On-Column Reaction Set-Up for High-Throughput Screenings and Mechanistic Investigations

Article abstract of DOI:10.1002/adsc.201500311

A screening platform, which offers a high-throughput approach as well as an easy investigation of kinetic isotope effects, applicable to a wide range of reactions is presented. To illustrate the high potential of this approach, the asymmetric transfer hydrogenation of methyl benzoylformate with copper(II) bis(oxazoline) and Hantzsch ester was examined. Accordingly, the enantioselectivities of the reaction performed on-column in a microcapillary were comparable to standard reaction conditions, however, we were able achieve catalysis and analysis in a single step in less than 30 min. The throughput can be increased by simultaneous investigation of different substrates without increasing the overall analysis time. Use of di-deuterated Hantzsch ester allowed us to investigate the kinetic isotope effect of the transfer hydrogenation reaction only requiring a minute amount of the deuterated transfer hydrogenation reagent. Hence we were able to get further insights into the mechanism of the asymmetric transfer hydrogenation using Hantzsch ester as hydrogen source. The here presented technique is broadly applicable to study isotope effects on a very small scale, which is a rapid and an inexpensive alternative compared to conventional experiments.

Full text of DOI:10.1002/adsc.201500311

FULL PAPERS
DOI: 10.1002/adsc.201500311
Very Important Publication
On-Column Reaction Set-Up for High-Throughput Screenings
and Mechanistic Investigations
a
a
a
a,
Skrollan Stockinger, Johannes Troendlin, Frank Rominger, and Oliver Trapp *
a
Organisch-Chemisches Institut, Ruprecht-Karls Universität Heidelberg, Im Neuenheimer Feld 270, 69120 Heidelberg,
Germany
Fax: (+49)-6221-544-904; e-mail: trapp@oci.uni-heidelberg.de
Received: March 28, 2015; Revised: June 5, 2015; Published online: September 11, 2015
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201500311.
Abstract: A screening platform, which offers a high- allowed us to investigate the kinetic isotope effect of
throughput approach as well as an easy investigation the transfer hydrogenation reaction only requiring
of kinetic isotope effects, applicable to a wide range a minute amount of the deuterated transfer hydroge-
of reactions is presented. To illustrate the high po- nation reagent. Hence we were able to get further in-
tential of this approach, the asymmetric transfer hy- sights into the mechanism of the asymmetric transfer
drogenation of methyl benzoylformate with cop- hydrogenation using Hantzsch ester as hydrogen
per(II) bis(oxazoline) and Hantzsch ester was exam- source. The here presented technique is broadly ap-
ined. Accordingly, the enantioselectivities of the re- plicable to study isotope effects on a very small
action performed on-column in a microcapillary scale, which is a rapid and an inexpensive alternative
were comparable to standard reaction conditions, compared to conventional experiments.
however, we were able achieve catalysis and analysis
in a single step in less than 30 min. The throughput Keywords: asymmetric transfer hydrogenation; cop-
can be increased by simultaneous investigation of per(II) bis(oxazoline); Hantzsch ester; high-through-
different substrates without increasing the overall put; kinetic isotope effect; on-column reaction gas
analysis time. Use of di-deuterated Hantzsch ester chromatography
Introduction
phoresis (CE) for highly efficient and fast separations,
[5]
can also be used as microreactor devices. Coating of
To optimize the yield and selectivity of a reaction re- the inner surface of a capillary with a catalyst species
quires the detailed knowledge of the mechanism and enables one to screen catalyzed reactions. This inte-
all factors influencing the reaction. Conventionally, gration of catalytic activity and separation selectivity
this information is obtained by tedious synthetic and in a single or consecutive microreactor column offers
consequent analytical steps in combination with sys- many advantages, like the investigation of intercon-
tematic variation of the reaction conditions, reactants, versions of stereoisomers by dynamic chromatogra-
[6]
catalysts and their concentrations. However, this pro- phy or as screening platform under varying reaction
[5]
cedure is often highly time consuming and associated conditions, substrates or catalysts. Furthermore, the
with high experimental costs. To minimize the overall screening process can be accelerated by simul-
number of steps, methods for a statistical experimen- taneous investigation of several reactants at the same
tal design were developed and represent a highly time by injection of reactant libraries, because of the
[
1]
promising approach. Despite this progress there is simultaneous separation no competitive reactions can
a demand for techniques to speed up the monitoring take place allowing the determination of reaction ki-
process, for example, by combination of synthesis and netics and selectivities under exactly the same reac-
[7]
analysis in a single step. In this area microfluidic devi- tion conditions. Additionally, such a set-up is pre-
ces (lab on a chip) and flow-through microreactors destinated for mechanistic investigations using isotope
[
2]
are broadly used for various applications in organic labelled compounds because only small amounts are
[
3]
[4]
synthesis and total analysis. Microcapillaries used required and analysis by mass spectrometry can be di-
routinely in gas chromatography (GC), open-tubular rectly performed.
liquid chromatography (OTLC) and capillary electro-
Adv. Synth. Catal. 2015, 357, 3513 – 3520
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3513

Skrollan Stockinger et al.
FULL PAPERS
Figure 1. Schematic overview of the reaction process taking place in a chromatographic reactor based on the theoretical
plate model.
Here we present an on-column reaction gas chro- and phase shift from one plate to the next plate takes
matographic (ocRGC) set-up applicable for high- place.
throughput measurements as well as for mechanistic
The experimental set-up we used in this study con-
investigations based on kinetic isotope experiments. sists of the catalytically active capillary coupled be-
[8]
We chose the asymmetric transfer hydrogenation of tween a pre-separation capillary (GE-SE-30; length
ketones by a copper(II) bis(oxazoline) catalyst and 0.5 m, I.D. 250 mm; film thickness 500 nm) and an
Hantzsch ester to demonstrate the features of such an enantioselective separation capillary (chiral stationary
on-column reaction chromatographic set-up.
phase heptakis-(2,6-di-O-methyl-3-O-pentyl)-b-cyclo-
[10]
dextrin; length 1.5–8 m, I.D. 250 mm; film thickness
50 nm). The pre-separation column achieves an opti-
mal equilibration after the injection (split injection at
508C) and spatial separation of the injected reac-
2
Results and Discussion
2
Establishing the Transfer Hydrogenation Screening
Set-Up
tants, enabling high-throughput investigations due to
the absence of competing reactions, whereas the
(
post)-separation column enables a complete separa-
To transfer the conventional batch reaction set-up to tion of the formed enantiomers. Reagents and prod-
a nanoliter capillary reactor we prepared in a first ucts were detected by flame ionization detection
step catalytically active capillaries with Cu(II) bis(ox- (FID) for quantification and identified by MS (quad-
azoline) {2,2-bis[(4R)-benzyl-4,5-dihydro-1,3-oxazol-2- rupole–ion trap MS) in a single step. The absolute
[11]
yl)propane]copper(II) bistriflate} (R,R)-1 embedded configuration
of the catalytically formed major
as catalyst and Hantzsch ester 2 as hydrogen donor. enantiomers was assigned by co-injection of the pure
Catalyst and transfer hydrogenation reagent were dis- (S)-enantiomers. Optimal reaction and separation
solved in an inert polysiloxane of medium polarity conditions were found to be at 608C and 250 kPa
(
8
PS086 containing 12–15% diphenylsiloxy and 85– helium to achieve reaction and separation at the same
8% dimethylsiloxy groups) and coated on the inner time. A temperature increase to 808C leads to an 8%
surface of fused-silica capillaries (I.D. 250 mm) by the decrease of conversion, which can be attributed to
[
9]
static method described by Grob resulting in a cata- thermal decomposition of the catalyst (see the Sup-
lytically active film with a defined thickness of porting Information). The enantioselectivity of the re-
2
9
50 nm. The final (R,R)-1 catalyst loading was only action shows on the other hand no temperature de-
À5
À10
À4
.82”10 mg (1.36”10 mol) and 2.62”10 mg pendence (30–808C). The slight decrease in enantio-
À9
(
1.03”10 mol) for the Hantzsch ester 2 per cm of selectivity compared to the results published by List
[8]
capillary. These nanoliter capillary reactors were em- et al. might be contributed to the here applied
ployed for asymmetric transfer hydrogenations in an higher reaction temperatures. Remarkably, the reac-
on-column reactions gas chromatographic set-up tor capillary, stored under argon, was stable for sever-
(
ocRGC), combining stereoselective catalysis and ste- al days without any loss of catalytic activity or selec-
reoselective analysis in a single step. Figure 1 shows tivity (see the Supporting Information).
schematically the theoretical background of this set-
up, considering the theoretical plates of a chromato-
graphic separation column as chromatographic flow-
through reactor, where phase distribution, catalysis
3
514
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 3513 – 3520

FULL PAPERS
On-Column Reaction Set-Up for High-Throughput Screenings
[
a]
Table 1. Summarized results of the asymmetric transfer hydrogenation by enantioselective ocRGC.
Entry
Substrate
Product
Major Product
Conversion [%]
er [%]
1
(S)
>99.5
79:21
2
3
(S)
(S)
98.8
76:24
77:23
>99.5
4
(S)
>99.5
65:35
[
[
[
b]
b]
c]
5
6
7
(S)
(S)
(S)
49.9
43.3
26.4
70:30
70:30
67:33
[
[
[
a]
Reaction conditions: 608C, inlet pressure 250 kPa He as inert carrier gas; experimental set-up: see text or the Supporting
Information.
Chiral stationary phase of the separation capillary was heptakis-(2,6-di-O-methyl-3-O-pentyl)-b-cyclodextrin; length
b]
c]
11.0 m; I.D. 250 mm; film thickness 250 nm at 608C, 40 kPa He.
Inlet pressure 40 kPa.
Substrate Screening
by ocRGC we performed the (R,R)-1-catalyzed asym-
metric transfer hydrogenation reactions in a flask
Various aromatic and aliphatic a-keto esters (Table 1) under conventional reaction conditions with four of
were investigated as substrates. Aryl alkyl a-keto the on-column analyzed a-keto esters. Because of its
esters were hydrogenated with high conversions and similar polarity compared to polysiloxane, toluene
moderate enantioselectivites. EWG groups, for exam- was chosen as solvent and the reaction temperature
ple, F,F groups, led to a slight decrease in the catalytic was set to room temperature. The obtained selectivi-
selectivity. Hydrogenation of the investigated aliphat- ties were comparable to the results published by List
[8]
ic a-keto ester resulted in lower conversions and et al. and were slightly higher than the selectivities
enantioselectivities. determined by ocRGC (Figure 2). In agreement with
Even with capillaries as long as only 12 cm we ob- the on-column measurements, the F,F group at the ar-
served very high conversions (>96%) at contact omatic system lowered the catalystꢁs selectivity. Addi-
times as short as 55 s because of the high specific in- tionally, no temperature influence on the selectivity
terfacial area per volume. In comparison, ocRGC re- was observed in the investigated temperature range
actions performed under conventional conditions re- between 258C and 608C. Interestingly, ketopantolac-
quire much longer reaction times to achieve these tone gave not only lower enantioselectivities com-
[
8]
conversions.
pared to the here investigated aromatic a-keto esters
Performing hydrogenation experiments using differ- but was also in very good agreement with the results
ent capillary lengths (1.1, 2.3, 4.0, 5.2, 7.0, 10.0 cmÆ obtained by ocRGC catalysis.
0.05 cm) at a constant loading of (R,R)-1 resulted in
To demonstrate the possibility of achieving a higher
contact times between 8 and 54 sÆ0.3 s. Here, the throughput, a small substrate library containing 3 dif-
contact time and conversion correlate linearly with ferent a-keto esters (entries 1, 3 and 4 in Table 1) was
the length of the capillary (see the Supporting Infor- simultaneously injected onto the experimental set-up.
mation). To compare the results, which we obtained All three compounds were stereoselectively trans-
Adv. Synth. Catal. 2015, 357, 3513 – 3520
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3515

Skrollan Stockinger et al.
FULL PAPERS
Study of Isotope Effects by On-Column Reaction Gas
Chromatography
To demonstrate the advantages for mechanistic stud-
ies we investigated the kinetic isotope effect using at
position
4
di-deuterated Hantzsch ester D -2
2
(
Figure 4). Deuterated Hantzsch ester was synthe-
[12]
sized according to Westheimer et al.
and crystal-
lized in the monoclinic space group C2m in contrast
¯
to the triclinic P1 space group described in the litera-
[13]
ture.
The di-deuterated Hantzsch ester D -2 was
2
used under the same conditions like the undeuterated
Figure 2. Comparison of enantioselectivites achieved by
batch catalysis in a flask and ocRGC catalysis.
formed to the corresponding (S)-a-hydroxy ester with
complete conversion by performing the reaction in
a 50-cm reactor capillary. This capillary length was
chosen to achieve a complete conversion of all sub-
strates.
It has to be pointed out that catalysis and analysis
could be achieved in less than 34 min and each prod-
uct could be identified by three-dimensional MS anal-
ysis (Figure 3). Mass spectrometric detection allows
us to deconvolute overlapping peaks of different com-
pounds eluting at the same retention time and makes
quantitative analysis and determination of enantio-
meric ratios even in complex cases possible.
Figure 4. Conversions in dependence on the capillary length
of methyl benzoylformate 4 in presence of the catalyst
(
R,R)-1 and Hantzsch ester 2 or D -2 dissolved in an inert
2
polysiloxane matrix.
Figure 3. An example of a high-throughput measurement of a substrate library containing 3 different a-keto esters (entries 1,
and 4 in Table 1), which were simultaneously injected.
3
3
516
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 3513 – 3520

FULL PAPERS
On-Column Reaction Set-Up for High-Throughput Screenings
measurements, the batch catalysis showed a kinetic
isotope effect of 1.51Æ0.29.
Hitherto only investigations of kinetic isotope ef-
fects of the hydrogenation of ethyl a-cyanocinna-
mates and benzylidenemalononitriles with Hantzsch
ester under non-catalytic conditions have been de-
[14]
scribed. Kinetic isotope effects in the range of 1.2
to 1.3 for mono-deuterated Hantzsch ester at position
1
(ND) and 5.3 to 6.0 for an at position 4 di-deuterat-
ed Hantzsch ester D -2 were observed. Consequently
2
it was assumed that the CÀH bond dissociation at po-
sition 4 of the Hantzsch ester is the rate-limiting step,
[15]
in terms of a direct hydride transfer.
Figure 5. Comparison of batch (offline) and ocRGC cataly-
Taking into account that we only detect a kinetic
isotope effect for the batch reaction, it seems that in
the ocRGC this rate-limiting step is already complet-
ed before the substrate is converted into the product.
Therefore no kinetic isotope effect is detectable. The
only difference is that in ocRGC the catalyst (R,R)-
sis using either Hantzsch ester 2 or D -2. For the tabulated
2
values obtained by the online measurements reactor capilla-
ries with 2 to 10 cm for the category <10 cm and 15 to
40 cm for the category>10 cm are summarized. The conver-
sion trend related to all measured capillary lengths is plotted
as Figure 4.
1
together with the Hantzsch esters 2 or D -2 are em-
2
bedded in an inert polysiloxane. We made the obser-
compound 2 for the ocRGC catalysis as well as for vation that if methyl benzoylformate 4 is added to
the batch synthesis of methyl benzoylformate 4. The a pre-mixture of catalyst (R,R)-1 and Hantzsch ester
mono-deuterated methyl mandelate D-5 was identi- 2 or D -2 in the conventional batch experiment with-
2
fied as product (Figure 5 and Supporting Informa- out the addition of a polysiloxane, the reaction is sig-
tion). In ocRGC the conversions for both Hantzsch nificantly slowed down compared to the classical pro-
esters depend on the capillary lengths as expected tocol where methyl benzoylformate 4 is added to cat-
and complete conversions are achieved for 10 and alyst (R,R)-1 followed by the Hantzsch ester. Inter-
15 cm capillary lengths (Figure 4), respectively. Statis- estingly, if polysiloxane is present no influence on the
tical analysis of the experimental data shows an excel- reaction rate is detectable when the reactant order is
lent agreement of the dependences in the range of ex- changed. Goodman et al. proposed for the hydrogena-
perimental error (Table in Figure 5). Surprisingly, this tion of imines using Hantzsch ester under BINOL-
suggests that the conversions are isotope independent. phosphoric acid catalysis the formation of a three
As shown in Figure 5 also the enantioselectivity molecules-containing transition state, where the imine
seems to be independent. Only when shorter capilla-
ries (<10 cm) and di-deuterated Hantzsch ester D -2
2
were employed did the enantiomeric ratio increase
slightly from 79:21 to 81:19 (D=4% ee). Although
the conversion is not affected by the isotope labelling,
it suggests that the enantioselectivity is influenced.
Generally we observed that the enantioselectivities of
ocRGC are slightly lower compared to conventional
batch catalysis, which is due to higher reaction tem-
peratures in ocRGC. Thus the increase of 4% ee by
using Hantzsch ester D -2 can be explained by an iso-
2
tope effect, which only affects the post-hydrogenation
racemization and therefore is independent of the orig-
inal hydrogenation rate. With increasing capillary
length there is a higher probability to interact with
the catalysts, which in turn decreases the influence of
the isotope effect, resulting in a constant enantiomeric
ratio for deuterated and non-deuterated Hantzsch
esters.
For comparison, the conventional batch catalysis
was investigated for the non-deuterated 2 and the
Figure 6. Calculated transition state (DFT calculations at
the B3LYP/6-31G** level of theory) for the transfer hydro-
genation of methyl benzoylformate 4 with (R,R)-1 and
Hantzsch ester 2. The path of the hydride transfer is illus-
deuterated Hantzsch ester D -2 under comparable trated by a dotted green line. For clarity the innocent hydro-
2
conditions (Figure 5). In contrast to the on-column gen atoms were not plotted.
Adv. Synth. Catal. 2015, 357, 3513 – 3520
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3517

Skrollan Stockinger et al.
FULL PAPERS
and the Hantzsch ester are coordinated across the ni- volatility of the substrates and reaction products,
trogen proton to the phosphoric acid moiety of the which is mandatory to obtain quantitative data.
[
16]
catalyst.
Transferring this concept to the system
presented here we get a transition state as depicted in
Figure 6. Here, both the Hantzsch ester and the Experimental Section
methyl benzoylformate 4 are coordinated to Cu of
catalyst (R,R)-1. According to the mentioned kinetic Materials and Methods
influence in the batch synthesis and the ocRGC ex-
a-Keto esters which were used for ocRGC experiments
periment, the rate-limiting step has to be the coordi-
nation of the Hantzsch ester to catalyst (R,R)-1. In
the ocRGC approach such a complex can be pre-
were purchased from Sigma–Aldrich (St. Louis, USA) and
were used without any further purification. Ethyl pyruvate
and methyl pyruvate were freshly distilled before use. The
formed. This explanation fits to the batch synthesis inert polysiloxane PS086 which was used as stabilizing poly-
experiments applying a different reactant order mer matrix to prepare catalyst capillaries was purchased
whereas when catalyst and Hantzsch ester were pre- from ABCR chemicals (Karlsruhe, Germany). Anhydrous
dichloromethane was used after boiling over calcium hy-
mixed in presence of the polysiloxane matrix before
dride and distillation under an inert atmosphere of argon.
addition of methyl benzoylformate. In this case no
GC and GC-MS-measurements were performed on
isotope effect was detectable. In addition without
a Thermo Trace GC-PolarisQ MS (Thermo, San Jose, Cali-
polysiloxane the reaction was significantly slower.
fornia, USA) equipped with a split/splitless (SSL) injector
(at 2508C), an autosampler (Thermo AS 3000) a flame ioni-
zation detector (FID, operated at 2508C) and a quadrupole–
ion trap mass spectrometer. Electron impact mass data were
recorded at an ion source temperature of 2008C and elec-
Conclusions
In summary, we have presented a reactor set-up, tron energy of 70 eV. Gas chromatograms and mass traces
which integrates catalysis with direct analysis in were recorded using the Xcalibur software package
(
Thermo, San Jose, California, USA). Enantioselective GC
a nanoliter capillary and which can be used as a pow-
erful tool to rapidly explore kinetic data (e.g., reactiv-
ity and enantioselectivity) as well as mechanistic data
was performed using fused-silica columns coated with the
chiral stationary phase (CSP) heptakis-(2,6-di-O-methyl-3-
O-pentyl)-b-cyclodextrin
O.D. 365 mm, 250 nm film thickness). Helium was used as
inert carrier gas. Fused-silica capillaries to prepare catalyti-
cally active reactor columns (I.D. 250 mm, O.D. 365 mm)
were purchased from Microquartz (Munich, Germany).
[
10]
(length 1.5–11 m, I.D. 250 mm,
(
e.g., isotope labelling experiments). Therefore both
catalyst and hydrogen donor were dissolved in an
inert polysiloxane matrix. In contrast to previously
published results, a time-consuming modification of
the catalystꢁs ligand backbone was not necessary to
obtain an immobilized catalytic system. The here de-
scribed catalytically active system is easy to handle,
stable for several days and shows great activity exem-
plified in quantitative conversions in less than 60 s.
Both aromatic and aliphatic a-keto esters were enan-
tioselectively hydrogenated to the corresponding
Preparation of 4,4’-D -Hantzsch Ester (D -2)
2
2
4
,4’-Dideuterio-2,6-dimethyl-3,5-dicarboethoxy-1,4-dihydro-
pyridine D -2 was prepared according to the procedure re-
ported by Norcross, Klinedinst and Westheimer. Parafor-
maldehyde-d2 (100 mg, 3.10 mmol), p-toluenesulfonic acid
2
[12]
(2.30 mg, 10.0 mmol), ethyl acetoacetate (754 mg, 5.80 mmol)
products. High-throughput applications can be realiz- and ammonia (2.00 mL of 2N solution in ethanol, 3.2 mmol)
ed with this set-up as demonstrated by injection of was transformed to 4,4’-D Hantzsch ester; yield: 298 mg
2
1
a substrate library containing 3 different substrates, (1.17 mmol, 38%). H NMR (600 MHz, CDCl
): d=1.28 (t,
3
J=7.15 Hz, 6H), 2.19 (s, 6H), 4.17 (q, J=6.97 Hz, 4H), 5.21
bs, 1H); HR-MS (EI+): m/z=255.1435, calcd. for
C H D NO [M]: 255.1440. Crystal data: C H NO , M =
which reduced the overall experimental time to just
(
34 min.
1
3
17
2
4
13 19
4
r
By straightforward use of deuterated Hantzsch
3
2
53.29, 0.25”0.24”0.15 mm , monoclinic, space group C2m,
a=24.3328(5) Š, b=6.9207(2) Š, c=7.4758(2) Š, b=
1.6095(10)8, V=1258.43(6) Š , Z=4, d
l=0.71073 Š, T=200(2) K, qrange =1.674–27.5078. Reflec-
tions measured 18735, independent 1557, Rint =0.0228. Final
ester as transfer hydrogenation reagent kinetic iso-
tope effects could be analyzed. In combination with
the chosen model reaction it enabled us not only to
synthesize mono-deuterated a-hydroxy esters with
3
À3
9
=1.337 gcm ,
calcd.
high stereoselectivity but also to propose a model for R indices [l>2s(l)]: R1=0.043, wR2=0.132. Details of data
the asymmetric transfer hydrogenation using Cu(II) collection, structure solution and refinement of crystal struc-
bis(oxazoline) catalyst (R,R)-1 with Hantzsch ester 2. tures of D -2 are contained in CCDC 1016121. These data
2
can be obtained free of charge from The Cambridge Crystal-
lographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
The here presented reactor set-up is not limited to
the here investigated reaction, but can be used for
other catalytic transformations using a reactive carrier
gas, i.e., hydrogen in hydrogenations, or using re-
agents in the stationary phase. Limitations are the
3
518
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 3513 – 3520

FULL PAPERS
On-Column Reaction Set-Up for High-Throughput Screenings
Preparation of Cu(II) Bis(oxazoline) Complex (R,R)-
1
clodextrin (1.5–11.0 m, I.D. 250 mm, film thickness 250 nm).
Helium was used as inert carrier gas. All measurements
were repeated 3 times.
{2,2-Bis[(4R)-benzyl-4,5-dihydro-1,3-oxazol-2-yl]-propane}-
copper(II) bistriflate (R,R)-1 was prepared according to the
[8]
procedure reported by List et al. Cu(OTf)2 (5.80 mg,
6.0 mmol) and 2,2-bis[(4R)-4-benzyloxazoline]propane
5.80 mg, 16.0 mmol) were added to a flame-dried Schlenk
1
(
Acknowledgements
tube. After drying the mixture under vacuum for 0.5 h fresh-
ly distilled dichloromethane (2.0 mL) was added and the
mixture was stirred for 1 h.
We thank the European Research Council for financial sup-
port (ERC Grant Agreement No. StG 258740). We thank Ute
Gärtner for coating fused-silica capillaries.
Coating of Catalytically Active Capillaries
(
(
R,R)-1, freshly prepared by the procedure described above References
0.6 mg, 0.82 mmol), Hantzsch ester 2 (1.6 mg, 6.2 mmol) or
D -2 (1.6 mg, 6.2 mmol) and PS086 (12 mg) were dissolved in
anhydrous dichloromethane (3 mL) and fused-silica capilla-
2
ries were coated with this solution by the static method de-
[
9]
scribed by Grob. This capillary (I.D. 250 mm, O.D. 365 mm)
was filled with this solution and the solvent was removed
under high vacuum after closing one end of the capillary
with silicon to obtain a 250 nm polymer film on the inner
surface of the capillary. Afterwards the capillary was flushed
with argon for several minutes at room temperature to
remove volatile components.
On-Column Reaction Gas Chromatographic
Experiments of the Asymmetric Transfer
Hydrogenation
On-column reaction gas chromatographic experiments of
the asymmetric transfer hydrogenation were performed by
enantioselective gas chromatography on a Thermo Trace Po-
larisQ GC-MS. For the stereoselective asymmetric transfer
hydrogenation of the a-keto esters a fused-silica capillary
(0.01–1 m in length) coated with the catalyst (R,R)-1 and
the Hantzsch ester 2 or D -2 embedded in the polysiloxane
2
matrix PS086 (I.D. 250 mm, film thickness 250 nm) was em-
ployed. This reactor column was coupled between a pre-sep-
aration capillary coated with GE-SE-30 (0.50 m, I.D.
250 mm, film thickness 500 nm) and an enantioselective sep-
aration capillary coated with heptakis-(2,6-di-O-methyl-3-O-
pentyl)-b-cyclodextrin (1.5–11.0 m, I.D. 250 mm, film thick-
ness 250 nm) to quantify reactants and products of the hy-
drogenation. Helium was used as inert carrier gas. All meas-
urements were repeated 3 times.
Batch Reactions of the Asymmetric Transfer
Hydrogenation
The batch synthesis measurements were realized based on
[8]
the work of List et al. Therefore freshly prepared (R,R)-
(1.8 mg, 5.0 mmol), following the procedure described
1
above with chloroform as solvent unless otherwise indicated,
were added to methyl benzoylformate 4 (4.1 mg, 25 mmol).
After addition of Hantzsch ester 2 (8.86 mg, 35 mmol) or D2-
[4] a) J. Knight, Nature 2002, 418, 474–475; b) D. Janasek,
J. Franzke, A. Manz, Nature 2006, 442, 374–380; c) S. J.
Haswell, Nature 2006, 441, 705–705; d) S. Fritzsche, S.
Ohla, P. Glaser, D. S. Giera, M. Sickert, C. Schneider,
D. Belder, Angew. Chem. 2011, 123, 9639–9642; Angew.
Chem. Int. Ed. 2011, 50, 9467–9470.
[5] a) O. Trapp, S. K. Weber, S. Bauch, W. Hofstadt,
Angew. Chem. 2007, 119, 7447–7451; Angew. Chem. Int.
Ed. 2007, 46, 7307–7310; b) O. Trapp, S. K. Weber, S.
2
(8.94 mg, 35 mmol) the sealed vials were stirred for 19 h
under argon atmosphere. After filtration over celite the
sample was analyzed by GC/MS using a pre-separation ca-
pillary coated with GE-SE-30 (0.50 m, I.D. 250 mm, film
thickness 500 nm) and an enantioselective separation capil-
lary coated with heptakis(2,6-di-O-methyl-3-O-pentyl)-b-cy-
Adv. Synth. Catal. 2015, 357, 3513 – 3520
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3519

Skrollan Stockinger et al.
FULL PAPERS
Bauch, T. Bäcker, W. Hofstadt, B. Spliethoff, Chem.
Eur. J. 2008, 14, 4657–4666; c) S. Stockinger, O. Trapp,
Chirality 2014, 26, 243–248; d) O. Trapp, S. Bremer,
S. K. Weber, Anal. Bioanal. Chem. 2009, 395, 1673–
Organic Synthesis II, Vol. 9, (Eds.: G. A. Molander, P.
Knochel), Elsevier, Oxford, 2014, pp 94–110.
[8] J. W. Yang, B. List, Org. Lett. 2006, 8, 5653–5655.
[9] K. Grob, Making and Manipulating Capillary Columns
for Gas Chromatography, Dr. Alfred Hüthig Verlag,
Heidelberg, Basel, New York, 1986.
[10] W. A. Kçnig, B. Gehrcke, D. Icheln, J. Dçnneke, W.
Wang, J. High Res. Chromatogr. 1992, 15, 367–372.
[11] a) P. Herwig, K. Zawatzky, M. Grieser, O. Heber, B.
Jordon-Thaden, C. Krantz, O. Novotny, R. Repnow, V.
Schurig, D. Schwalm, Z. Vager, A. Wolf, O. Trapp, H.
Kreckel, Science 2013, 342, 1084–1086; b) P. Herwig, K.
Zawatzky, D. Schwalm, M. Grieser, O. Heber, B.
Jordon-Thaden, C. Krantz, O. Novotny, R. Repnow, V.
Schurig, Z. Vager, A. Wolf, O. Trapp, H. Kreckel, Phys.
Rev. A 2014, 90, 052503; c) K. Zawatzky, P. Herwig, M.
Grieser, O. Heber, B. Jordon-Thaden, C. Krantz, O.
Novotny, R. Repnow, V. Schurig, D. Schwalm, Z.
Vager, A. Wolf, H. Kreckel, O. Trapp, Chem. Eur. J.
2014, 20, 5555–5558.
1679; e) S. K. Weber, S. Bremer, O. Trapp, Chem. Eng.
Sci. 2010, 65, 2410–2416; f) J. Troendlin, J. Rehbein, M.
Hiersemann, O. Trapp, J. Am. Chem. Soc. 2011, 133,
16444–16450; g) C. Lang, U. Gärtner, O. Trapp, Chem.
Commun. 2011, 47, 391–393; h) S. Fuessl, O. Trapp,
Electrophoresis 2012, 33, 1060–1067; i) S. Sandel, S. K.
Weber, O. Trapp, Chem. Eng. Sci. 2012, 83, 171–179;
j) S. Stockinger, O. Trapp, Beilstein J. Org. Chem. 2013,
9
, 1837–1842; k) J. Gmeiner, M. Seibicke, C. Lang, U.
Gärtner, O. Trapp, Adv. Synth. Catal. 2014, 356, 2081–
087.
6] a) O. Trapp, G. Schoetz, V. Schurig, Chirality 2001, 13,
2
[
4
6
03–414; b) C. Wolf, Chem. Soc. Rev. 2005, 34, 595–
08; c) I. D’Acquarica, F. Gasparrini, M. Pierini, C. Vil-
lani, G. Zappia, J. Sep. Sci. 2006, 29, 1508–1516; d) R.
Sabia, A. Ciogli, M. Pierini, F. Gasparrini, C. Villani, J.
Chromatogr. A 2014, 1363, 144–149; e) F. Maier, O.
Trapp, Angew. Chem. 2012, 124, 3039–3043; Angew.
Chem. Int. Ed. 2012, 51, 2985–2988; f) O. Trapp, Top.
Curr. Chem. 2013, 341, 231–270.
[12] B. E. Norcross, P. E. Klinedinst, F. H. Westheimer, J.
Am. Chem. Soc. 1962, 84, 797–802.
[13] A. T. H. Lenstra, G. H. Petit, R. A. Dommisse, F. C. Al-
derweireldt, Bull. Soc. Chim. Belg. 1979, 88, 133–141.
[14] X.-Q. Zhu, H.-L. Zou, P. -W. Yuan, Y. Liu, L. Cao, J.-P.
Cheng, J. Chem. Soc. Perkin Trans. 2 2000, 1857–1861.
[15] X.-Q. Zhu, Y.-C. Liu, J.-P. Cheng, J. Org. Chem. 1999,
64, 8980–8981.
[
7] a) O. Trapp, J. Chromatogr. A 2008, 1184, 160–190;
b) O. Trapp, Chem. Today 2008, 26, 26–28; c) O. Trapp,
Electrophoresis 2010, 31, 786–813; d) O. Trapp, in:
Rapid Enantiomeric Excess Determination, (Eds.: M.
Christmann, S. Bräse), VCH, Weinheim, 2012; e) O.
Trapp, in: Microcapillary Catalysis in Comprehensive
[16] L. Simón, J. M. Goodman, J. Am. Chem. Soc. 2008,
130, 8741–8747.
3
520
asc.wiley-vch.de
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 3513 – 3520