continuous-flow generation of
Grignard reagents by magnesium–halogen exchange, see: a) T.
Brodmann, P. Koos, A. Metzger, P. Knochel, S. V. Ley, Org.
Process Res. Dev. 2012, 16, 1102; b) T. Tricotet, D. F. OꢁShea,
Chem. Eur. J. 2010, 16, 6678; c) H. Wakami, J. Yoshida, Org.
Process Res. Dev. 2005, 9, 787.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Synthetic Methodology
J. Wu, X. Yang, Z. He, X. Mao,
T. A. Hatton,*
T. F. Jamison*
&&&& — &&&&
Continuous Flow Synthesis of Ketones
from Carbon Dioxide and Organolithium
or Grignard Reagents
Known chemistry in flow: The continuous
flow synthesis of ketones using CO2 and
organolithium or Grignard reagents
exhibits significant advantages over con-
ventional batch conditions. Undesired
symmetric ketone and alcohol byprod-
ucts are suppressed and an unprece-
dented solvent-dependence of the orga-
nolithium reactivity was used to achieve
the desired selectivity.
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!