Can Accelerated Reactions in Droplets Guide Chemistry at Scale?

Article abstract of DOI:10.1002/ejoc.201601270

Mass spectrometry (MS) is used to monitor chemical reactions in droplets. In almost all cases, such reactions are accelerated relative to the corresponding reactions in bulk, even after correction for concentration effects, and they serve to predict the likely success of scaled-up reactions performed in microfluidic systems. The particular chemical targets used in these test studies are diazepam, atropine and diphenhydramine. In addition to a yes/no prediction of whether scaled-up reaction is possible, in some cases valuable information was obtained that helped in optimization of reaction conditions, minimization of by-products, and choice of catalyst. In a variant on the spray-based charged droplet experiment, the Leidenfrost effect was used to generate larger, uncharged droplets and the same reactions were studied in this medium. These reactions were also accelerated but to smaller extents than in microdroplets, and they gave results that correspond even more closely to microfluidics data. The fact that MS was also used for online reaction monitoring in the microfluidic systems further enhances the potential role of MS in exploratory organic synthesis.

Full text of DOI:10.1002/ejoc.201601270

A Journal of
Accepted Article
Title: Can Accelerated Reactions in Droplets Guide Chemistry at Scale?
Authors: Michael Wleklinski; Caitlin Falcone; Bradley Loren; Zinia Jaman; Kiran Iyer;
H. Samuel Ewan; Seok-Hee Hyun; David Thompson; R Graham Cooks
This manuscript has been accepted after peer review and the authors have elected
to post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early View as
soon as possible and may be different to this Accepted Article as a result of editing.
Readers should obtain the VoR from the journal website shown below when it is pu-
blished to ensure accuracy of information. The authors are responsible for the content
of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201601270
Link to VoR: http://dx.doi.org/10.1002/ejoc.201601270
Supported by

European Journal of Organic Chemistry
10.1002/ejoc.201601270
COMMUNICATION
Can Accelerated Reactions in Droplets Guide Chemistry at Scale?
Michael Wleklinski, Caitlin E. Falcone, Bradley P. Loren, Zinia Jaman, Kiran Iyer, H. Samuel Ewan,
Seok-Hee Hyun, David H. Thompson*, and R. Graham Cooks*
Abstract: Mass spectrometry (MS) is used to follow chemical
reactions in droplets. In almost all cases, such reactions are
accelerated relative to the corresponding reactions in bulk, even
after correction for concentration effects, and they serve to predict
the likely success of scaled up reactions performed in microfluidic
flow systems. The particular chemical targets used in these test
studies are diazepam, atropine and diphenhydramine. In addition to
a yes/no prediction of whether scaled up reaction is possible, in
some cases valuable information was obtained which helped in
optimization of reaction conditions, minimization of by-products and
choice of catalyst. In a variant on the spray-based charged droplet
experiment, the Leidenfrost effect was used to generate larger,
uncharged droplets and the same reactions were studied in this
medium. These reactions were also accelerated but to smaller
extents than in microdroplets and they gave results that
corresponded even more closely to microfluidics data. The fact that
MS was also used for on-line reaction monitoring in the microfluidic
systems further enhances the potential role of MS in exploratory
organic synthesis.
differ from ESI based droplets in that they are i) larger ii) net
neutral and iii) involve elevated temperatures. The difference in
droplet size means that larger amounts of reagent can be
studied, but the surface/volume ratio is greatly decreased. The
measured reaction acceleration factors for three previously
studied Leidenfrost reactions, hydrazone formation, Katritsky
pyrylium to pyridinium conversion and 5C] laisen-Schmidt
[
condensation, are about an order of magnitude.
Note that we do not expect to be able to transfer optimized
conditions exactly from the droplet scale to the microfluidics
scale, in part because of uncertainty about the origins of
acceleration effects in the two systems. We do expect that
these optimized conditions will represent a starting point for
efficient optimization of the conditions in the microfluidics
reactor. We also expect that information on reaction
intermediates and mechanisms might be acquired from the
study of droplet reactions. This information is already being
obtained in experiments in which the degree of desolvation of
the initial droplets is varied by changing the distance that the
[1b]
droplets travel before analysis. Information obtained from the
droplet reactor on experimental parameters including solvent,
catalyst, pH, etc. is also readily acquired and we examine how
transferable this information is in optimizing the microfluidics
flow reactor.
The identification of suitable pathways to target molecules is
just one step towards an online, automated flow-through
synthesizer, for which the groundwork has been laid by several
groups. Notable are the mole-scale, end-to-end, continuous
Introduction
This study is part of a larger project, the overall goal of which is
to develop an automated scalable and continuous synthesis
system. A key objective is to test possible synthetic pathways
quickly on a small scale seeking a go/no-go result. We “spot
test” particular routes using a chemical pruning step which
employs reaction acceleration in droplets with independent mass
spectrometric analysis. A simple yes/no answer to product or (in
multistep reactions) intermediate formation is sought using the
charged droplet reactor. We use electrospray (ESI) for both
synthesis and analysis with careful control of parameters to
[
6]
manufacturing pilot plant developed by MIT/Novartis,
the
refrigerator sized, reconfigurable, on demand synthesizer of
[
7]
pharmaceuticals of MIT,
the nanomole-scale robotic high-
[8]
throughput synthesizer of Merck and the automated synthesis
laboratory of Eli Lilly. The mole-scale MIT synthesizer was
used to produce aliskiren hemifumarate in tablet form, from a
[
9]
[
6]
complex intermediate in a continuous fashion. The Merck
nanomole system was used to screen 1,500 reactions per day to
identify potential candidate reactions for large scale synthesis.
The Lilly system combined automated synthesis with analysis
performed remotely controlled in real-time, to produce gram-
scale products.
[
1]
avoid unwanted reaction during analysis.
Charged microdroplets are produced by ESI. It is known
that reaction rates increase as the solvent evaporates because
of changes in concentration, pH, surface/volume ratios, and
interfacial effects.
remarkably large.
accelerated reactions in droplets, including evidence that partial
solvation of reagents at interfaces contributes to the orders of
magnitude reaction rate acceleration that can be seen.
The main question underlying this study is whether droplet
reactions may be used to predict chemical reactivity in flow
chemistry systems, in particular in microfluidics. The mechanism
for acceleration observed in microdroplets is certainly different
from that in microfluidics in that evaporation is not significant in
microfluidics; however, interfacial effects may still play an
[
1b, 1c, 2]
The acceleration factors can be
recent review covers the topic of
[
3]
A
[
1a]
The
[10]
important role especially in droplet microfluidics. The speed of
data acquisition in droplets makes this approach attractive. Note
that false negatives (predict no reaction, but reaction can be
observed) is not expected to be a serious problem because
there are usually many available routes to test. On the other
hand, a false positive result will lead to wasted effort in seeking
an analogous flow reaction. Note, too, that use of droplets for a
simple yes/no regarding occurrence of reaction represents only
one level of enquiry, even though it is the most important one.
As will be seen in the results now to be discussed, information
on reaction conditions is also obtained, although the quality of
this information remains to be evaluated further by studying
more cases.
hypothesis investigated here is that the accelerated reactions
that occur in droplets might assist in rapidly evaluating reactivity
in microfluidic systems.
A second method of producing droplets is based on the
[4]
Leidenfrost effect. It has recently been shown that accelerated
[5]
organic reactions occur in Leidenfrost droplets. These droplets
[*]
M. Wleklinski, C. E. Falcone, B. P. Loren, Z. Jaman, K. Iyer, H. S.
Ewan, Dr. S. Hyun, Prof. D. H. Thompson*, and Prof. R. G. Cooks*
Department of Chemistry, Purdue University
560 Oval Drive, West Lafayette, IN (USA)
E-mail: cooks@purdue.edu
E-mail: davethom@purdue.edu
Homepage: http://aston.chem.purdue.edu
Results and Discussion
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601270
COMMUNICATION
The charged droplet and microfluidic based synthesis of
amide 3, generated by N-acylation of 1 with the 2-haloacetyl
chloride 2, was examined due to its importance as a synthetic
step in the pathway to diazepam (Scheme 1). An electrospray
droplet reactor was used to evaluate potential solvents for the
amount of S
N
2 reaction product was observed, especially at
higher temperatures (100 and 150°C). The presence of a by-
product (ion m/z 288) arising from the initial S 2 reaction product
was confirmed by NMR and MS/MS (data not shown). The
droplet reactor predicted formation of the desired intermediate,
which was observed in flow. However, under higher temperature
N
N
conditions in flow, the proportion of S 2 product increased. This
difference can be explained by the fact that evaporative cooling
occurs during flight in the droplet reaction at room temperature,
[
11]
thus reducing the reaction temperature.
Nonetheless, the
droplet reaction demonstrated reaction feasibility and showed
that specific solvents (ACN, toluene) are better than others a
fact reiterated under microfluidic conditions for this
transformation.
Scheme 1. Reaction of 1 and 2 (X = Cl, Br) to form amide 3.
The ability of droplet reactivity to guide chemistry at scale
was investigated for another important drug, diphenhydramine.
The flow based synthesis of diphenhydramine was
demonstrated by the reaction of dimethylaminoethanol (DMAE)
N-acylation reaction using chloroacetyl chloride. Offline charged
droplet reactions were performed using a mixture of 1 and 2 (X =
Cl) in various solvents, and conversion to product as analyzed
by ESI MS was compared to a 30-minute batch reaction (Figure
S1). Before analysis, samples are quenched in order to ensure
no further reaction by diluting the collected product into the
solvent used in the prior step. The results indicate that there is
significant acceleration of the reaction when the solvent is DMF,
ACN, or toluene. Acceleration in microdroplets is associated with
evaporation and is proposed to be due, in part, to intrinsic rate
[
12]
with chlorodiphenylmethane.
This synthesis featured 100%
atom economy; however, chlorodiphenylmethane is an
expensive starting material, which motivated an effort to develop
a
more
cost
effective
process
by
replacing
chlorodiphenylmethane using a commodity starting material,
benzhydrol, 4. To synthesize diphenhydramine 5, benzhydrol (4)
was converted to the corresponding mesyl ester, which was
subsequently treated with DMAE to produce 5 (Scheme 2). The
droplet reactor synthesis of 5 was demonstrated by performing
two sequential charged droplet reactions in either a toluene or
acetonitrile solvent system. First benzhydrol and mesyl chloride
were sprayed to produce the mesyl ester. This material was
recovered and re-dissolved before
[
2c]
acceleration at the interface. These initial results encouraged
a more extensive reaction screening, where the effect of the
chosen 2-haloacetyl chloride (Cl or Br) and the solvent (ACN,
toluene) was investigated using a droplet reactor. Interestingly,
the droplet reactor data indicated nearly complete conversion of
starting materials to product for both the chloro and bromo
starting materials in acetonitrile (Figure 1) and in toluene (Figure
S2). Starting material 1, is observed at m/z 246 and product, 3
appears in either its protonated (X = Cl, m/z 322-326 or X = Br,
m/z 366-370) or sodiated (X = Cl, m/z 344-348 or X = Br, m/z
3
88-392). A small amount of S
observed for bromoacetyl chloride in ACN but not in toluene. In
the case of the chloro starting material (2), the S 2 and acylation
N
2 product (m/z 322) was
Scheme 2. Mesylation of 4 followed by reaction with dimethylaminoethanol
(DMAE) to form diphenhydramine,5.
N
products have identical molecular formulae and are not
differentiable; however, the presence of only small amounts of
introduction of the second reagent, dimethylaminoethanol (20
equivalents) and repetition of the spray process (Figure 2). The
MS analysis of the two-step spray product indicated that ACN is
overall a better solvent for the synthesis of diphenhydramine, 5
N
the S 2 product using
bromo starting material, provides
evidence that the chloro reacts mainly to form the desired
acylation product.
(
m/z 256), than toluene. Unreacted DMAE (m/z 90), mesylated
DMAE (m/z 168), and a dimer of DMAE with methanesulfonic
acid (m/z 275) were observed with the charged microdroplets. A
similar trend was observed in flow (Figure S3, S4), where
optimized conditions gave diphenhydramine in 35% and <1%
yield in ACN and toluene, respectively, using one equivalent of
dimethylaminoethanol. Good agreement is observed between
the charged droplet reactor and flow in this synthesis.
Figure 1. Synthesis of 3 in ACN using a) chloroacetyl chloride and b)
bromoacetyl chloride in droplet reactor and microfluidics (µ-Fl)
Flow experiments were performed using the same
concentrations as in the droplet reactor, while screening the
effect of temperature for a fixed residence time of 30 seconds.
High conversion to 3 was observed with chloroacetyl chloride in
both solvents at 50°C. More interestingly, a major difference was
observed with bromoacetyl chloride in ACN, wherein a major
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601270
COMMUNICATION
Figure 2. Charged microdroplet (a,c) and microfluidic (b,d) reaction
The occurrence of accelerated organic reactions in
Leidenfrost droplets is at least in part a surface property
partial solvation of reagent molecules at the surface reduces
[5]
telescoping for diphenhydramine synthesis in two solvents (ACN, toluene)
(
The charged droplet and microfluidic synthesis of atropine
also was achieved by telescoping two reaction steps. The
intermediate ester 8 (Scheme 3) was prepared from the
commercially available starting material tropine
phenylacetyl chloride 7.
activation energies). Consistent with this, and the smaller
surface/volume ratios of Leidenfrost droplets, acceleration
factors are smaller than in electrosprayed microdroplets.
However, the larger droplets (0.5 mL volume) mean that
conditions in the Leidenfrost droplets are closer to those in
microfluidic solutions and in the bulk, so the predictive power of
Leidenfrost droplet reactions might be even greater than that of
electrospray generated droplets. The lack of a formal charge
also strengthens the expected analogy with scaled-up chemistry.
This expectation is met when one considers data for the first
step of the diazepam synthesis using bromoacetyl chloride and
chloroacetyl chloride. First, consider the mass spectra recorded
for charged droplets in ACN and in toluene versus those for
Leidenfrost droplets in the same two solvents (Figure 4, Figure
S6). The assignment of m/z is the same as figure 1 and scheme
6
and
Scheme 3. Esterification reaction of 6 with 7 to synthesize 8 followed by base
catalyzed aldol condensation with formaldehyde to synthesize atropine, 9.
1
(
. The conversion in the Leidenfrost experiment is not as great
more starting material seen) as in the charged droplet
reactions, but both methods give almost exclusively the desired
acylation intermediate as opposed to the S 2 product. There is
The intermediate 7 was used without further purification for the
aldol condensation reaction to produce the final product,
atropine. The first step was optimized for solvent and reactant
stoichiometry. The droplet reactor indicated dimethylacetamide
as the best solvent and this was confirmed in microfluidics (data
not shown). Using the unpurified intermediate ester 8 (m/z 260),
N
not a large difference in the results for ACN versus toluene as
solvent, except that the conversion is slightly higher in ACN. If
we now consider the difference between microfluidic flow and
Leidenfrost droplet data, we find remarkable similarities.
Microfluidic synthesis at 50°C results primarily in desired
acylation product as is the case in the Leidenfrost droplets. One
difference is in the formation of a minor species seen at m/z 260
in the Leidenfrost case.
a
base screen with the droplet reactor determined the
effectiveness of three bases in synthesizing atropine, 9 (m/z
90). Each base was successful in the droplet reactor,
producing significant amounts of atropine (and byproducts), with
,5-diazabicyclo[4.3.0]dec-5-ene being the most effective
Figure 3, Figure S5). In flow, each base produced atropine (and
2
1
(
byproducts) with 1,5-diazabicyclo[4.3.0]dec-5-ene again being
the most efficient. There is also some agreement between flow
and charged droplets on the type and extent of byproduct
formation. For example, using 1,5-diazabicyclo[4.3.0]dec-5-ene,
atropine and its dehydration product (m/z 272, 10) are observed.
However, with MeOK the same type of byproducts could be
observed, but their proportions were quite different. (Scheme 3).
The major byproduct in flow, characterized by m/z 304, 11, is
barely formed in charged droplets. One possible reason for this
difference is that formaldehyde with its high vapor pressure
escapes the droplets rapidly obviating formation of this
byproduct. Finally, good agreement between the two methods
was observed with NaOH, where the ,-unsaturated product
(
m/z 272, 10) is the major byproduct for both droplet reactor and
microfluidics. Thus, for the synthesis of atropine, the charged
droplet reactor was useful in guiding the choice of solvent and
base.
Figure 4. Synthesis of 3 in toluene using chloroacetyl chloride a) and bromo
acetylchloride b) in Leidenfrost, charged microdroplets and microfluidics
The ion m/z 260 is believed to be due to a ring closure product
resulting from acylation. The uncharged Leidenfrost droplets
closely mirror the chemistry in microfluidics, except when the
temperature in the microfluidics reaction is greatly elevated.
Under these circumstances different byproducts are generated
as S
The complete synthesis of diazepam in Leidenfrost droplets
was demonstrated by adding 7 or 70 equivalents of NH to
N
2 becomes more competitive.
3
intermediate 3 (not isolated) in a telescoped reaction. The
chloro intermediate was unable to produce diazepam in quantity.
However, the bromo intermediate produced diazepam
(
confirmed by MS/MS) in agreement with a reported flow
[7]
synthesis.
Conclusions
Figure 3. Charged microdroplet (a,c) and microfluidic (b,d) reaction
telescoping for atropine synthesis using potassium methoxide or 1,5-
diazabicyclo[4.3.0]dec-5-ene
This study provides evidence for the importance of MS, not
only in the traditional sense as an analytical method, but as a
fast, predictive means to perform small scale continuous
synthesis. The data encourage the use of spray ionization as a
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601270
COMMUNICATION
[
5]
method of screening for successful reaction pathways in flow
reactions. At a secondary level, we find some parallels in the
favored catalysts, solvents, mole ratios of reagents, and other
operating conditions. However, little is known of droplet reaction
mechanisms (an important topic in its own right), so
extrapolation from conditions that favor reactions in
nanodroplets to microfluidic chemistry may not be simple or
universal. Nevertheless, as we show in this study, a useful
guide to the global aspects of flow chemistry is obtained in these
cases.
Leidenfrost Droplet Experiments: Reactions in Leidenfrost droplets
differ in that these are i) larger ii) net neutral and iii) involve elevated
temperatures. Reaction mixtures were added in aliquots over a 2 min
period to maintain a constant droplet volume (Figure S7). The droplet (ca.
2 mm diameter) was levitated in a petri dish atop a heater with a surface
temperature of 400
temperatures close to, but below, the boiling point of the solvent.
– 500 C [CARE!]. Reactions occurred at
[14]
Mass spectrometry: Mass spectral analysis of reaction products was
performed using an LTQ ion trap mass spectrometer (Thermo Fisher
Scientific, San Jose, CA) with nanoESI ionization. All product samples
Limitations in further extending this approach to reaction
screening (and to on-line reaction monitoring) are to be found in
the size, cost and complexity of commercial mass
spectrometers, many of the features of which are unnecessary
for this type of study. What is needed for the purposes
(
spray, Leidenfrost, or flow reactions) were diluted 1:100 into acetonitrile
before analysis unless otherwise noted. The distance between the tip of
the spray emitter and ion transfer capillary to the MS was held constant
at ca. 1 mm. Experiments were performed using borosilicate glass pulled
to a ca. 1-3 um aperture. A spray voltage of either positive or negative
.0 kV was used for all analysis. Positive ion mode was used for all
chemical analysis unless otherwise noted. Product ion (MS/MS) spectra
described here is
a small, portable, unit resolution, low
mass/charge range (to m/z 1000) instrument which has ambient
ionization and tandem mass spectrometry capabilities. A recent
review of miniature MS instruments describes a few systems
of this type.
2
[
13]
were recorded using collision induced dissociation (CID) with
normalized collision energy of 25 (manufacturer’s unit).
a
Synthesis of Diazepam Precursor: Reactions were performed with 100
mM 5-Chloro-2-(methylamino)benzophenone and 100 mM 2-haloacetyl
chloride (halo = Cl, Br) dissolved in either toluene or acetonitrile.
Solutions were either mixed prior to use (droplet reactor, Leidenfrost) or
mixed online (microfluidics) to give a final reaction concentration of 50
mM. Other conditions were explored as indicated in the results section.
Experimental Section
Droplet Reactor Experiments: These experiments used electrospray
ionization (ESI) by spraying the reaction mixture either i) directly into the
MS or ii) onto a collection surface before taking up the residue in solvent
and performing ESI-MS product analysis. Figure 5 illustrates these two
options, which are distinguished by the fact that i) is virtually
instantaneous (10 – 15 s) while ii) can take a few minutes, but ii) is more
versatile in that the reaction and analysis occur in separate steps, more
product is formed, and the procedure allows temperature and other
conditions to be varied and optimized. In both cases, the primary
question being asked is whether the desired reaction occurs or not.
Products, byproducts and residual reagents were identified from mass
spectra recorded at unit resolution while tandem mass spectrometry
Diphenhydramine
Reactions were performed in
a two-step manner. First, 500 mM
benzhydrol was mixed with 500 mM mesyl chloride, then in the second
step 20 equivalents of dimethylaminoethanol was used in the droplet
reactor, while 1 equiv was used for microfluidics. Other conditions were
explored as indicated in the results section.
(MS/MS) was used to confirm identifications.
Atropine: The atropine intermediate was first synthesized by reacting 1
M phenylacetyl chloride and 1 M tropine dissolved in DMA. For the
second step, 7 equiv of base in DMA and 7 equiv of formaldehyde in H O
2
was used. Other conditions were explored as indicated in the Results
section.
Microfluidics: Microfludic reactions were performed using a Labtrix S1
system from Chemtrix, Ltd. The Chemtrix system is comprised of syringe
pumps to deliver regents, microfluidic chips, a Peltier element, back
pressure regulator, and a collection carousel. For the synthesis of
diphenhydramine a homebuilt Peltier controlled system coupled with the
Chemtrix microfluidic platform was used to allow for multi-step reactions
to be performed with control over the temperature of each step.
Chemicals and reagents: All chemicals were purchased from Sigma
Aldrich and used without further purification.
Figure 5. Methods used to perform microdroplet reactions, based on ESI
either a) with on-line product analysis by MS or b) with sprayed droplet
deposition and subsequent off-line MS product analysis
Acknowledgements
Offline Droplet Reactor: Offline droplet experiments were performed
using a homebuilt electrospray ionization source. In the cases of atropine
and diazepam, the reagents were premixed and subjected to offline
Support is acknowledged from the Department of Defense:
Defense Advanced Research Projects Agency, Award no:
W911NF-16-2-0020.
2
electrospray at 10 µl/min with +5kV voltage and 100 PSI N . For
diphenhydramine, reagents were mixed inline using a mixing tee and
offline electrospray was performed under the same conditions. After the
electrospray deposition was complete, reaction product was rinsed from
the collection surface and then analyzed by nanoESI. Samples were
diluted at least 100-fold before analysis in order to quench the reaction
and ensure no further reaction could occur during the analysis step. For
two- step reactions, the washed material was drawn back into a syringe
and mixed with the second step reagent and then electrosprayed,
collected, and washed as before.
Keywords: reaction kinetics • mass spectrometry • reaction
acceleration • flow chemistry • Leidenfrost effect
[1]
aX. Yan, R. M. Bain, R. G. Cooks, Angewandte Chemie International
Edition 2016, n/a-n/a; bR. M. Bain, C. J. Pulliam, R. G. Cooks,
Chemical Science 2015, 6, 397-401; cR. M. Bain, C. J. Pulliam, S. T.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601270
COMMUNICATION
Ayrton, K. Bain, R. G. Cooks, Rapid Communications in Mass
Spectrometry 2016, 30, 1875-1878.
[
2]
aJ. K. Lee, S. Kim, H. G. Nam, R. N. Zare, Proceedings of the National
Academy of Sciences 2015, 112, 3898-3903; bR. M. Bain, C. J. Pulliam,
S. A. Raab, R. G. Cooks, Journal of Chemical Education 2016, 93, 340-
344; cY. Li, X. Yan, R. G. Cooks, Angewandte Chemie International
Edition 2016, 55, 3433-3437.
[
[
[
[
3]
4]
5]
6]
S. Banerjee, R. N. Zare, Angewandte Chemie International Edition
2
015, 54, 14795-14799.
C. Josserand, S. T. Thoroddsen, Annual Review of Fluid Mechanics
016, 48, 365-391.
R. M. Bain, C. J. Pulliam, F. Thery, R. G. Cooks, Angewandte Chemie
016, n/a-n/a.
2
2
S. Mascia, P. L. Heider, H. Zhang, R. Lakerveld, B. Benyahia, P. I.
Barton, R. D. Braatz, C. L. Cooney, J. M. B. Evans, T. F. Jamison, K. F.
Jensen, A. S. Myerson, B. L. Trout, Angewandte Chemie International
Edition 2013, 52, 12359-12363.
[
[
7]
8]
9]
A. Adamo, R. L. Beingessner, M. Behnam, J. Chen, T. F. Jamison, K. F.
Jensen, J.-C. M. Monbaliu, A. S. Myerson, E. M. Revalor, D. R. Snead,
T. Stelzer, N. Weeranoppanant, S. Y. Wong, P. Zhang, Science 2016,
352, 61-67.
A. Buitrago Santanilla, E. L. Regalado, T. Pereira, M. Shevlin, K.
Bateman, L.-C. Campeau, J. Schneeweis, S. Berritt, Z.-C. Shi, P.
Nantermet, Y. Liu, R. Helmy, C. J. Welch, P. Vachal, I. W. Davies, T.
Cernak, S. D. Dreher, Science 2015, 347, 49-53.
[
[
A. G. Godfrey, T. Masquelin, H. Hemmerle, Drug Discovery Today
2013, 18, 795-802.
10] aH. Song, D. L. Chen, R. F. Ismagilov, Angewandte Chemie
International Edition 2006, 45, 7336-7356; bK. Jähnisch, V. Hessel, H.
Löwe, M. Baerns, Angewandte Chemie International Edition 2004, 43,
4
06-446.
11] S. C. Gibson, C. S. Feigerle, K. D. Cook, Analytical Chemistry 2014, 86,
64-472.
[
4
[
[
12] D. R. Snead, T. F. Jamison, Chemical Science 2013, 4, 2822-2827.
13] D. T. Snyder, C. J. Pulliam, Z. Ouyang, R. G. Cooks, Analytical
Chemistry 2016, 88, 2-29.
[
14] R. Abdelaziz, D. Disci-Zayed, M. K. Hedayati, J.-H. Pöhls, A. U. Zillohu,
B. Erkartal, V. S. K. Chakravadhanula, V. Duppel, L. Kienle, M. Elbahri,
Nat Commun 2013, 4.
This article is protected by copyright. All rights reserved

European Journal of Organic Chemistry
10.1002/ejoc.201601270
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
COMMUNICATION
This paper asks whether the course of
reaction in charged microdroplets and
uncharged Leidenfrost droplets can be
used as a guide to reactions in
microfluidic flow systems. Synthesis of
three important pharmaceutical drugs
is explored with the droplet reactor
providing information on solvent,
catalysts, and byproducts in flow.
Michael Wleklinski, Caitlin E. Falcone,
Bradley P. Loren, Zinia Jaman, Kiran
Iyer, H. Samuel Ewan, Seok-Hee Hyun,
David H. Thompson*, and R. Graham
Cooks*
Page No. – Page No.
Can Accelerated Reactions in
Droplets Guide Chemistry at Scale?
This article is protected by copyright. All rights reserved