Thieme Chemistry Journal Awardees - Where are they now? Catalytic transport with an amine carrier in a fluorous triphasic reaction

Article abstract of DOI:10.1055/s-0028-1087916

Several aromatic aldehydes are transported by a fluorous amine from one organic phase through a fluorous phase to another organic phase. The derived imines react with phenylhydrazine to immobilize the transported product as a hydrazone and release the ami

Full text of DOI:10.1055/s-0028-1087916

554
LETTER
Thieme Chemistry Journal Awardees – Where are They Now?
Catalytic Transport with an Amine Carrier in a Fluorous Triphasic Reaction
Catalytic
F
luorous
i
Transp
t
ort torio Montanari,a Marvin S. Yu,*a Dennis P. Curran*b,1
a
Fluorous Technologies, Inc., UPARC, 970 William Pitt Way, Pittsburgh, PA 15238, USA
Fax +1(412)9656448; E-mail: myu@fluorous.com
b
Department of Chemistry, University of Pittsburgh, Pittsburgh, PA 15260, USA
E-mail: curran@pitt.edu
Received 1 August 2008
Abstract: Several aromatic aldehydes are transported by a fluorous
amine from one organic phase through a fluorous phase to another
organic phase. The derived imines react with phenylhydrazine to
immobilize the transported product as a hydrazone and release the
amine for reuse. In this way, catalytic transport is accomplished for
the first time.
Key words: triphasic reaction, transport, fluorous chemistry, U-
tube, fluorous amine
The production of new organic molecules is typically
comprised of a reaction and a separation, and these two
stages are almost always conducted separately in space
and time. Conducting a reaction and a separation coinci-
dently has the potential to result in a synthesis device (or
machine) where starting materials are fed in and products
are harvested out continuously. Such devices could poten-
tially be piggybacked to provide for continuous, realtime,
multistep reaction and separation.
Dennis P. Curran received his B.Sc. in 1975 from Boston College.
His Ph.D. was granted from the University of Rochester in 1979 where
he worked under Professor Andrew S. Kende. After a two year post-
doctoral stay with Barry M. Trost at the University of Wisconsin, Dr.
Curran joined the faculty of the Chemistry Department at the Univer-
sity of Pittsburgh in 1981. He now holds the ranks of Distinguished
Service Professor and Bayer Professor of Chemistry, and is the foun-
der of Fluorous Technologies, Inc. Among other awards, Dr. Curran
has received the American Chemical Society Award for Creative
Work in Fluorine Chemistry (2008), the University of Pittsburgh In-
novator Award (2007), the American Chemical Society Award for
Creativity in Organic Synthesis (2000) and the Cope Scholar Award
(1988). He is currently an ISI "Highly Cited Researcher". Dr. Curran
is well known for his work at the interface of radical chemistry and
organic synthesis. More recently, he has made significant contribu-
tions to the emerging discipline of fluorous chemistry.
The separation component of any synthesis device re-
quires some kind of partitioning, and the immiscibility of
fluorous and organic liquids offers attractive opportuni-
ties.2 In 2001, we introduced fluorous triphasic devices for
transportative deprotection of fluorous-tagged materials.3
In the example shown in Figure 2, a mixture resulting
from enzymatic enantioselective hydrolysis of a racemic
fluorous-tagged alcohol was simultaneously separated
and detagged.4 The mixture was comprised of roughly
equal amounts of the free alcohol (R)-1 and its enantio-
meric counterpart (S)-2 still bearing the fluorous tag.
Three steps are needed to complete the resolution: 1) the
alcohol (R)-1 and the ester (S)-2 must be separated; 2) the
retrieved ester (S)-2 must be hydrolyzed, and 3) the resid-
ual tag must be separated from the resulting alcohol (S)-1.
to the receiving phase, where it is detagged by the base.
This strands the resulting alcohol (S)-1 in the receiving
phase, while the residual tag (which is now highly fluo-
rous) partitions back to the fluorous phase. The starting al-
cohol (R)-1 stays put in the source phase. Thus, in an ideal
process, there is one product in each of the three phases.
The triphasic device in Figure 1, a simple U-tube, accom-
plishes all three tasks simultaneously. The U-tube con-
tains two organic phases (source and receiving) separated
by a fluorous phase that prevents direct contact between
these phases and thereby regulates exchange. A mixture
of (R)-1 and (S)-2 is added to the source side of the device
and sodium methoxide is to the receiving side. Over time,
the ester (S)-2 partitions through the fluorous phase over
This kind of ‘transportative detagging’ of fluorous-tagged
substrates is a potentially general way to remove non-
tagged contaminants along with removal of the fluorous
tag. But such processes are by definition stoichiometric in
the fluorous component (the tag). Herein, we report the
first catalytic fluorous triphasic reaction and separation
process.
SYNLETT 2009, No. 4, pp 0554–0557
x
x.
x
x.
2
0
0
9
Advanced online publication: 16.02.2009
DOI: 10.1055/s-0028-1087916; Art ID: S06808ST
© Georg Thieme Verlag Stuttgart · New York
In principle, it should be possible in a fluorous–organic
triphasic device to convert an organic substrate to an or-
ganic product with the aid of a substoichiometric amount

LETTER
Catalytic Fluorous Transport
555
O
start
+
finish
H
H
OH
C8F17
O
O
O
(R)-1
(S)-2
(R)-1
from source phase
(S)-2
from receiving phase
source
phase
receiving
phase
MeOH
MeO–
MeOH
(R)
(S)
RfCH2CH2CO2Me
FC-72
Figure 1 Transportative detagging of a ester–alcohol mixture generated by enzymatic kinetic resolution
of a fluorous carrier. We set out to prove this principle by amine 3 were vortexed together in a vial to allow reaction
transporting an aldehyde from the source phase though the and subsequent partitioning of the resulting imine 5. The
fluorous phase with a fluorous amine carrier via the inter- phases were then separated and the forward fluorous-
mediacy of a fluorous imine. Aldehydes are important raw phase-containing part of this imine 5 was vortexed with
materials, and their feedstocks can be contaminated with the receiving phase to allow reaction to form the hydra-
alkenes, alcohols, and other impurities. Thus, their sepa- zone 7 with release of the imine 5. Then the back fluorous
ration with a simultaneous onward reaction has potential phase containing the amine 3 was returned to the source
value.
phase and the whole process was repeated in a series of
identical cycles.
The reactions involved in the process are summarized
with 4-bromobenzaldehyde (4) in Scheme 1. In the source
organic phase, 4 reacts with fluorous amine carrier 3 to
give fluorous imine 5. In turn, the imine 5 is transported
through the fluorous phase to the receiving organic phase,
where it reacts with added phenyl hydrazine 6 to provide
hydrazone 7. The energy released in this reaction drives
the separation, and the amine carrier is also released for
return through the fluorous phase to the source organic
phase.
source phase-reaction
C8F17
NH2
+
Br
CHO
3
4
C8F17
N C C6H4Br
H
5, transported forward
receiving-phase reaction
H
C8F17
N
C
H
C6H4Br
Ph
N
NH2
+
+
We expected the rates of both reaction steps to be relative-
ly fast, and thus the rate of the overall reaction and sepa-
ration process will be limited by the transport rate. Our
previous stoichiometric triphasic reactions were also
transport-limited, and often took several days to go to
completion in standard U-tubes. Reducing the fluorous
component to a substoichiometric amount will further
slow the process, and accordingly we deemed the U-tube
device design unsuitable.
5
6
C8F17
NH2
Br
C NNHPh
H
3, tranported back
7
Scheme 1 Reactions involved in the catalytic transport with fluo-
rous amine carrier 3
In principle, the transport rate problem can be solved by
engineering solutions in which the fluorous phase is circu-
lated between the source and receiving phases. We decid-
ed to mimic this engineering solution with serial
extractions, as summarized in Figure 2. Thus, the source
phase with the aldehyde 5 and the fluorous phase with the
If the reactions are rapid and quantitative, and if the parti-
tioning of the fluorous amine/imine into the fluorous
phase is also quantitative, then it is possible to transport in
each cycle an amount of aldehyde that is equal to the
amount of amine added. Control experiments suggested
Synlett 2009, No. 4, 554–557 © Thieme Stuttgart · New York

556
V. Montanari et al.
LETTER
source phase, MeCN
Br
CHO
4
fluorous phase
forward, FC-72
fluorous phase
back, FC-72
C8F17
N
C
H
C6H4Br
C8F17
NH2
5
3
Figure 3 Plot of transport of 4 to form 7 as a function of cycle
there was no detectable back transport of either the phe-
nylhydrazine (6) or the product hydrazone 7 to the source
phase.
receiving phase, MeCN
H
Br
C
H
NNHPh
Ph
N
NH2
Figure 3 plots the accumulation of hydrazone 7 in the re-
ceiving phase as a function of cycle. As expected, the plot
is linear, and about 0.2 mmol of aldehyde 4 is transported
from the source phase to the receiving phase in each cycle.
This means that about 35% of amine 3 that was added was
productive in each cycle.
6
7
Figure 2 Simplified illustration of the iterative reaction and extrac-
tion cycle with infinite partition coefficients. In reality, the organic
components have high partition coefficients, but the fluorous com-
ponents 3 and 5 distribute between both phases
After the 10th cycle, the yield of phenylhydrazone based
on aldehyde was 89%, and the yield based on amine was
350%. This shows the catalytic nature of the transport.
The GC analysis showed that the amine 3 was largely in-
tact after the 10th cycle, distributed over the three phases.
We conducted a similar experiment with about 10% of the
amine catalyst, and about 40% of the phenylhydrazone 7
was produced in the receiving vial after ten cycles (about
4% transport per cycle). This is consistent with the expec-
tations based on the amount of productive amine from the
first experiment.
that the two reactions were rapid and quantitative, as ex-
pected.5
But the partitioning of the amine 3 and imine 5 into the
fluorous phase will surely not be quantitative since the
fluorine content of these compounds is not nearly high
enough.6,7 In that case, after a few cycles, a kind of quasi-
equilibrium will be reached in which there is unproductive
amine 3 in both the source (as imine) and receiving phases
and productive amine 3 in the fluorous phase (the imine is
present in the forward direction and the free amine in the
reverse direction). In this scenario, the amount of alde-
hyde transported in each cycle will equal the amount of
productive amine. Provided that the aldehyde is present in
excess, a plot of transport amount versus time should be
linear and should not depend on the aldehyde concentra-
tion.
These experiments need to be controlled for direct (pas-
sive) transport of the aldehyde through the fluorous phase
to the receiving phase. In these first experiments, we did
this indirectly by measuring the partition coefficient of the
aldehyde between FC-72 and acetonitrile. This was <0.01,
so significant passive transport seems unlikely.
We then conducted a second series of cycles with the
same procedure mentioned above with 4-fluorobenzalde-
hyde and 10% amine 3. This time, we did a parallel series
of control cycles with all the same components, but omit-
ting the fluorous amine. The plots of hydrazone yield ver-
sus cycle for both experiments are shown in Figure 4. The
control experiment, which was stopped after the 4th cycle,
shows that the background (passive) transport rate is not
zero, but it is rather small (and 1% per cycle). This is pre-
sumably due to direct partitioning of the aldehyde into the
FC-72.9
In a typical experiment, 4-bromobenzaldehyde (4, 2.2
mmol) and (perfluorooctyl)propylamine (3, 0.56 mmol,
25 mol%) in acetonitrile (7 mL) and FC 72 (7 mL) were
vortexed in the source vial for one hour. The phases were
allowed to separate, and the FC-72 phase was transferred
to the receiving vial containing phenylhydrazine (6, 2.3
mmol) in acetonitrile (7 mL). This was similarly vortexed
for one hour. The FC phase was then returned to source
vial to start the second cycle. Cycles were repeated until
phenylhydrazine was largely consumed.8 The progress of
the reaction was followed by GC. We could not detect
either imine 5 or aldehyde 4 in the receiving phase. Also,
Synlett 2009, No. 4, 554–557 © Thieme Stuttgart · New York

LETTER
Catalytic Fluorous Transport
557
ing solvents, catalysts, etc. to alter partition coefficients)
and through engineering (for example, by using continu-
ous extractors).
Acknowledgment
FTI thanks the National Science Foundation (SBIR) and DPC
thanks the National Institutes of Health for funding of this work.
References and Notes
(1) DPC holds an equity interest in Fluorous Technologies, Inc.
(2) (a) Gladysz, J. A.; Emnet, C. In The Handbook of Fluorous
Chemistry; Gladysz, J. A.; Curran, D. P.; Horvath, I. T.,
Eds.; Wiley-VCH: Weinheim, 2004, 11–23. (b) Ryu, I.;
Matsubara, H.; Emnet, C.; Gladysz, J. A.; Takeuchi, S.;
Nakamura, Y.; Curran, D. P. In Green Reaction Media in
Organic Synthesis; Blackwell: Ames IO, 2005, 59–124.
(c) Iskra, J. Lett. Org. Chem. 2006, 3, 170–175. (d) Ryu, I.;
Matsubara, H.; Nakamura, H.; Curran, D. P. Chem. Rec.
2008, 351.
Figure 4 Active transport with amine 3 (10%) and 4-fluorobenzal-
dehyde (top line) plotted with control for passive transport (bottom
line)
The results for the transport experiment with 4-fluorobenz-
aldeyde were similar to those with 4-bromobenzaldehyde.
With 10% of the amine 3, about 40% of the aldehyde was
transported to form the hydrazide after the 10th cycle. The
initial rate of transport in the experiment with the amine
exceeds that rate of the background transport by a factor
of about 5. This again demonstrates the effect of the
amine/imine on the transport. We did similar experiments
with benzaldehyde and 4-trifluoromethylbenzaldeyde
with generally similar results (not shown). In each of these
case the initial rate of transport in the experiment with the
amine exceeded that of the control by a factor of 3.
(3) Nakamura, H.; Linclau, B.; Curran, D. P. J. Am. Chem. Soc.
2001, 123, 10119.
(4) Luo, Z. Y.; Swaleh, S. M.; Theil, F.; Curran, D. P. Org. Lett.
2002, 4, 2585.
(5) The controls were conducted similarly to the transport
experiments. Aldehydes were vortexed with 1 equiv of
amine or hydrazine for 1 h in MeCN. Formation of the imine
or hydrazone was quantitative as assayed by GC. These
samples then served as standards for the transport
experiments.
(6) Gladysz, J. A.; Emnet, C.; Rábai, J. In The Handbook of
Fluorous Chemistry; Gladysz, J. A.; Curran, D. P.; Horvath,
I. T., Eds.; Wiley-VCH: Weinheim, 2004, 56–100.
(7) The Kp ([FC-72]/[CH3CN]) of 3 and 5 were measured by
direct partitioning as 2.3 and 0.8, respectively.
(8) About 5–10% of the volume of the FC-72 phase was lost in
each cycle, presumably due to evaporation. This was made
up by adding additional FC-72 at the beginning of each new
cycle to return the volume to 7 mL.
(9) We also considered that transport of the aldehyde might
occur by accidental transfer of some of the organic source
phase to the receiving phase during the cycles. But we did
not observe back transfer of the hydrazine or hydrazone, so
These experiments demonstrate for the first time that an
organic reactant can be transported from one organic
phase through a fluorous phase to another organic phase
by a substoichiometric amount of a fluorous transport
agent. The experiments are prototypes for synthesis devic-
es that accomplish a simultaneous reaction and separation
and that are catalytic in the fluorous component. Assum-
ing that the reaction rates are fast, the efficiency of such
devices will depend on the efficiency of the transport.
This is subject to optimization through chemistry (chang-
Synlett 2009, No. 4, 554–557 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.