Continuous Flow Asymmetric Transfer Hydrogenation with Long Catalyst Lifetime and Low Metal Leaching

Article abstract of DOI:10.1002/ejoc.201901547

Homogenous iridium complexes with asymmetric ligands and different tether lengths have been solid-supported and prepared in multi-gram quantities. Packed in a fixed-bed, they have been used in continuous flow for up to 120 hours in the asymmetric transfer hydrogenation of APs in 2-propanol to make 1-phenethyl alcohols in > 95 % conversion and ee. The CsDPEN ligand and C₅ tether showed higher performance than TsDPEN and the C₁₄ tether, whilst the ketone feed concentration and reaction temperature were optimized to enable the catalyst to be used at 5 mol-% loading with a residence time of 39 minutes. The total amount iridium leaching of from the support during sustained operation was 58–147 ppm. The flow system gives higher catalyst turnover numbers than the related batch reactions, but the nature and concentration of the base were found to influence strongly the catalyst's performance, with the finding that triethylamine maintains high enantioselectivity but slowly deactivates the catalyst, whilst potassium tert-butoxide does the opposite. The utility of the system is shown in the high ee's and good conversions achieved for a range of aryl alcohols.

Full text of DOI:10.1002/ejoc.201901547

A Journal of
Accepted Article
Title: Continuous Flow Asymmetric Transfer Hydrogenation with Long
Catalyst Lifetime and Low Metal Leaching
Authors: Yuji Kawakami, Antonia Borissova, Michael Chapman, Gert
Goltz, Eleonora Koltsova, Ivan Mitrichev, and Andrew John
Blacker
This manuscript has been accepted after peer review and appears as an
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 published
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.201901547
Link to VoR: http://dx.doi.org/10.1002/ejoc.201901547
Supported by

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL PAPER
Continuous Flow Asymmetric Transfer Hydrogenation with Long
Catalyst Lifetime and Low Metal Leaching
Yuji Kawakami,[a] Antonia Borissova,[b] Michael R. Chapman,[b] Gert Goltz,[b] Eleonora Koltsova,[c]
Ivan Mitrichev, [c] A. John Blacker.[b]*
Dedicated to the memory of Prof. Jonathan M. J. Williams
Abstract: Homogenous iridium complexes with asymmetric ligands
and different tether lengths have been solid-supported and prepared
in multi-gram quantities. Packed in a fixed-bed, they have been used
in continuous flow for up to 120 hours in the asymmetric transfer
hydrogenation of APs in isopropanol to make 1-phenethyl alcohols in
>95% conversion and ee. The CsDPEN ligand and C5 tether showed
higher performance than TsDPEN and the C14 tether, whilst the
ketone feed concentration and reaction temperature were optimized
to enable the catalyst to be used at 5 mol% loading with a residence
time of 39 minutes. The total amount iridium leaching of from the
support during sustained operation was 58-147 ppm. The flow system
gives higher catalyst turnover numbers than the related batch
reactions, but the nature and concentration of the base were found to
influence strongly the catalyst’s performance, with the finding that
Figure 1. Structures, turnover and recycle numbers of some immobilized
asymmetric transfer hydrogenation catalysts.[6a-d]
triethylamine maintains high enantioselectivity but slowly deactivates
the catalyst, whilst potassium tert-butoxide does the opposite. The
utility of the system is shown in the high ee’s and good conversions
achieved for a range of aryl alcohols.
bidentate ligands with catalysts used in batch, separated and re-
used, Figure 1,[6] Whilst the total turnover numbers (TTN) are
satisfactory, problems include: the fall in activity after a few
recycles due to metal leaching or catalyst poisoning; the solid
support itself causing lower activity because of reduced mass
Introduction
transfer; the increased cost of complex ligands versus the benefit
in TTN. Surprisingly, there are few reports of transfer
hydrogenation in continuous flow, where a benefit of operating in
steady-state is information about catalyst deactivation and
leaching rates. Liese et al attached a functionalized tetradentate
ruthenium catalyst to a soluble polysiloxane polymer and used
Since the discovery of the Noyori Ru(arene) catalysts in
1995,[1] and the isoelectronic Ir/Rh Cp* catalysts in 1997,[2]
catalytic asymmetric transfer hydrogenation (CATHy) has been
used widely to make chiral alcohol and amine building blocks, with
numerous scaled-up industry examples.[3] Despite this, the
turnover number (TON) and frequencies (TOF) are not as high as
those of asymmetric hydrogenation catalysts.[4] To overcome this,
loadings of 0.1-1 mol% are used, that both increase the cost and
complicate separation of the catalyst from the product.
Immobilization is a strategy that has been greatly researched, and
is used widely with enzymes, but is unsuccessful for
homogeneous catalysts[5] Most reports use functionalized
this in
a
continuous flow membrane reactor with
isopropanol/dichloromethane.[7] This was operated over 175
reaction volumes, but, due to deactivation, required a top-up of
0.5% fresh catalyst per reaction volume. In this way, they
achieved a steady-state conversion of 79% AP giving the alcohol
in 91% ee, with a productivity of 578 g/L/day, giving a catalyst TTN
of 2630. Van Leeuwen et al tethered a ruthenium norephedrine
catalyst to silica.[8] They used this in a fixed-bed continuous flow
mode over 1 week, to reduce AP with isopropanol at 95% steady-
state conversion to 1-phenylethanol with 90% ee. The leaching of
ruthenium was less than 1% during 3-11 h of catalysis and a fall
in performance was only seen after 170 h of operation.
[a]
[b]
Mr Y.Kawama Chemical Process R&D, CMC & Production HQs,
Ono Pharmaceutical Co., Ltd. 3-1-1 Sakurai, Shimamoto-Cho,
Mishima-Gun, Osaka, 618-8585, Japan.
Dr A.Borissova, Dr M.R.Chapman, Dr G.Goltz, Prof. A.J. Blacker.
Institute of Process Research and Development, School of
Chemistry and School of Chemical and Process Engineering,
University of Leeds, Woodhouse Lane, Leeds, LS2 9JT, United
Kingdom. www.chem.leeds.ac.uk/john-blacker
This study reports a new application of our published alkyl
tethered tetramethylcyclopentadienyl (Cp**) metal complex
immobilized on Wang-resin.[9][10] Here, the modular design
anchors iridium through an anionic 5 complex, and retains three
free coordination sites.
[c]
Prof. E.Koltsova, Dr I.Mitrichev. D. Mendeleev University of
Chemical Technology of Russia, IT department, Miusskaya sq., 9,
125047, Moscow, Russia
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.

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL PAPER
Results and Discussion
Table 1. Activity and selectivity of soluble and immobilized catalysts used in
batch with AP, isopropanol, KOtBu.
[a]
The (S,S)-N-tosyl-1,2-diphenylethylene diamine (TsDPEN, 2a),
and (S,S,S)-N-camphorsulfonyl-1,2-diphenylethylene diamine
(CsDPEN, 2b), ligands were selected for this work, as these have
previously given good results in homogeneous asymmetric
transfer hydrogenation reactions.[11] The catalyst’s performance
with C5 and C14 linkers (expressed as the number of carbon atoms
in the chain) were compared, as this gives an indication of mass
transfer effects in the boundary layer around the solid support.
The sequence in which the metalation, ligation and addition to the
support were carried out, was found to be important. Attempts to
coordinate the diamine ligand directly to the tethered
dichloroiridium Cp** complex immobilized on Wang resin were
unsuccessful, whilst ligation of the soluble complex, followed by
activation of the terminal hydroxyl group as a triflate, then reaction
with the Wang resin solid support, gave the desired immobilized
catalysts in reasonable yields, Scheme 1.
#
Catalyst
Mode
Ee[b]
(%)
79
95
82
95
85
96
24
82
46
67
95
Conv.
TON
Time
(h)
5
TOF
(h-1)
3.0
4.6
3.0
4.8
3.4
4.8
1.5
1.0
1.1
1.0
4.6
[b]
(%)
1
IrClCp*TsDPEN
Batch
Batch
Batch
Batch
Batch
Batch
Batch
Batch
Batch
Batch
Batch
59
90
60
96
67
97
62
95
86
96
93
15
23
15
24
17
24
16
24
22
24
23
2
IrCp*CsDPEN
5
3
3a
5
4
3b
5
5
3c
5
6
3d
5
7
4a
24
24
24
24
5
8
4b
9
4c
10
11
4d
IrCp*CsDPEN
(no base)
3b (no base)
4b (no base)
12
13
Batch
Batch
91
3
86
33
22
8
5
4.4
0.3
24
[a] Substrate/Catalyst/Base (mole ratio) = 100/4/10 (entries 1-6), 100/4/100
(entries 7-10), 100/4/0 (entries 11-13). [b] Measured by GC (ESI 11.1)
The specific surface area was measured by BET analysis at
11.2 m2/g.[12] It is calculated for 4a that each tethered catalyst
molecule at the polymer surface is 1.9Å apart, 4b 2.5Å, 4c 3Å and
4d 4.5Å. The extended length of the tether & iridium catalyst from
the surface are about 28Å 4a and 40Å 4d. From these estimates,
it is clear that each site is not isolated and tethered catalyst
molecules can interact, and might entangle.
Table 1 shows the results a comparison of the homogenous
and solid-supported catalysts in the asymmetric transfer
hydrogenation of acetophenone (AP) to 1-phenylethanol by
isopropanol (IPA). Potassium tert-butoxide (KOtBu) is used
commonly in this system to activate the catalyst from the 18-
electron chloroiridium precatalyst, to the 16-electron species.[1]
However, it was found that when the base and ketone were
combined into a single isopropanol solution, the mixture darkened
in a few hours, due to the formation of Claisen condensation
products, which over extended run times was problematic.
Accordingly, separate solutions of each were pumped and mixed
together with the catalyst (ESI 4.0). The conventional IrCp*
homogenous catalysts used in batch mode, entries 1-2, showed
similar performance to the soluble C5 and C14 Cp** analogues, 3a-
d, entries 3-6, indicating that the tether does not interfere in the
reaction. Also, the CsDPEN ligand gave significantly higher ee
and conversion than TsDPEN ligand. Comparing the solid-
supported catalysts, entries 7-10, with the homogenous, in which
1 equiv. of base was used, the conversions and TONs were
similar, but the enantioselectivities and TOFs were considerably
lower. The base was suspected to be the problem, so the
catalysts were retested without it. Surprisingly, the homogenous
catalysts IrCp*CsDPEN and 3b showed little change in
performance, entries 11-12, and investigation by 1HNMR, showed
the isolated pre-catalysts were present as the 16-electron species,
as opposed to the IrCp*TsDPEN and 3a that were observed as
the 18-electron chloroiridium species (ESI 2.0). As is frequently
observed with immobilized catalysts, 4a-d were also less active
than the homogenous ones. For example the turnover frequency
(TOF) of 3b is 4.8 h-1 compared with 1.0 h-1 with 4b, entries 4 and
Scheme 1. Reaction sequence and results for making the C5 and C14 tethered
TsDPEN and CsDPEN Cp** iridium complexes.
This highlights the difficulties in doing reactions on solid supported
materials. Both the C5 and C14 tethered chloro Cp* iridium
complexes, 1a and 1b, were prepared in multi-gram quantities by
previously described methods.[9] The ligands TsDPEN, 2a, and
CsDPEN, 2b, were reacted with 1a or 1b in dichloromethane with
triethylamine to give the soluble complexes, 3a-d in near
quantitative yield. Interestingly, 3a and 3c were obtained with the
chloride ligand, however 3b and 3d were isolated as the 16-
electron species lacking HCl, as confirmed by 1HNMR and HRMS,
despite using the same conditions (ESI 2.0). The primary alcohol
tethers were then activated with triflic anhydride and reacted with
Wang resin (1% degree of crosslinking, 100-200 mesh) to give
the solid-supported catalysts, 4a-d. After extensive washing, first
with dichloromethane, then water, 1M HCl (aq), water, methanol
and isopropanol, the solid was vacuum dried. ICP analysis of the
iridium showed immobilization reaction yields of 30% for 4a, 20%
for 4b, 14% for 4c, and only 4% for 4d. As previously observed,
the longer tether seems to cause problems with attachment. The
solid supported catalysts were analyzed by EDX TEM (ESI 10.0),
showing spherical particles of average 120 m diameter with an
even surface distribution of the catalyst’s constituent elements.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL PAPER
8. This time however, when the base was omitted from the
reaction using catalyst 4b it showed poor activity, TOF=0.3, and
very low ee, entry 13.
From these initial studies, it was clear that the base is
important in delivering both high conversion and ee over a
sustained time, however no studies were found to help us develop
the IPA CATHy system. Accordingly, we evaluated different
bases in batch mode using catalyst 4b, with the results shown in
Figure 2.
To understand further the effects of base, experiments were
carried out in continuous flow. 0.5 g of each solid-supported
catalyst, were blended together with 2.5 g washed sand and
packed into 170 mm length, 4.5 mm diameter, empty HPLC
columns, and placed into a temperature controlled electrical
heating block. The column void space was determined to be 1.4
mL (ESI 8.0), and a combined flow rate of 0.036 mL/min substrate
and base in IPA, gave a residence time (Tres) of 39 minutes. After
reaching steady state, each sample was collected over 2 minutes
and analysed for conversion of AP to 1-phenylethanol, along with
its ee. Figure 3 shows the effect of KOtBu on catalyst performance
over 48 hours.
Figure 2. The effect of 100 mol% of different bases on the conversion and
enantioselectivity in the IPA reduction of AP with 3.7 mol% solid-supported
catalyst 4b used in batch-mode.
Figure 3. Experiments comparing the effect of base feed regime on conversion
and ee. Expt.1 started at 5 mol% KOtBu and reduced after 3 h to 0.5 mol%.
Expt.2 started at 5 mol% KOtBu and reduced at 0.9 h to 0.5 mol%. Dotted lines
show the overall trend.
It was found that the conversions and ee’s correlate with the
base pKa. The traditional base used is KOtBu (pKa = 16-19),[13a]
,
whilst 2,6-lutidine (6.75),[13b] and N-methylmorpholine (NMM)
(7.4),[13c] were found to be too weak to fully activate the catalyst
by dehydrochlorination to generate the 16-electron species. On
the other hand, amine bases with a moderate pKa performed well,
giving both high ee and conversion: DBU (11.9),[13d] Et3N
(10.7),[13a] Hünig’s base (10.7),[13e] and TMEDA (10.4).[13f]
Triethylamine is the cheapest amine base, so its reaction
stoichiometry was investigated further, Table 2.
In experiment 1, it was found that a base feed concentration of 50
mM was too high (5 mol% of substrate, 100 mol% of catalyst), and
the product ee fell rapidly from 82 to 47% (0.2%ee.min-1) whilst
the conversion remained constant at ~83%. When the base
concentration was reduced at 3 h to 5 mM, the ee remained
steady for a further 26 hours (40 reactor volumes (RV)). The fall
in ee might be loss of ligand or its racemization, both of which
have been observed in related studies.[14] In experiment 2, the
base was used at 50mM for 0.9 h, then at a tenth of this thereafter.
In this way, the selectivity was more consistent at ~85% ee for 48
hours however the conversion fell slowly from 77 to 60%.
Purification of the feed materials had no effect on the loss of
conversion, showing that residual impurities were unlikely to be
the problem. The total amount of iridium collected in the eluent
over 48 h continuous flow was measured by ICP-MS at 147 ppm
(ESI 1), indicating that loss of activity is not caused by metal
leaching.[9] Similarly, the leaching of ligand was investigated,
however non was found in the eluent and including 0.5 mol% of
CsDPEN into the feed only slightly improved the TTN.
Table 2. The influence of Et3N equivalents on AP reduction in batch using
catalyst 4b.
#
Equivalents
Conv.(%)(Ee %)
of Et3N[a]
5 h
24 h
1[b]
2[b]
3[c]
4[d]
5[c]
0
9 (2)
33 (3)
0.1
31 (93)
85 (95)
87 (95)
94 (95)
94 (96)
95 (92)
93 (94)
98 (94)
1.0
3.0
10.0
[a] Mole ratio based on AP with catalyst 4b in IPA. [b] 3.7 mol% catalyst at 60oC.
[c] 2.3 mol% catalyst at 30oC. [d] 2.3 mol% catalyst at 60oC.
The continuous flow reaction conditions were optimized
using catalyst 4a. It was found that both the reaction temperature
and substrate concentration influenced the conversion and
enantioselectivity, Figures 4a and 4b.
Without base the catalyst performed poorly, entry 1. Using 10
mol% with respect to AP, a slow reaction occurred with 31%
conversion after 5h, entry 2. One equivalent gave a faster rate,
whilst 10 equivalents gave near full conversion in 5 h with a high
ee, entry 4. Based on this, triethylamine works well at ≥1 equiv. to
substrate, whilst previous results show that KOtBu works best
sub-stoichiometrically.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL PAPER
In the former, the ee varied between 82 and 59%, but the
conversion fell only slightly from 92 to 80 over the 5 days
operation. Using triethylamine, the ee remained consistently
around 92% over the entire experiment, however the conversion
fell increasingly from 95 to 25%. ICP-MS showed only 58 ppm
iridium in the eluent. The TTN is 294 in the KOtBu system and 388
in the triethylamine system, both of which exceed the TON of 24
achieved with the homogenous catalyst in batch. The results also
indicate that that the system could be operated for longer to give
increased catalyst use, though there is clearly some fall-off in
performance discussed below.
Figure 4. AP conversion and 1-phenethyl alcohol ee at steady-state continuous
flow. a) The effect of temperature at 0.3 M AP, b) The effect of AP concentration
at 60oC.
The length of the tether was shown to affect the catalyst
lifetime and TOF. Figure 6 shows a comparison between the C5
and C14 tethered catalysts.
o
A temperature of up to 80 C enhanced the reaction rate, giving
up to 100% conversion during the 39 minutes residence time,
however this was at the expense of enantioselectivity falling from
58% ee to 36% ee, Figure 4a. The intermediate temperature of
60 oC was selected for future studies. To improve productivity, the
effect of increasing the AP feed concentration was evaluated: 0.3
M AP gave 79% conv. and 51% ee, whilst 3 M AP gave 67% and
30% respectively, Figure 4b. By calculation, at a residence time
of one hour the 3 M concentration would give 100% conv. giving
a productivity of 1.65 moles/L/h, however, the lower ee favored
the use of 0.3 M AP. The lower ee at higher concentrations, is due
to the acetone by-product that accelerates the reverse
dehydrogenation reaction (ESI Figure S3).
Using the optimized conditions, experiments were
performed over 120 hours (185 RV) continuous flow using the C5
tethered catalyst 4b and either 5 mol% KOtBu, Figure 5a, or 300
mol% Et3N, Figure 5b.
Figure 6. Comparison of catalysts 4b (C5) and 4d (C14) with different tether
length. S/C(4b)/B(Et3N)=100/33/300, S/C(4d)/B(Et3N) = 100/13/300. The dotted
lines show the overall trends.
At normalized iridium concentrations, the C14 catalyst 4d lost
activity more quickly in 24 hours than the C5 catalyst 4b. However
the maximum TOF of 4d was 8.9 h-1, twice that of the
homogenous catalysts or 4b (4.4 h-1), which implies a higher
concentration in the catalytic cycle, but a higher deactivation rate.
To investigate the scope of the continuous flow asymmetric
transfer hydrogenation, a series of different substrates were
tested, Figure 7. After reaching steady state, each sample was
analysed by GC for conversion and ee. AP analogues with ortho-,
meta- and para- electron-withdrawing substituents (-Br, -Cl, -F, -
NO2) gave good conversion (94-100%) and enantioselectivity (78-
94% ee). Electron donating groups, -pOMe, -pMe, gave 59% and
87% conversion and 93 and 98% ee, respectively. Surprisingly,
both 1-indanol and pinacol were obtained in only 12% (93% ee)
and 3% conversion respectively, which is surprising given
previous good results with these substrates with related
homogenous catalysts.[15]
Conclusions
This study has shown that a continuous flow asymmetric transfer
hydrogenation process, using IPA as the hydrogen donor and
solvent, can been catalyzed efficiently by an immobilized chiral
iridium Cp** complex of modular design. The process has
Figure 5. The conversion, ee and total turnover number (TTN) in continuous
flow asymmetric reduction of AP (0.3 M) at 60oC with catalyst 4b. a) 5mM KOtBu.
b) 300 mol% Et3N. The dotted lines show the overall trends.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL PAPER
systems, with the additional problem of CO2 (g) evolution.[15] The
immobilized catalysts are also being used in amine racemization
and alkylation reactions which will be reported elsewhere.
Experimental Section
Typical procedure of ATH in batch mode:
Homogenous catalyst.
Under
a
nitrogen atmosphere, the
Ir(Cp*C5H10OH){(S,S)-TsDPEN}Cl 3a (16.0 mg, 0.02 mmol, 0.04 equiv.)
and KOtBu (5.6 mg, 0.05 mmol, 0.10 equiv.) were added to 2-propanol (2.0
ml). AP (58 l, 0.50 mmol, 1.0 equiv.) was added, followed by 2-propanol
(0.50 ml). After stirring for 5 hours at 60oC, the reaction mixture was
sampled and analysed by GC for conversion and enantioselectivity.
Immobilized catalyst. Under a nitrogen atmosphere, the immobilized C5
tethered/CsDPEN catalyst 4b (41.7 mg, 9.2 mol, 0.22 mmol/g, 0.037
equiv.) and Et3N (35 l, 0.25 mmol, 1.0 equiv.) were added to 2-propanol
(1.0 ml). AP (29 l, 0.25 mmol, 1.0 equiv.) was added, followed by 2-
propanol (0.25 ml). After stirring for 24 hours at 60oC, the reaction mixture
was sampled and analysed by GC for conversion and enantioselectivity.
Typical procedure of ATH in flow mode:
Two pump system. The immobilized C5 tethered/TsDPEN catalyst 4a (500
mg, 0.35 mmol/g, 0.18 mmol) and sand (low iron, 2.50 g) were mixed by
shaking and placed in the steel column (length: 170 mm, inner diameter:
4.5 mm, volume: 2.7 ml) with filter. Two HPLC pumps (JASCO, PU-980)
were connected to the T-shape mixer and packed column with heating
block. After being filled with isopropanol (each flow rate: 0.2 ml/min, 30
Figure 7. Other ketones reduced by CATHy in continuous flow, to give chiral
alcohols, showing steady-state conversions (%) (ee(%)).
o
min), the heating block was maintained at 50 C. Feed A: AP (3.60 g, 30
mmol) was dissolved and make up to 100 ml by isopropanol at room
temperature. Feed B: KOtBu (168 mg, 1.5 mmol) was dissolved and make
up to 100 ml by isopropanol at room temperature. Each solution was fed
into the packed column at each flow rate of 18 l/min whilst maintaining
the reaction temperature at 50oC. For reaction monitoring, the reactor
eluent was collected for 6 minutes each time and analysed by GC.
Single pump system. The immobilized C5 tethered/CsDPEN catalyst 4b
(500 mg, 0.14 mmol/g, 0.07 mmol) and Sand (low iron, 2.50 g) were mixed
by shaking and placed in the steel column (Length: 170 mm, Inner
diameter: 4.5 mm, Volume: 2.7 ml) with filter. The HPLC pump (JASCO,
PU-980) was connected to the packed column with heating block. After
being filled with isopropanol (0.3 ml/min, 30 min), the heating block was
maintained at 60oC. AP (3.60 g, 30 mmol), Et3N (9.11 g, 90 mmol) and
(S,S,S)-CsDPEN (621 mg, 0.15 mmol) were dissolved at 30-40oC in
isopropanol and make up to 200 ml by isopropanol at room temperature.
The prepared mixture was fed into the packed column at the flow rate of
36 l/min whilst maintaining the reaction temperature at 60oC. The reactor
eluent was continuously collected in the flask at room temperature. Once
the feed solution was nearly empty, the feed solution was prepared anew
and was replaced with previous one. The reactor eluent was collected for
3 minutes each time and analysed by GC.
been operated over 5 days with low levels of iridium (58-147 ppm)
leaching over this period. Two asymmetric ligands and two tether
lengths have been evaluated, with CsDPEN and the C5 tether
giving an average 94% ee and TTN of 388 over 120 hours, which
is significantly better than the homogeneous catalyst TON of 24
used in batch. The type and concentration of base used in the
reaction has been identified as an important condition. The
correlation of activity with base pKa and strength is likely related
to the concentration of the catalytically active 16-electron species.
An initial stoichiometric quantity of KOtBu is sufficient to activate
the catalyst, and a sub-stoichiometric feed neutralizes any
residual acidity that would allow it to revert to the inactive 18-
electron species, in this way high activities and ee’s are
maintained over days. On the other hand, an excess KOtBu is
able to remove the ligand but retain a catalytically active species,
causing the ee to fall not the conversion. The same solid-
supported catalyst, lacking an asymmetric ligand was observed
by x-ray synchrotron techniques to react with tert-butoxide.[14a]
Triethylamine is unable to displace the ligand so the optical
purities remain consistently high over a long time, however the
conversion falls increasingly. This unusual behavior may be
explained by physical effects such as entanglement of the tethers.
Based on the loadings used, an estimate of site-isolation showed
the catalysts are close enough to entangle, and it might be
anticipated that the longer the tether the more readily this might
happen. Indeed, the C14 is initially more active than the C5 but
loses activity more rapidly. Clearly more research is required to
understand the cause of these observations, so it would be
interesting to prepare supported catalysts with lower loadings to
increase site-isolation and to compare their turnover. The use of
formate hydrogen donor is being tested in continuous flow
Acknowledgements
This work was granted by the Newton Institutional Links UK-
Russia project (277335368) funded by the British Council. We
would like to thank Ono Pharmaceutical, Japan for the
secondment of YK to the iPRD.
Keywords: homogenous supported catalyst • immobilized
iridium catalyst • continuous flow • asymmetric transfer
hydrogenation • phenylethanol
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL PAPER
[1]
a) D. Wang, D. Astruc, Chem. Rev. 2015, 115, 6621-6686; b) S.
Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
1995, 117, 7562–7563; c) R. Noyori, S. Hashiguchi, Asymmetric Transfer
Hydrogenation Catalyzed by Chiral Ruthenium Complexes. Acc. Chem.
Res. 1997, 30, 97–102.
Pericas, P. R. Hanson, O. Raiser, ACS Sustainable Chem. Eng. 2016, 4,
2698-2705; l) Y. Arakawa, A. Chiba, N. Haraguchi, S. Itsuno, Adv. Synth.
Catal. 2008, 350, 2295-2304. m) P. N. Liu, P. M. Gu, F. Wang, Y. Q. Tu,
Org. Lett. 2004, 6, 169-172; n) S. Itsuno, S. Takahashi, Chem. Cat.
Chem. 2017, 9, 385-388.
[2]
[3]
a) J. Mao, D. C. Baker, Org. Lett. 1999, 1, 841-843; A. J. Blacker, B. M.
Mellor WO9842643. Zeneca Ltd.
[7]
[8]
[9]
S. Laue, L. Greiner, J. Woltinger, A. Liese, Adv. Synth. Catal. 2001, 343,
711-720.
T. Ikariya, A. J. Blacker, Acc. Chem. Res. 2007, 40, 1300–1308; b) T.
Touge, T. Hakamata, H. Nara, T. Kobayashi, N. Sayo, T. Saito, Y. Kayaki,
T. Ikariya, J. Am. Chem. Soc. 2011, 133, 14960–14963; c) F. K. Cheung,
A. J. Clarke, G. J. Clarkson, D. J. Fox, M. A. Graham, C. Lin, A. L. Criville,
M. Wills, Dalton Trans. 2010, 39, 1395-1402; d) A. M. Hayes, D. J. Morris,
G. J. Clarkson, M. A. Wills, J. Am. Chem. Soc. 2005, 127, 7318; e) N. A.
Cortez, R. Rodriguez-Apodaca, G. Aguirre, M. Parra-Hake, T. Cole, R.
Somanathan, Tetrahedron Lett. 2006, 47, 8515–8518; f) M. Renom-
Carrasco, P. Gajewski, L. Pignataro, J. G. DeVries, U. Piarulli, C.
Gennari, L. Lefort, Adv. Synth. Catal. 2016, 358, 515–519.
A. J. Sandee, D. G. I. Petra, J. N. H.; Reek, P. C. J. Kamer, P. W. N. M.
van Leeuwen, Chem. Eur. J. 2001, 7, 1202-1208.
S. J. Lucas, B. D. Crossley, A. J. Pettman, A. D. Vassileiou, T. E. O.
Screen, A. J. Blacker, P. C. McGowan, Chem. Commun. 2013, 49,
5562−5564.
[10] S. S. Wang, J. Am. Chem. Soc. 1973, 95, 1328-1333.
[11] a) X. Li, A. J. Blacker, I. Houson, X. Wu, J. Xiao, Synlett 2006, 8, 1155-
1160; b) A. J. Blacker, M. J. Stirling, Avecia Pharmaceuticals Ltd.
WO2006046062, 2006.
[12] B Altava, M. I. Burguete, E. Garcı´a-Verdugo. S. V. Luis, Chem. Soc.
Rev., 2018, 47, 2722-2771; A. Deratani, G. D. Darling, D. Horak and J.
M. J. Frechet, Macromolecules, 1987, 20, 767–772; L.-J. Zhao, C. K.-W.
Kwong, M. Shi, P. H. Toy, Tetrahedron, 2005, 61, 12026–12032; F.
Guendouz, R. Jacquier and J. Verducci, Tetrahedron, 1988, 44, 7095–
7108
[4]
a) P. Barbaro, F. Liguori, Chem. Rev. 2009, 109, 515–529; b) A. Molnar,
A. Papp, Coord. Chem. Rev. 2017, 349, 1-65; c) C. M. Monteiro, A. F.
Trindade, P. M. P. Gois, C. A. M Afonso in Catalytic Methods in
Asymmetric Synthesis, (Eds. M. Gruttadauria and F. Giacalone), John
Wiley & Sons, Inc., 2011, pp. 3–82; d) C. V. Doorslaer, J. Wahlen, P.
Mertens, K. Binnemans, D. D. Vos, Dalton Trans. 2010, 39, 8377–8390.
a) S. Hubner, J. G. de Vries, V. Farina, Adv. Synth. Catal. 2016, 358, 3–
25; b) B. Pugin, H.-U. Blaser in Heterogenized Homogeneous Catalysts
for Fine Chemicals Production, (Eds.: P. Barbaro, F. Liguori), Springer,
Netherlands, 2010, 33, pp. 231–245.
[13] a) W. Reeve, C. M. Erikson, P. F. Aluotto, Can. J. Chem. 1979, 57, 2747-
2754; b) J. T. Muckerman, J. H. Skone, M. Ning, Y. Wasada-Tsutsui,
Biochimica et Biophysica Acta 2013, 882-891; c) H. K. Hall Jr., J. Am.
Chem. Soc. 1957, 79, 5441–5444; d) R. Srivastava, J. Mol. Catal. A:
Chem. 2007, 264, 146–152; e) K. L. Sorgi, Encyclopedia of Reagents for
Organic Synthesis, 2001; f) T. B. Nguyen, P. Retailleau, Org. Lett. 2017,
19, 4858−4860.
[5]
[6]
a) K. Masuda, T. Ichitsuka, N. Koumura, K. Sato, S. Kobayashi,
Tetrahedron 2018, 74, 1705-1730; b) B. R. Vaddula, R. S. Varma, M. A.
A. Gonzalez, Curr. Org. Chem. 2013, 17, 2268–2278; c) J. Dimroth, J.
Keilitz, U. Schedler, R. Schomaecker, R. Haag, Adv. Synth. Catal. 2010,
352, 2497-2506; d) Y. Shen, Q. Chen, L. Lou, K. Yu, F. Ding, S. Liu,
Catalysis Letters 2010, 137, 104-109; e) S. Parambadath, A. P. Singh,
Catalysis Today 2009, 141, 161-167; f) A. Hu, G. T. Yee, G. T. W. Lin, J.
Am. Chem. Soc. 2005, 127, 12486-12487; g) X. Huang, J. Y. Ying, Chem.
Commun. 2007, 18, 1825-1827 h) H. Siklova, E. Leitmannova, P. Kacer,
L. Cerveny, Reaction Kinetics and Catalysis Letters 2007, 92, 129-136;
i) J. Li, Y. Zhang, Y. D. Han, Q. Gao, C. Li, J. Mol. Catal. A: Chem. 2009,
298, 31-35. j) K. Zhang, J. An, Y. Su, J. Zhang, Z. Wang, T. Cheng, G.
Liu, ACS Catal. 2016, 6, 6229-6235. k) C. M. Eichenseer, B. Kastl, M. A.
[14] a) G. J. Sherborne, M. R. Chapman, A. J. Blacker, R. A. Bourne, T. W.
Chamberlain, B. D. Crossley, S. J. Lucas, P. C. McGowan, M. A. Newton,
T. E. O. Screen, P. Thompson, C. E. Willans, B. N. Nguyen, J. Am. Chem.
Soc., 2015, 137, 4151-4157; b) A. J. Blacker, S. B. Duckett, J. Grace, R.
N. Perutz, A. C. Whitwood, Organometallics 2009, 28, 1435-1446.
[15] a) S. H. Deshpande, V. S. Shende, S. K. Shingote, D. Chakravarty, V. G.
Puranik, R. V. Chaudhari, A. A. Kelkar, RSC Advances 2015, 5 (64),
51722-51729; b) F. K. Cheung, A. M. Hayes, D. J. Morris, M. Wills,
Organic & Biomolecular Chemistry 2007, 5 (7), 1093-1103.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201901547
European Journal of Organic Chemistry
FULL
PAPER
Continuous Flow Asymmetric
Transfer Hydrogenation
Yuji Kawakami, Antonia Borissova,
Michael R. Chapman, Gert Goltz,
Eleonora Koltsova, Ivan Mitrichev, A.
John Blacker.*
Page No. – Page No.
Continuous Flow Asymmetric
Transfer Hydrogenation with Long
Catalyst Lifetime and Low Metal
Leaching
Continuous flow catalytic asymmetric transfer hydrogenation of ketones
is reported using iridium complexes attached to Wang resin. The system
gives higher TONs than homogenous batch, but lower TOFs. The base
was found to affect the conversion and ee, and only low levels of iridium
were found to leach. A range of aryl ketones were reduced in good ee
and conversion.
This article is protected by copyright. All rights reserved.