Flow chemistry and polymer-supported pseudoenantiomeric acylating agents enable parallel kinetic resolution of chiral saturated N-heterocycles

Article abstract of DOI:10.1038/nchem.2681

Kinetic resolution is a common method to obtain enantioenriched material from a racemic mixture. This process will deliver enantiopure unreacted material when the selectivity factor of the process, s, is greater than 1; however, the scalemic reaction product is often discarded. Parallel kinetic resolution, on the other hand, provides access to two enantioenriched products from a single racemic starting material, but suffers from a variety of practical challenges regarding experimental design that limit its applications. Here, we describe the development of a flow-based system that enables practical parallel kinetic resolution of saturated N-heterocycles. This process provides access to both enantiomers of the starting material in good yield and high enantiopurity; similar results with classical kinetic resolution would require selectivity factors in the range of s = 100. To achieve this, two immobilized quasienantiomeric acylating agents were designed for the asymmetric acylation of racemic saturated N-heterocycles. Using the flow-based system we could efficiently separate, recover and reuse the polymer-supported reagents. The amide products could be readily separated and hydrolysed to the corresponding amines without detectable epimerization.

Full text of DOI:10.1038/nchem.2681

ARTICLES
PUBLISHED ONLINE: 12 DECEMBER 2016 | DOI: 10.1038/NCHEM.2681
Flow chemistry and polymer-supported
pseudoenantiomeric acylating agents enable
parallel kinetic resolution of chiral saturated
N-heterocycles
Imants Kreituss and Jeffrey W. Bode
*
Kinetic resolution is a common method to obtain enantioenriched material from a racemic mixture. This process will
deliver enantiopure unreacted material when the selectivity factor of the process, s, is greater than 1; however, the
scalemic reaction product is often discarded. Parallel kinetic resolution, on the other hand, provides access to two
enantioenriched products from a single racemic starting material, but suffers from a variety of practical challenges
regarding experimental design that limit its applications. Here, we describe the development of a flow-based system that
enables practical parallel kinetic resolution of saturated N-heterocycles. This process provides access to both enantiomers
of the starting material in good yield and high enantiopurity; similar results with classical kinetic resolution would require
selectivity factors in the range of s = 100. To achieve this, two immobilized quasienantiomeric acylating agents were
designed for the asymmetric acylation of racemic saturated N-heterocycles. Using the flow-based system we could
efficiently separate, recover and reuse the polymer-supported reagents. The amide products could be readily separated
and hydrolysed to the corresponding amines without detectable epimerization.
inetic resolution (KR) is a widely used method to separate one of the product and gain access to both enantiomers of the racemate,
enantiomer of a chiral compound from a racemic mixture1,2. Vedejs introduced the concept of PKR5. In PKR, two KR reactions are
K
This occurs by selective chemical transformation of one conducted simultaneously to yield two distinct non-enantiomeric
enantiomer to a new product, which is typically formed with only products (Fig. 1a). If both enantiomers of the racemate react with
modest enantiomeric excess. Unless reagents, such as enzymes, similar rates, the optimal 1:1 enantiomer concentration is maintained
with selectivity factors (s) > 500 are used, KR is useful for producing throughout the course of the resolution and both products are formed
the enantioenriched starting material with excellent enantiopurity in significantly improved enantiopurity as compared with a classical
and modest yields, but is typically not appropriate if both enantio- KR (Fig. 1b). For a successful PKR, the implemented resolution reac-
mers are needed in high enantiopurity as the product enantiopurity tions should: (a) have similar (preferably identical) rates kR ≈ kS; (b)
is modest3. Parallel kinetic resolution (PKR)—in which both enan- occur without mutual interference; (c) have opposite enantioselectiv-
tiomers of the starting material undergo reactions with pseudo- ity with respect to the substrate; and (d) yield separable reaction pro-
enantomeric reagents—can lead to distinct products, each with ducts6. Due to these requirements the reaction design of PKR is
high enantiomeric excess. So far this strategy has been difficult to difficult; while several conceptual examples of PKR have been
implement in a practical fashion due to the challenge of physically reported, very few practical implementations have emerged7–9
.
separating the reagents and readily isolating the pseudoenantiomeric Enantiodivergence can also be achieved using a single chiral reagent
products4. In this report, we describe a combination of flow chem- or catalyst, most often leading to enantioenriched regioisomeric10–15
istry, solid supported resolving agents, and readily cleavable tertiary or diastereomeric16–18 products. Such processes are referred to as regio-
amides for the successful PKR of chiral, saturated N-heterocycles divergent or stereodivergent reactions of racemic mixtures and they are
including piperazines, morpholines, piperdines, tetrahydroisoquino- more common in organic synthesis4. These are attractive processes
lines, and others. Despite an intrinsic selectivity of only s = 10–20 when they can be implemented, however they do not fully benefit
in most cases, flow-based PKR allows the two enantiomers to be from the inherent advantages of a true PKR19.
isolated in good yield and outstanding enantiomeric excesses, an
outcome that would normally require reagents with selectivities in acid (PEARL, polymeric reagents for KR ) for the KR of saturated
the range of s = 50–100. N-heterocycles on both small (100 mg) and preparative (>20 g)
We have recently reported immobilized chiral acyl hydroxamic
In a typical KR the enantioenriched starting material is recovered scales (Fig. 1c)20. Using the PEARL reagents, we obtained enantio-
at the expense of enantiopurity of the product by running the reaction pure amines from their racemates, albeit in diminished yields at
to higher conversion; the scalemic product is often discarded. the expense of the scalemic product. As both enantiomers of the
Although KR can in principle provide the recovered starting PEARL reagents are readily available on scale, we sought to tackle
material in high enantiopurity, in practice it is often difficult and the resolution of saturated N-heterocycles via PKR. Herein, we
time consuming to run the reactions to sufficient conversion. To report successful application of polymer bound quasienantiomeric
increase the efficiency of the resolution, maximize the enantiopurity reagents for PKR of a broad range of chiral N-heterocycles. To
Laboratorium für Organische Chemie, Department of Chemistry and Applied Biosciences, ETH-Zürich, 8093 Zürich, Switzerland.
e-mail: bode@org.chem.ethz.ch
*
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
1

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2681
ARTICLES
a
Kinetic resolution
Parallel kinetic resolution
B*
kS
B*
kS
A(S)
A(S)
D
F
C
kS ≈ kR
s = kS/kR
B* and E*: quasienantiomeric
reagents
A(S) : A(R) = 1:1
A(S) : A(R) = 1:1
B*: chiral reagent
B*
kR
E*
kR
A(R)
A(R)
ent-C
At t0: [A(S)] = [A(R)] & kS >> kR (Initial relative rates)
As the reaction proceeds...
At t0: [A(S)] = [A(R)] & kS ≈ kR (Initial relative rates)
As the reaction proceeds...
At tx: [A(S)] << [A(R)] & kS ≈ kR
At tx: [A(S)] = [A(R)] & kS ≈ kR
Increase of the relative rate of the less reactive enantiomer
No change in relative concentrations or rates
b
Recovered amine (KR)
100
Amide products (PKR)
Amide product (KR)
95% conversion
(Parallel kinetic resolution)
80
50% conversion
(Kinetic resolution)
60
40
20
0
0
5
10
15
Time (h)
20
25
30
R1
O
2–3 ml min–1
c
d
O
O
O
X
N
N
O
N
H
R
45 °C, 24 h
R2
O
X
O
+
N
H
R
X
N
O
R
X
O
X
Ph
O
X
O
Ph
N
O
N
R
N
H
R
R
N
H
R
R1
O
O
O
X
N
X
R
N
H
R2
Figure 1 | Principles of kinetic and parallel kinetic resolution. a, Theoretical background of KR and PKR. b, Simulation of a KR and PKR reaction system (with
s = 20) showing the % enantiomeric excess of the products or starting materials during the course of the reaction (for more detailed discussion regarding
reaction progress simulations see Supplementary page 5–9). In PKR, the enantiopurity of the two pseudoenantiomeric products remains constant. c, KR with
PEARL (polymeric reagents for amine resolution) reagents. Enantiopure amines can be recovered in <40% yield, the scalemic amide product can be
discarded, selectivity is in the range s = 10–20, and non-cleavable acyl groups are delivered. d, Flow enabled PKR. A simple reaction setup with cleavable
acyl groups, s of up to 100 and both amines recovered in higher yields.
achieve the essential physical separation of PEARL reagents, while Results
still allowing for near simultaneous reaction times and facile Selection of orthogonal, cleavable acyl groups. To prepare the
regeneration, we developed a flow-based implementation of quasienantiomeric resolving agents necessary for PKR, we first set out
this chemistry. We further developed acyl groups that allow for to identify two orthogonal acyl groups that would meet strict
facile separation of the enantioenriched amide products and requirements. They should be inexpensive, easy to prepare, exhibit
subsequent deprotection in good yields without epimerization, good selectivity, react with similar rates (that is, provide similar
resulting in an overall KR of the enantiomeric N-heterocycles selectivity to classical KR), and afford separable products. Carbonate
with selectivities and yields far exceeding those possible with and carbamate derivatives 1a and 1b—as well as acetate 1c—did not
traditional approaches (Fig. 1d).
afford any selectivity, and hydrocinnamoyl group 1d proved very
2
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2681
ARTICLES
O
R
available and inexpensive; 3-(2-nitrophenyl)propanoic acid was
prepared in two steps on a decagram scale (Supplementary page 10).
O
O
MeO
N
Parallel kinetic resolution with mixed beads. To test if the
polymeric reagents could be applied for PKR, we combined the
quasienantiomeric acylating agents in an equimolar ratio in THF
and treated the mixture with racemic amines. After 24 h at 45 °C
the resulting amide products were isolated in good yields and
high enantiopurity. The enantiomeric ratio of the amides was
independent of the amount of the polymer used in the process,
indicating that the resolution proceeds without interactions
between the acylating reagents. A broad range of cyclic secondary
amines were evaluated and all products were obtained with useful
enantiomeric ratios. To obtain similar results in classical KR of
cyclic secondary amines very high selectivities would be required
(s up to 90), so far unprecedented in amine resolution with small
molecule resolving agents (Table 1).
NH
MeO
O
MeO
MeO
Ph
1a–f
NH
+
MeO
MeO
Ph
N
R
O
Ph
Me
R =
R =
O
1a
s = 1
1c
s = 2
1e
s = 20
O2N
H
N
Batch reactors for recovery of PEARL reagents. Previously, we
demonstrated that PEARL reagents could be readily recovered,
recycled by treatment with the corresponding acid anhydride
and reused in KR without erosion in reactivity or selectivity.
We have established that a single batch can be used and
regenerated dozens of time for different substrates without loss
of activity. However, for PKR experiments the two resins were
mixed together and were not recoverable after the reaction. In
order to recycle the acylating agents after the reaction and
render the process practical, we designed an H-tube reactor
with two compartments. The polymeric reagents were separated
with a Teflon membrane, which was permeable to the solution
1b
s = 1
1d
s = 23
1f
s = 22
Figure 2 | Acylating agents and s factors afforded in kinetic resolution.
Enantiomeric ratios of the unreacted starting material and product were
determined by supercritical fluid chromatography (SFC) or HPLC on chiral
stationary phases. s was calculated according to ref. 2.
difficult to hydrolyse. After screening multiple variants, we identified the
pent-4-enoyl 1e and 3-(2-nitrophenyl)propanoyl 1f groups, which
afforded reasonable selectivities in KR and could be cleaved from the
amide product (Fig. 2). The requisite pent-4-enoic acid is commercially
Table 1 | Parallel kinetic resolution of saturated N-heterocycles.
NO2
X
X
X
N
R
O
O
+
N
R
O
O
O
O
N
R
N
O
N
H
O
O2N
O
O
THF
45 °C
2–10a
2–10b
18–32 h
O
O
Ph
NH
OEt
OEt
OEt
N
H
N
N
N
H
H
N
O
H
2a
3a
4a
5a
6a
e.r. = 92:8 (42% yield)
e.r. = 93:7 (35% yield)
e.r. = 94:6 (31% yield)
e.r. = 91:9 (40% yield)
e.r. = 91:9 (30% yield)
2b
e.r. = 94:6 (26% yield)
s = 42
3b
e.r. = 96:4 (26% yield)
s = 67
4b
e.r. = 96:4 (20% yield)
s = 70
5b
6b
e.r. = 92:8 (21% yield)
s = 30
e.r. = 93:7 (32% yield)
s = 34
H
OH
MeO
MeO
O
O
O
F
N
H
NH
N
H
N
Ph
H
F3C
N
F
CF3
7a
8a
9a
10a
e.r. = 95:5 (21% yield)
e.r. = 92:8 (26% yield)
e.r. = 93:7 (46% yield)
e.r. = 93:7 (25% yield)
7b
e.r. = 95:5 (27% yield)
s = 58
8b
9b
e.r. = 97:3 (50% yield)
s = 90
10b
e.r. = 95:5 (21% yield)
s = 53
e.r. = 92:8 (35% yield)
s = 30
Reactions were carried out on a 0.4 mmol scale in THF at 45 °C for 18–32 h using equimolar ratios of the immobilized acylating agents and racemic amine. Yields correspond to isolated products purified by
column chromatography. Enantiomeric ratios were determined by SFC or HPLC on chiral stationary phases. The s required to obtain given enantiopurities under classical KR conditions are shown.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
3
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2681
ARTICLES
Table 2 | H-tube reactor for parallel kinetic resolution.
NO2
X
N
X
O
N
X
O
O
R
O
N
O
O
N
R
N
R
O
H
O
O
O2N
THF, 45 °C
O
shaker in an oven
2a, 6a, 7a, 11a, 12a 2b, 6b, 7b, 11b, 12b
F
Me
N
O
F
OEt
NH
N
N
N
H
H
N
H
Ph
H
O
S
2a
6a
7a
11a
12a
e.r. = 89:11 (46% yield)
e.r. = 87:13 (27% yield)
e.r. = 93:7 (28% yield)
e.r. = 84:16
e.r. = 86:14 (37% yield)
2b
e.r. = 89:11 (48% yield)
s = 19
6b
e.r. = 91:9 (38% yield)
s = 22
7b
e.r. = 92:8 (35% yield)
s = 35
11b
e.r. = 91:9
s = 20
12b
e.r. = 90:10 (50% yield)
s = 19
Reactions were carried out on a 1.0–2.0 mmol scale in THF at 45 °C for 24 h using equimolar ratios of the immobilized acylating agents and the racemic amine. The acylating agents were separated by a Teflon
membrane mesh and the H-tube reactor was placed on a shaker in an oven to maintain reaction temperature and mixing. Yields correspond to products purified by column chromatography. Enantiomeric ratios
were determined by SFC or HPLC on chiral stationary phases. The s required to obtain given enantiopurities under classical KR conditions are shown.
Table 3 | Parallel kinetic resolution in flow.
X
2–3 ml min–1
N
R
X
O
45 ºC, 24 h
NO2
O
2a, 5a, 7a, 9a, 10a
N
H
R
+
X
O
N
O
O
O
O
N
N
R
O
O
O
O2N
2b, 5b, 7b, 9b, 10b
H
OH
N
H
MeO
MeO
O
F
NH
Ph
F3C
N
N
H
CO2Et
N
N
H
CF3
H
2a
5a
7a
9a
10a
e.r. = 95:5 (50% yield)
e.r. = 89:11 (45% yield)
e.r. = 96:4 (28% yield)
e.r. = 97:3 (40% yield)
e.r. = 94:6 (18% yield)
2b
e.r. = 93:7 (48% yield)
s = 53
5b
e.r. = 95:5 (35% yield)
s = 45
7b
e.r. = 94:6 (35% yield)
s = 70
9b
10b
e.r. = 96:4 (17% yield)
s = 70
e.r. = 96:4 (48% yield)
s = 106
Reactions were carried out on a 1.5 mmol scale in THF at 45 °C for 18–32 h at a flow rate of 2–3 ml min–1 using 2 × 5 ml glass columns. Equimolar ratios of the immobilized acylating agents (500 mg, 1.5 mmol g–1
)
and racemic amine (1.5 mmol) were used. Yields correspond to isolated products purified by column chromatography. Enantiomeric ratios were determined by SFC or HPLC on chiral stationary phases. The s
required to obtain given enantiopurities under classical KR conditions are shown.
of the amine (Table 2). To maintain the requisite 45 °C reaction reaction mixing, insufficient swelling of the resin and other
temperature the H-tube reactor was placed on a shaker in a physical factors. These results were rather disappointing, as the
heated oven. Such an experimental setup allowed us to run the enantiopurities were similar to those obtained by classical KR
reactions on a more convenient (1.0–2.0 mmol) scale and using PEARL reagents.
recover the PEARL reagents after each reaction. The amide
products were obtained in good yields albeit with diminished Flow-enabled parallel kinetic resolution. The batch reactor
enantiopurity, although the best results were not reproducible. possessed shortcomings including difficult reaction temperature
The diminished selectivity could be explained by unequal control, as well as solvent evaporation and leaking. To address
4
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2681
ARTICLES
X
N
b
a
X
R
Pd/C, H2
then AcOH 90 °C
X
X
I2 (3.0 equiv.)
O
N
R
THF/H2O 1:1
or
N
R
N
H
R
H
Zn, AcOH 90 °C
NO2
O
13a–e
14a–e
15a–e
16a–e
O
F
O
F
N
H
CO2Et
N
H
CO2Et
N
N
H
N
H
N
H
H
14a 83%
14b 83%
14c 25%
16a 77%
16b 83%
16c 52%
F
F
MeO
MeO
MeO
MeO
NH
NH
NH
NH
S
S
Ph
Ph
16d 40%
16e 64%
14d 89%
14e 80%
c
H
N
H
H
H
N
H
Pd/C
H2
OH
OH
OH
OH
OH
I2 (3.0 equiv.)
N
NO2
N
N
H
NO2
N
N
H
+
+
CO
CO
CO
THF:H2O
1:1
then
AcOH 90 °C
F3C
F3C
N
F3C
N
F3C
F3C
N
CF3
CF3
CF3
CF3
CF3
10a
e.r. 97:3
10b
e.r. 94:6
17
10b
e.r. 94:6
60% yield
18
e.r. 94:6
78% yield
e.r. 97:3
90% yield
Products easily separated
~ 1:1 Mixture of amides
Difficult to separate
Figure 3 | Amide product hydrolysis. a, Pent-4-enoyl amides (1.0 equiv.) were hydrolysed by treatment with I2 (3.0 equiv) in a 1:1 mixture of THF:H2O for
2–8 h. b, 3-(2-nitrophenyl)propanoyl amides were reduced to corresponding aniline derivatives using Pd/C and H2 or Zn/AcOH followed by hydrolysis in AcOH
at 90 °C. Yields correspond to products purified by column chromatography. Product enantiopurity was confirmed by SFC or HPLC analysis on chiral stationary
phases. c, Sequential hydrolysis protocol. The mixture of amides was first treated with I2 (3.0 equiv.) in a 1:1 mixture of THF:H2O. Resulting unprotected
N-heterocycle and 3-(2-nitrophenyl)propanoyl amide were separated and the latter was hydrolysed using Pd/C and H2 followed by heating in AcOH at 90 °C.
Yields correspond to products purified by column chromatography. Product enantiopurity was confirmed by SFC or HPLC analysis on chiral stationary phases.
these issues we turned to advances in flow chemistry. Flow systems resolution cycle. Automation of the column regeneration step
offer multiple practical advantages for reaction setup and control should be possible with more advanced flow instrumentation. The
such as, safer handling of hazardous and highly reactive more dense column seals prevented leakage of the polymer and the
intermediates, better heat and mass transfer, improved reactant adjusted flow rate improved the mixing; column heating was easily
residence time control, reagent and reaction chamber separation and achieved with a standard column-heating block. Overall, the flow
full process automation21–23. Flow processes have been implemented system turned out to be far more reliable and user-friendly than the
for numerous homogenous24–30 and heterogeneous31,32 reactions batch reactors and was essential for practical resolutions (Table 3).
including asymmetric transformations using immobilized reagents
or catalysts33–36
.
Amine deprotection. With sufficient isolated quantities of the
A simple but suitable system for flow PKR was constructed from amides we proceeded to study the hydrolysis with enantioenriched
two glass columns, an HPLC pump and a column heater. The products. Amides are remarkably stable (half-life > 100 years at
columns were charged separately with the acylating agents and room temperature and pH 7)37 and traditional conditions for
sealed with a membrane to ensure that the polymer could not hydrolysis, such as strong acids (HCl or H2SO4) or strong bases
elute. A solution of the amine in THF was cycled through the (NaOH or LiOH), did not afford the amine products in reasonable
columns at a flow rate 2–3 ml min–1 for 24 h at 45 °C. This flow yields. Instead, we utilized olefin and nitro groups to induce an
rate proved optimal for good mixing between the solid and the intramolecular cyclization to form a readily cleaved intermediate.
solution phases. The rate of the mixing is much faster than the The pent-4-enoyl amide could be hydrolysed by treatment with
rate of the acylation allowing the amine to be in simultaneous molecular iodine in an aqueous THF mixture and the
contact with both acylating agents. After the reaction the system corresponding amines could be obtained in excellent yields38. For
was flushed with THF and Et2O and the resulting amides were the 3-(2-nitrophenyl)propanoyl39 derived amides the nitro group
separated. The obtained yields and enantiopurities were in the was reduced to the corresponding aniline followed by
same range or slightly higher compared with the scenario where intramolecular lactamization and concomitant deprotection under
both resins were mixed together in one pot. The flow system simpli- acidic conditions. After screening multiple acids, we found that full
fied polymer recovery and reuse; the columns were disconnected conversion and good yields of the amine could be obtained simply
and the resins separately reacylated and reused for the subsequent by heating the amide in acetic acid. Importantly, the deprotection
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.
5

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2681
ARTICLES
8. Duffey, T. A., MacKay, J. A. & Vedejs, E. Catalytic parallel kinetic resolution
under homogeneous conditions. J. Org. Chem. 75, 4674–4685 (2010).
9. Liao, L. A., Zhang, F., Dmitrenko, O., Bach, R. D. & Fox, J. M. A reactivity/
affinity switch for parallel kinetic resolution: α-amino acid quasienantiomers
and the resolution of cyclopropene carboxylic acids. J. Am. Chem. Soc. 126,
4490–4491 (2004).
10. Kamlet, A. S., Préville, C., Farley, K. A. & Piotrowski, D. W. Regioselective
hydroarylations and parallel kinetic resolution of vince lactam. Angew. Chem.
Int. Ed. 52, 10607–10610 (2013).
11. Wu, B., Parquette, J. R. & RajanBabu, T. V. Regiodivergent ring opening of chiral
aziridines. Science 326, 1662 (2009).
12. Langlois, J. B. & Alexakis, A. Identification of a valuable kinetic process in
copper-catalyzed asymmetric allylic alkylation. Angew. Chem. Int. Ed. 50,
1877–1881 (2011).
of both pent-4-enoyl and 3-(2-nitrophenyl)propanoyl amides
proceeded without detectable epimerization (Fig. 3a,b).
In cases when separation of the amide products after resolution
proved difficult, such as in the case of mefloquine, a sequential
hydrolysis protocol was applied (Fig. 3c). The resolution reaction
mixture, containing an ∼1:1 mixture of enantioenriched amides 25
and 26, was treated with molecular iodine to selectively hydrolyse
the pentenoyl amide 25. The amine product 27 was easily separated
from the unreacted amide 26 and was isolated in high yield (90%
yield in respect to amide 25). The unreacted amide 26 was hydrogen-
ated over Pd/C and heated in AcOH to afford the amine 28 in 78%
yield. Both of these steps proceeded without detectable epimerization.
This protocol proved applicable to other amide products of the PKR.
13. Bertozzi, F., Crotti, P., Macchia, F., Pineschi, M. & Feringa, B. Highly
enantioselective regiodivergent and catalytic parallel kinetic resolution. Angew.
Chem. Int. Ed. 40, 930–932 (2001).
14. Tanaka, K. & Fu, G. C. Parallel kinetic resolution of 4-alkynals catalyzed by Rh
(I)/Tol-BINAP: synthesis of enantioenriched cyclobutanones and
cyclopentenones. J. Am. Chem. Soc. 125, 8078–8079 (2003).
15. Webster, R., Böing, C. & Lautens, M. Reagent-controlled regiodivergent
resolution of unsymmetrical oxabicyclic alkenes using a cationic rhodium
catalyst. J. Am. Chem. Soc. 131, 444–445 (2009).
16. Chavez, D. E. & Jacobsen, E. N. Catalyst-controlled inverse-electron-demand
hetero-Diels-Alder reactions in the enantio- and diastereoselective synthesis of
iridoid natural products. Org. Lett. 5, 2563–2565 (2003).
17. Dehli, J. R. & Gotor, V. Preparation of enantiopure ketones and alcohols
containing a quaternary stereocenter through parallel kinetic resolution of β-
keto nitriles. J. Org. Chem. 67, 1716–1718 (2002).
18. Abril, O. & Whitesides, G. M. Hybrid organometallic/enzymic catalyst
systems: regeneration of NADH using dihydrogen. J. Am. Chem. Soc. 104,
1552–1554 (1982).
19. Doyle, M. P. et al. Highly selective enantiomer differentiation in intramolecular
cyclopropanation reactions of racemic secondary allylic diazoacetates. J. Am.
Chem. Soc. 117, 11021–11022 (1995).
20. Kreituss, I. et al. Robust, recyclable resin for decagram scale resolution of
( )-mefloquine and other chiral N-heterocycles. Angew. Chem. Int.
Ed. 55, 1553–1556 (2015).
21. Pastre, J. C., Browne, D. L. & Ley, S. V. Flow chemistry syntheses of natural
products. Chem. Soc. Rev. 42, 8849–8869 (2013).
22. Atodiresei, I., Vila, C. & Rueping, M. Asymmetric organocatalysis in
continuous flow: opportunities for impacting industrial catalysis. ACS Catal. 5,
1972–1985 (2015).
Discussion
The increasing interest in tandem and cooperative catalysis reactions,
often with catalysts that must be recovered or may be mutually incom-
patible, demands innovative approaches to maintain separation.
Current methods include catalyst separation in different solution
phases40, the use of micelles41,42, or application of heterogeneous or
immobilized catalysts43. By performing reactions in flow, catalyst sep-
aration and regeneration is easily achieved with no detriment to the
reaction yields or selectivities. As far as we are aware, this is the first
example of two asymmetric reactions occurring in parallel, rather
than in sequence, in a flow-based system. This concept should be
easily transferred to many other cases, such as the combination of
enzymes and transition metals44–46 or metals and organic catalysts47.
Importantly, it should work well in cases where the product of one cat-
alyst reaction is the substrate of the other but the catalysts are not com-
patible with one another48,49
.
In conclusion, PKR with reusable PEARL reagents allows access
to both enantiomers of cyclic secondary amines in high enantiopur-
ity. Two complimentary acyl groups have been identified alongside
suitable conditions for their hydrolysis. The use of a user-friendly
flow-based reaction allowed facile separation of the pseudoenantio-
meric products and greatly simplifies the regeneration and reuse of
the reagents. This concept should prove broadly useful in other
chemical processes that benefit from physical separation of distinct
reagents or catalysts.
23. Webb, D. & Jamison, T. F. Continuous flow multi-step organic synthesis. Chem.
Sci. 1, 675–680 (2010).
24. Snead, D. R. & Jamison, T. F. A three-minute synthesis and purification of
ibuprofen: pushing the limits of continuous-flow processing. Angew. Chem. Int.
Ed. 54, 983–987 (2015).
25. Battilocchio, C. et al. Iterative reactions of transient boronic acids enable
sequential C–C bond formation. Nat. Chem. 8, 360–367 (2016).
26. Adamo, A. et al. On-demand continuous-flow production of pharmaceuticals in
a compact, reconfigurable system. Science 352, 61–67 (2016).
27. Chen, M. & Buchwald, S. L. Rapid and efficient trifluoromethylation of aromatic
and heteroaromatic compounds using potassium trifluoroacetate enabled by a
flow system. Angew. Chem. Int. Ed. 52, 11628–11631 (2013).
28. Shu, W. & Buchwald, S. L. Enantioselective β-arylation of ketones enabled by
lithiation/borylation/1,4-addition sequence under flow conditions. Angew.
Chem. Int. Ed. 51, 5355–5358 (2012).
Methods
Two separate 5 ml glass columns were charged with each of the polymeric reagents
(500 mg, ∼1.50 mmol g–1, 0.50 equiv.). The polymers were allowed to swell by
flushing the system with THF at a flow rate of 3 ml min–1 at 45 °C for 15–20 min.
The amine (1.5 mmol, 1.00 equiv.) was flushed through the system for 18–24 h at a
flow rate of 3 ml min–1 and temperature of 45 °C. After the reaction the polymers
were washed with THF (3 × three column volumes) and Et2O (2 × three column
volumes). The amide products were separated by column chromatography and the
polymers were regenerated by treatment with the corresponding acid anhydride.
29. Ganiek, M. A., Becker, M. R., Ketels, M. & Knochel, P. Continuous flow
magnesiation or zincation of acrylonitriles, acrylates, and nitroolefins.
application to the synthesis of butenolides. Org. Lett. 18, 828–831 (2016).
30. Mallia, C. J. & Baxendale, I. R. The use of gases in flow synthesis. Org. Process
Res. Dev. 20, 327–360 (2015).
31. Dong, K., Sun, C. H., Song, J. W., Wei, G. X. & Pang, S. P. Synthesis of
2,6,8,12-tetraacetyl-2,4,6,8,10,12-hexaazaisowurtzitane (TAIW) from 2,6,8,12-
tetraacetyl-4,10-dibenzyl-2,4,6,8,10,12-hexaazaisowurtzitane (TADBIW) by
catalytic hydrogenolysis using a continuous flow process. Org. Process Res. Dev.
18, 1321–1325 (2014).
32. Baxendale, I. R., Ley, S. V., Mansfield, A. C. & Smith, C. D. Multistep synthesis
using modular flow reactors: Bestmann-Ohira reagent for the formation of
alkynes and triazoles. Angew. Chem. Int. Ed. 48, 4017–4021 (2009).
33. Annis, D. A. & Jacobsen, E. N. Polymer-supported chiral Co (Salen) complexes:
synthetic applications and mechanistic investigations in the hydrolytic kinetic
resolution of terminal epoxides. J. Am. Chem. Soc. 121, 4147–4154 (1999).
34. Adint, T. T. & Landis, C. R. Immobilized bisdiazaphospholane catalysts for
asymmetric hydroformylation. J. Am. Chem. Soc. 136, 7943–7953 (2014).
35. Kamahori, K., Ito, K. & Itsuno, S. Asymmetric diels-Alder reaction of
methacrolein with cyclopentadiene using polymer-supported catalysts: design of
highly enantioselective polymeric catalysts. J. Org. Chem. 61, 8321–8324 (1996).
Received 8 May 2016; accepted 18 October 2016;
published online 12 December 2016
References
1. Keith, J. M., Larrow, J. F. & Jacobsen, E. N. Practical considerations in kinetic
resolution reactions. Adv. Synth. Catal. 343, 5–26 (2001).
2. Kagan, H. B. & Fiaud, J.-C. Kinetic resolution. Top. Stereochem. 18,
249–330 (1988).
3. Vedejs, E. & Jure, M. Efficiency in nonenzymatic kinetic resolution. Angew.
Chem. Int. Ed. Engl. 44, 3974–4001 (2005).
4. Russell, T. A. & Vedejs, E. in Separation of enantiomers: synthetic methods
(ed. Todd, M.) 216–262 (Wiley-VCH, 2014).
5. Vedejs, E. & Chen, X. Parallel kinetic resolution. J. Am. Chem. Soc. 119,
2584–2585 (1997).
6. Dehli, J. R. & Gotor, V. Parallel kinetic resolution of racemic mixtures: a new
strategy for the preparation of enantiopure compounds? Chem. Soc. Rev 31,
365–370 (2002).
7. Vedejs, E. & Rozners, E. Parallel kinetic resolution under catalytic conditions : a
three-phase system allows selective reagent activation using two catalysts. J. Am.
Chem. Soc. 123, 2428–2429 (2001).
6
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.

NATURE CHEMISTRY DOI: 10.1038/NCHEM.2681
ARTICLES
36. Csajagi, C. et al. Enantiomer selective acylation of racemic alcohols by lipases in
continuous-flow bioreactors. Tetrahedron Asymmetry 19, 237–246 (2008).
37. Hutchby, M. et al. Switching pathways: room-temperature neutral solvolysis and
substitution of amides. Angew. Chem. Int. Ed. 51, 548–551 (2012).
38. Debenham, J. S. Madsen, R. Roberts, C. & Fraser-Reid, B. Two new orthogonal
amine-protecting groups that can be cleaved under mild or neutral conditions.
J. Am. Chem. Soc. 117, 3302–3303 (1995).
47. Allen, A. E. & MacMillan, D. W. C. Synergistic catalysis: a powerful synthetic
strategy for new reaction development. Chem. Sci. 3, 633–658 (2012).
48. Hafez, A. M., Taggi, A. E., Dudding, T. & Lectka, T. Asymmetric catalysis on
sequentially-linked columns. J. Am. Chem. Soc. 123, 10853–10859 (2001).
49. Hafez, A. M., Taggi, A. E. & Lectka, T. Column asymmetric catalysis. Chem.
Eur. J. 8, 4114–4119 (2002).
39. Entwistle, I. D. The use of 2-nitrophenylpropionic acid as a protecting group for
amino and hydroxyl functions to be recovered by hydrogen transfer
hydrogenation. Tetrahedron Lett. 6, 555–558 (1979).
40. Shirakawa, S. & Maruoka, K. Recent developments in asymmetric phase-transfer
reactions. Angew. Chem. Int. Ed. 52, 4312–4348 (2013).
Acknowledgements
This work was supported by the European Research Council (ERC Starting Grant no.
306793-CASAA) and the Swiss Federal Commission for Technology and Innovation
(CTI 15523.1). S.-Y. Hsieh, B. Wanner and K.-Y. Chen (all ETH) are thanked for helpful
discussions and advice with flow-based systems.
41. Isley, N. A., Linstadt, R. T. H., Kelly, S. M., Gallou, F. & Lipshutz, B. H.
Nucleophilic aromatic substitution reactions in water enabled by micellar
catalysis. Org. Lett. 17, 4734–4737 (2015).
42. La Sorella, G., Strukul, G. & Scarso, A. Recent advances in catalysis in micellar
media. Green Chem. 17, 644–683 (2015).
Author contributions
J.W.B. and I.K. contributed equally to the design of the study. J.W.B. and I.K. co-wrote the
paper. I.K. performed the experiments and wrote the Supplementary Information.
43. Kobayashi, S. Flow ‘fine’ synthesis: high yielding and selective organic synthesis
by flow methods. Chem. Asian J. 11, 425–436 (2016).
44. Hyster, T. K., Knorr, L., Ward, T. R. & Rovis, T. Biotinylated Rh(III) complexes
in asymmetric C–H activation. Science 338, 500–503 (2012).
45. Paetzold, J. & Bäckvall, J. E. Chemoenzymatic dynamic kinetic resolution of
primary amines. J. Am. Chem. Soc. 127, 17620–17621 (2005).
Additional information
Supplementary information and chemical compound information are available in the
online version of the paper. Reprints and permissions information is available online at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to J.W.B.
46. Denard, C. A. et al. Development of a one-pot tandem reaction combining
ruthenium-catalyzed alkene metathesis and enantioselective enzymatic
oxidation to produce aryl epoxides. ACS Catal. 5, 3817–3822 (2015).
Competing financial interests
The authors declare no competing financial interests.
NATURE CHEMISTRY | ADVANCE ONLINE PUBLICATION | www.nature.com/naturechemistry
7
© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.