Deactivation mechanisms of iodo-iridium catalysts in chiral amine racemization

Article abstract of DOI:10.1016/j.tet.2020.131823

The homogenous, [IrCp?I₂]₂, SCRAM catalyst (1) is active in the racemization of chiral amines. NMR, kinetic and structural mechanistic studies have determined the cause of catalyst deactivation to occur when ammonia or methylamine are liberated by hydrolysis or aminolysis of the intermediate imine, which tightly coordinate to the iridium centre to block turnover. Control of moisture and substrate concentration can suppress deactivation, whilst partial reactivation of spent catalyst was identified using hydroiodic acid.

Full text of DOI:10.1016/j.tet.2020.131823

Journal Pre-proof
Deactivation Mechanisms of Iodo-Iridium Catalysts in Chiral Amine Racemization
Maria H.T. Kwan, Nisha P.B. Pokar, Catherine Good, Martin F. Jones, Rachel
Munday, Thomas Screen, A. John Blacker
PII:
S0040-4020(20)31071-1
https://doi.org/10.1016/j.tet.2020.131823
TET 131823
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 27 October 2020
Revised Date: 23 November 2020
Accepted Date: 26 November 2020
Please cite this article as: Kwan MHT, Pokar NPB, Good C, Jones MF, Munday R, Screen T, Blacker
AJ, Deactivation Mechanisms of Iodo-Iridium Catalysts in Chiral Amine Racemization, Tetrahedron,
https://doi.org/10.1016/j.tet.2020.131823.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Elsevier Ltd. All rights reserved.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Deactivation Mechanisms of Iodo-Iridium
Catalysts in Chiral Amine Racemization
Leave this area blank for abstract info.
Maria H. T. Kwan,a Nisha P. B. Pokar,a Catherine Good,b Martin F. Jones,b Rachel Munday,b Thomas Screenc
and A. John Blacker a *
a School of Chemistry and School of Chemical and Process Engineering. Institute of Process Research and
Development. University of Leeds, Woodhouse Lane, Leeds, LS2 9JT, UK; b Chemical Development,
Pharmaceutical Technology & Development, Operations, AstraZeneca, Macclesfield, SK10 2NA, UK; c Apex
Molecular, Alderley Park, SK10 4TG, UK

1
Tetrahedron
journal homepage: www.elsevier.com
Deactivation Mechanisms of Iodo-Iridium Catalysts in Chiral Amine Racemization
Maria H. T. Kwan,a Nisha P. B. Pokar,a Catherine Good,b Martin F. Jones,b Rachel Munday,b Thomas
Screenc and A. John Blacker a*
a School of Chemistry and School of Chemical and Process Engineering. Institute of Process Research and Development. University of Leeds, Woodhouse Lane,
Leeds, LS2 9JT, UK
b Chemical Development, Pharmaceutical Technology & Development, Operations, AstraZeneca, Macclesfield, SK10 2NA, UK
c Apex Molecular Ltd, Alderley Park, Macclesfield, SK10 4TG, UK
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
The homogenous, [IrCp*I2]2, SCRAM catalyst (1) is active in the racemization of chiral amines.
NMR, kinetic and structural mechanistic studies have determined the cause of catalyst
deactivation to occur when ammonia or methylamine are liberated by hydrolysis or aminolysis
of the intermediate imine, which tightly coordinate to the iridium centre to block turnover.
Control of moisture and substrate concentration can suppress deactivation, whilst partial
reactivation of spent catalyst was identified using hydroiodic acid.
Available online
Keywords: Chiral amine racemization; Catalytic
transfer hydrogenation dehydrogenation; Catalyst
deactivation
borrowing
mechanism;
Iridium;
Hydrogen
2009 Elsevier Ltd. All rights reserved
.
Dedicated to the memory of our colleague,
collaborator and friend Professor Jonathan Williams
1. Introduction
Previous work on the [IrCp*I2]2 catalyst, reports the kinetics
and mechanism of chiral amine racemization and shows amine
coordination to the iridium followed by hydride transfer.22 With
N-substituted amines it was shown that the racemization occurs
by release of the imine/iminium into the bulk solvent prior to
non-selective reduction by a hydridoiridium species. A similar
system has been reported by Yamaguchi et al, in which
IrCp*(NH3)3I2 catalyst carries out alkylation of aqueous ammonia
with a range of alcohols.24, 25
In the pharma industry racemic amines are commonly
resolved by diastereomeric crystallization as the process is
generally simple, economic and provides a consistent, high
quality product.1 Examples include dextromethorphan,2
cetirizine,3 cinacalcet4 and vestipitant.5 A disadvantage is the
large amount of enantiomeric waste produced that exceeds half
the input. In this regard racemization is being studied as a means
of recycling the unwanted material.6-9 Amines that possess an
acidic chiral proton, produced for example by Schiff bases or
alpha-carbonyl groups, can be racemized by mild bases, and there
are many examples where this is combined with in-situ resolution
for dynamic resolution processes.10-12 Unfortunately, many chiral
amines lack labile protons, with examples that include N-
heterocycles and N-(di)alkylamines. Pd,13-14 Ru,15-16 and Ir,17-18
based catalysts have been shown to reversibly dehydrogenate
these types of amines. The [IrCp*I2]2 catalyst is active in the
racemization of N-substituted amines and has been used in both
enzyme dynamic kinetic resolution,19-20 and dynamic
thermodynamic resolution processes, which is a crystallization
induced deracemization.21 The latter study reports an
immobilized [IrCp*I2]2 catalyst used in a combined continuous
flow racemization and diastereomeric crystallization of amine
salts. The immobilized catalyst lifetime was up to 242 hours, but
with N-methylated amines, a linear deactivation of 2.8% ee.h-1
was seen. This paper reports a detailed investigation of the cause
of the catalyst deactivation, Figure 1.
Figure 1. Integrated amine resolution-racemization process
and study of catalyst deactivation reported in this paper.
2. Results/discussion
The racemization of different concentrations of (S)-N-methyl-
2-phenethylamine (S)-(2) by catalyst (1) in toluene at 105oC

2
Tetrahedron
shows unusual differences in both the initial rates and
the loss of catalyst activity. Adding acetophenone midway
racemization endpoint, Figure 2.
through, had no effect on the racemization of (S)-(2), whilst,
adding 20 equiv. methylamine (with respect to iridium), 30
minutes after the start of reaction, stopped the racemization
completely, Figure 3(b). The addition of 2 equiv. of methylamine
initially retarded the catalyst, but at 105oC it may be lost from the
system as a vapor, and the catalyst regained a little activity.
1
A HNMR experiment was done to investigate methylamine
complexation to the catalyst. Figure 4 shows the change in the
integrals of selected protons (chemical shifts A-F) in a d6-DMSO
solution of catalyst (1) upon titration of methylamine/THF
solution.
Figure 2. Racemization of different concentrations of amine
(S)-(2) by catalyst (1) 0.2 mol% starting from ~100% e.e.
The fastest initial rates were observed at 0.25, 0.5 and 1.0 M
concentrations, whereas at 0.1 and 2.0 M the rates were
significantly lower. Furthermore, none of the experiments gave a
racemate, depending upon concentration stopping between 23%
and 4% e.e after 480 minutes. First order plots were fitted to the
initial rates according to Equation 1, where [S]0 is the initial
concentration of enantiomer (S)-(2) and [R]t is the concentration
of enantiomer (R)-(2) at time t.
1
Figure 4. Changes in HNMR (d6-DMSO) chemical shift of
the (CH3)5Cp and CH3NH2 protons when methylamine/THF
solution is titrated into the catalyst dimer (1).
Equation 1
Four different catalyst species were identified, based on the
Cp* methyl and methylamine protons and were assigned as the
iridium dimer (1), mono- (1a), di-(1b) and tri-(1c) aminomethyl
iridium species, Scheme 1, (ESI 5).
Linear fits for 0.1 and 2.0 M concentrations were poor, with
R2 values of 0.88 and 0.64, however, good fits were obtained
with 0.25, 0.5 and 1 M concentrations and kobs of 0.092, 0.102
and 0.114 min-1 respectively.
To understand the reason for these observations, a spiking
experiment was carried out in which an additional 0.5 equiv. of
(S)-(2) was added at 24 h making the solution 55% e.e. The graph
shows the catalyst is inactive before 24 h since no racemization
of the extra amine occurs, Figure 3(a).
Scheme 1. Putative complexes formed by addition of
methylamine to (1)
1HNMR (CDCl3) signals for the IrCp*Cl2(H2NMe) complex,
analogous to (1a), have been reported at δ=1.70 (s, 15H, Me,
Cp*), 2.77 (t, 3H, Me, 3JHH 6.7 Hz) and for
[IrCp*Cl(H2NMe)2]Cl (d6-DMSO): 1.62 (s, 15H, Me, Cp*), 2.46
3
(t, 6H, Me, JHH 6.0 Hz).26 The relative concentration of each
species was determined as a function of the methylamine added,
with dimer (1) transformed into IrCp* species (1a-c). It appears
that these species are in dynamic equilibrium and the data was
fitted to the Scheme 1 model. The observed equilibrium
constants, Equations 2-4, were calculated at 3 equivalents of
methylamine, where all four species were present.
Figure 3. Racemization of (S)-(2) 0.5 M by (1) 0.2 mol% in
toluene at 105oC (a) Spiked addition of 0.5 equiv. (S)-(2) at 24 h
resulting in an increase in e.e. but little racemization (b) spiking
of 20 equiv. acetophenone or 20 or 2 equiv. methylamine (with
respect to (1)) at 30 minutes.
Equation 2
Interference of oxygen was eliminated as a cause of catalyst
deactivation, but a reaction mass balance indicated that around
5% of the amine was converted to acetophenone. This allowed us
to postulate hydrolysis of the imine intermediate as the cause of
Equation 3
Equation 4

3
A scoping study was done to determine whether catalyst
inactivation occurred with other N-alkylated substrates, Table 1.
Notable, is the wide variation in racemization rate between the
acyclic amines (2)-(6), entries 1-5 and cyclic amines (7) and (8),
entries 6&7. The N-methylamino chiral center in sertraline (S,S)-
(4), entry 3, was epimerized rapidly from cis to trans but stopped
at around 18% d.e. This observation was previously assumed to
be the thermodynamic equilibrium between the diastereomers.17
However, a substrate spiking experiment, Figure S7 shows a
largely inactive catalyst with adventitious hydrolysis of the imine
producing an amount methylamine that equates to 8.8 equiv. vs
(1), entry 3. The N,N-dimethyl analogue (S)-(5), also failed to
racemize completely, stopping at 16% e.e., entry 4, with 1 mol%
acetophenone (5 equiv. vs (1)) observed in the final samples.
This indicates N,N-dimethylamine can also coordinate and
inactivate (1). Contrary to this, the N-iso-propylamine (R)-(6)
was completely racemized under these conditions, albeit more
slowly than (S)-(2), entry 5. Furthermore, acetophenone was seen
(1.26 mol%, 6.3 equiv. vs (1)), indicating that iso-propylamine is
ineffective at inhibiting the catalyst. Racemization of the
heterocyclic amines (S)-(7) and (S)-(8) is fast. Indeed, the latter
was too rapid to determine accurately, even at 60oC rather than
105oC. In these cases, hydrolysis of the intermediate imine is
unfavorable and reduction to produce the racemic amine is more
likely. Their steric size may prevent multiple coordination and, as
previously reported, the secondary imine dissociates readily
enabling high catalyst turnover.22
Their integrations were converted to concentrations based on the
internal standard, benzene, see ESI5.
Under this set of conditions, the equilibrium constants lie
mainly towards the disubstituted methylamine-iridium complex
(1b); while the formation of tri-substituted methylamine-iridium
complex (1c) is less favored due to steric hindrance. Although
(1a) is the least sterically demanded, the dissociation of
methylamine is unfavorable resulting in a relatively small
equilibrium constant. Unfortunately, further characterization of
species (1a-c) was unsuccessful, either by isolation of complexes
from the mixture, or mass spectrometric data. Attempts to
prepare (1c) directly using an excess of methylamine under
pressure, were similarly unsuccessful. It is unclear which of the
aminomethyl species are catalytically inactive but based on the
level of methylamine in the racemizing solution and the
estimated equilibrium constants, species (1b) and (1c) may be
inactive whilst (1a) may retain some activity. Heating (1a-c) to
displace methylamine resulted in only minor reactivation of the
catalyst, which was also observed in Figure 3(b). Competition
experiments, done by adding iso-propylamine, di-iso-
propylamine or tert-butylamine into a racemization of (S)-(2) by
catalyst (1), failed to prevent catalyst deactivation, as evidenced
by the reaction stopping at ~6% e.e., further indicating the
strength of the methylamine complexes (1a-c), ESI 6. The
evidence indicates that bulky amines and secondary imines bind
reversibly but methylamine and ammonia bind strongly and can
multiply coordinate to deactivate the catalyst. The reason these
may be potent inhibitors is because the catalyst is unable to
dehydrogenate them. Methanol is similarly a poor hydrogen
transfer donor with this catalyst, with formaldehyde never
observed.
Having determined the cause of catalyst deactivation as
hydrolysis of the intermediate imine, the obvious remedy is to
exclude water.
A
control reaction, with N-methyl-2-
phenethylimine in wet toluene but without catalyst, showed little
hydrolysis, however the addition of catalyst (1) resulted in rapid
formation of acetophenone. Similarly, water added to an actively
racemizing solution of (S)-(2) was shown to affect the rate and
end point. These results indicate the imine is coordinated to the
iridium sufficiently long to allow reactions with nucleophiles.
Real time analysis of the catalytic racemization of (S)-(2) by (1)
The analogous IrCp*I2NH3 (1d), and IrCp*(NH3)3I2
complexes have been isolated and the crystal structures
24
reported.17, The low TOF and incomplete racemization with
primary amine (S)-2-phenethylamine (S)-(3), indicates a similar
deactivation process with ammonia, generated as a result of
substrate dimerization. A solid isolated from the reaction was
identified by single crystal x-ray crystallography as (1d) and
heating this failed to rejuvenate its racemization activity.
1
using high field continuous flow HNMR and +ESI-MS were
inconclusive but showed accumulation of complex (9), M+ = 717,
and a smaller, steady-state, concentration of (10), M+ = 589.
Using a selective excitation pulse sequences technique,23 hydrido
signals at -13.1 and -15.2 ppm with M+ of 783 and 908 and
isotope patterns that indicate bridging species (Cp*Ir)2IH2 and
(Cp*Ir)2I2H that may be off cycle. Complex (1e) was not seen but
may rapidly reduce the imine to generate racemic (2) and, with
re-addition of HI, catalyst species (1a). Unlike the racemization
of (S)-(3), dimer (12) and reduced diastereomers are never seen
with (S)-(2). A species with MH+ of 725.18 (11a) is observed and
this invokes the precursor (10), although (11b) has the same mass
and isotope pattern and cannot be excluded. A plausible
mechanism for the hydrolysis is shown in Scheme 2.
Table 1. Racemization of amines with homogeneous catalyst
(1) 0.2 mol%
N
Entry amine[a]
kobs
(min-1)
0.047
0.003
0.044
0.005
0.022
0.18
ketone[b] limit e.e.
Cp*
Ph
Me
Ph
I
(mol%)
0.72
N/D
(%)
15
10[d]
18
16
0
NH
Ir
(1a)
Me
+
Cp*
N
Cp*
Ir
H
N
H
+ HI
1
2[c]
3[e]
4
(S)-(2)
(S)-(3)
(S,S)-(4)
(S)-(5)
(R)-(6)
(S)-(7)
(S)-(8)
H2
Ir
Me
H
H
R
(1e)
H
I
O
N
I
- HI
I
(2)
H
Ph
1.77
1.05
1.26
N/D
Ph
Ph
Cp*
N
OH
H2O
Ir
Me
Ph
I
5
H
Cp*
N
Cp*
N
- HI
H
Ir
6
7[f][g]
0
Me
Ph
H
(9)
Ir
Me
Ph
Ph
I
I
Cp*
N
I
0.093
N/D
0
H
i
X
Me
Ph
Ir
Me
Ph
N
N
[a] 0.1 M; [b] at end of experiment 6-24h, N/D = not detected [c] in
solvent 2,4-dimethyl-3-pentanol/toluene 1:1(v/v) [d] however, this represents
only 5% yield, the main products are dimers [e] compares with 0,042 min-1 in
reference 22; [f] in solvent EtOAc/ iPrOH 7:1 (v/v) [g] carried out at 60oC
N
Ph
H
H
Ph
(S)-(2)
(10)
(11a)
(12)
(11b)

4
Tetrahedron
Scheme 2. Proposed mechanism of formation of complex (1a)
interaction of two different catalyst species transferring either a
deuteride or deuteron to the coordinated imine/enamine to
explain the labelling pattern.
showing the failed nucleophilic attack of (S)-(2) on the
coordinated imine (10) and no observed dimer, but its reaction
with water to give acetophenone, complex (1a) and the
racemized amine (2).
For deuterium to appear at the chiral center of (2), the
deuteride must come from a catalyst that has dehydrogenated d8-
iso-propanol. Deuterium incorporation is also seen beta- to the
chiral center, confirming the reaction of an iridium-bound
enamine with deuteron. Indeed, multiple deuteron incorporation
is seen at the C1 methyl but not C3, the N-methyl group in (2), ESI
9.
Reactivation of the catalyst was tried by adding sodium iodide
d6-DMSO solution into a reaction mixture containing (1a-c) in an
NMR tube. The spectra showed an increase in the concentration
of (1) and (1a), but also what might be the anionic triiodoiridium
species IrCp*I3Na (1f). Unfortunately, including NaI in the
racemization of (S)-(2) made no difference to the rate or end
point. Hydroiodic acid was also tested but resulted in loss of the
amine to side reactions.
3. Conclusions
The studies reported here explain the observed differences in
iodo-iridium mediated racemization rates of acyclic N-alkyl and
heterocyclic amines. Catalyst deactivation was suspected when
N-methylamines failed to completely racemize. Deactivation was
found to occur by strong coordination of ammonia from primary
amines, and methylamine from N-methyl amines. This occurs
through hydrolysis or aminolysis of the catalyst-activated imine.
Several N-methyl coordinated iridium species have been
observed by 1NMR and are in dynamic equilibrium. Although the
racemization activity of each is still unknown, the calculated
equilibria show the disubstituted methylamine-iridium complex
(1b) predominates, and this may be the inactive catalyst species.
This would explain why racemization of N-iso-propyl or
heterocyclic amines do not deactivate the catalyst, as their larger
size may preclude disubstitution. For N-methyl amines,
poisoning of the catalyst was prevented by ensuring anhydrous
conditions, whilst adding hydrogen donor alcohols helped shift
the equilibrium away from the imine and slightly improved the
racemization rates. The addition of methanol was detrimental to
the reaction as it seems to behave like water in alcoholising the
catalyst-activated imine. The addition of sodium iodide was also
successful in moving the equilibrium away from the amine
coordinated species, though its use within the reaction was
unsuccessful. Some insights into the catalytic mechanism were
gained with continuous flow NMR with observation of
hydridoiridium species and by +ESI-MS amine coordinated
complexes, though more work is required to more fully
characterize catalytic intermediates. The use of d8-iso-propanol
solvent showed deuterium incorporation at positions alpha- and
beta- to the chiral center through deuteride and deuteron addition.
This indicates both imine/iminium and enamine coordinated
species are present in the manifold, and that separate catalytic
species are interacting to produce the observed products. Further
studies on the use of the SCRAM catalyst to introduce deuterium
into amines will be reported elsewhere.
Another way to overcome the problem of catalyst inhibition is
to encourage reduction of the imine by adding hydrogen donor
solvents into the reaction. Figure 5 shows the effect of adding
20%(v/v) alcohols to the racemization of (S)-(2) in ethyl acetate.
Figure 5. The effect of alcohol donor solvents on
racemization rates of (S)-(2) with 1 mol% catalyst (1) at 60°C.
The racemization half-life (t1/2) with ethyl acetate is 66
minutes corresponding to second order rate constant of kcat of
1.77x10-2 M-1.s-1 (compared to 2.16x10-2 M-1.s-1 in toluene at the
same temperature)22 and complete racemization is observed.
Whereas, the addition of iso-propanol and ethanol slightly
increase the kcat to 1.83x10-2 M-1.s-1, with end points of 6% and
13% e.e. respectively. The effect of methanol is more marked,
with a reduced kcat of 1.49x10-2 M-1.s-1 and end point of 50% e.e.
Methanol may coordinate to the iridium in the same way as
methylamine, Scheme 1, or act as a nucleophile, as in Scheme 2,
though the mechanism has not yet been investigated. The small
rate increase seen with ethanol and iso-propanol can be attributed
1
to their hydrogen donor properties. Time resolved HNMR and
4. Experimental
+ESI-MS deuterium labelling studies using d8-iso-propanol,
indicate a bimolecular catalytic mechanism in which the
deuterido-iridium complex formed from the alcohol, reduces the
coordinated imine, Scheme 3 and ESI 8.
Materials and equipment. Unless otherwise stated, all
chemicals were obtained from Sigma-Aldrich, Fisher Scientific,
Alfa Aesar or Fluorochem and were used without further
purification. Catalyst (1) was prepared as described previously.20
All solvents used were HPLC grade.
General procedures for racemization in batch. The
racemization of (S)-(2) (0.5 M) in toluene by [IrCp*I2]2 (1) is
described as an example. Catalyst (1) (5.1 mg, 4.4 µmol) and a
stirrer bar were added to a three-neck round bottom flask and
flushed with nitrogen for 10 minutes. (S)-(2) (0.3 g, 2.2 mmol,
323 µL) and n-decane (31.2 mg, 0.22 mmol, 43 µL) were
dissolved into toluene (4 mL, dried over molecular sieves). The
solution was added to the catalyst. The reaction was stirred and
heated to 105°C for 24 hours. The reaction was sampled (20 µL)
before heating and during the reaction by diluting the sample into
ethyl acetate (980 µL). The samples were analyzed by GC for
Scheme 3. Proposed mechanism for the observed
incorporation of deuterium into (S)-(2) which requires the

5
conversion and chiral GC for the determination of e.e., after
derivatization with trifluoroacetic anhydride (30 µL).
A similar procedure was employed for the racemization of
other substrates under designated conditions (Table S1). The e.e.
of the samples were determined by either chiral GC, HPLC or
1HNMR (Table S2).
2. H. Morris and J. Wallach, From PCP to MXE: a comprehensive review
of the non-medical use of dissociative drugs. Drug Testing and Analysis
2014, 6 (7-8), 614-632.
3. D. A. Pflum, H. S. Wilkinson, G. J. Tanoury D. W. Kessler, H. B. Kraus,
C. H. Senanayake. and S. A. Wald, Org. Proc. Res. Dev., 2001, 5 (2), 110-
115.
Racemization of (S)-(2) (0.5 M) in batch with spiking of (S)-
(2). After 25 hours, where the racemization stopped or reached
0% e.e., (S)-(2) (0.5 equiv. with respect to the original amount of
(S)-(2) in the reaction) and n-decane (0.25 equiv.) were dissolved
into toluene and charged to the reaction mixture at 105°C. The
reaction was sampled immediately after the addition and at
specific times.
Racemization of (S)-(2) (0.5 M) in batch with spiking of
methylamine (2 or 20 equiv. with respect to iridium). A similar
procedure to above was employed. Catalyst (1) (25.8 mg, 22.2
µmol) and tetrahydrofuran (4 mL) were used and the reaction
was heated to 60°C. After 30 minutes, where the e.e. of (2)
reached approximately 50%, methylamine/ tetrahydrofuran
solution (1.7 M, 52.2 µL, 88.7 µmol) was added. The reaction
was sampled 15 minutes after the addition and at specific times
for 24 hours. The same procedures were repeated for spiking 20
equiv. of methylamine (520 µL, 0.88 mmol) after 30 minutes
from the start of the reaction.
4. V. T. Mathad, G. B. Shinde, S. S. Ippar, N. C. Niphade, R. K. Panchangam
and P. J. Vankawala, Syn. Commun. 2011, 41 (3), 341-346.
5. C. M. Brandel, J. W. B. Cooke, R. A. J. Horan, F. P. Mallet and D. R.
Stevens, Org. Proc. Res. Dev. 2015, 19 (12), 1954-1965.
6. E. J. Ebbers, G. J. A. Ariaans, J. P. M. Houbiers, A. Bruggink, A. and B.
Zwanenburg, Tetrahedron 1997, 53 (28), 9417-9476.
7. J. H. Lee, K. Han, M.-J. Kim and J. Park, Eur. J. Org. Chem. 2010, (6),
999-1015.
8. N. J. Turner, Curr. Opin. Chem. Biol. 2010, 14 (2), 115-121.
9. C.-J. Jiang and G.-L. Cheng, Curr. Org. Chem. 2013, 17 (11), 1225-1234.
10. K. M. J Brands and A. J. Davies, Chem. Rev. 2006, 106 (7), 2711-2733.
11. W. L Noorduin, T. Izumi, A Millemaggi, M. Leeman, H. Meekes, W. J. P
Van Enckevort, R. M Kellogg, B. Kaptein, E.Vlieg and D. G. Blackmond, J.
Am. Chem. Soc., 2008, 130 (4), 1158-1159.
12. O. Verho and J.-E. Backvall, J. Am. Chem. Soc. 2015, 137 (12), 3996-
4009.
13. M. T. Reetz and K. Schimossek, Chimia International Journal for
Chemistry 1996, 50 (12), 668-669.
NMR titration of catalyst (1) with methylamine. Catalyst (1)
(3.1 mg, 2.7 µmol, 5.3 µmol iridium) and benzene (9.1 mg, 10
µL, 0.117 mmol) were dissolved into d6-DMSO (0.6 mL). The
sample was analyzed by 1HNMR (500 MHz). Methylamine/
methanol solution (2 M) was added as shown in Table S4. After
each addition, the sample was shaken for 60 seconds and the
1HNMR was recorded.
14. A. N. Parvulescu, P. A Jacobs and D. E. De Vos, Chemistry – A
European Journal 2007, 13 (7), 2034-2043.
15. O. Pamies, A. H. Ell, J. S. M. Samec, N. Hermanns and J.-E. Backvall,
Tetrahedron Letts. 2002, 43 (26), 4699-4702.
16. J. Paetzold and J. E. Backvall, J. Am. Chem. Soc., 2005, 127 (50), 17620-
1.
17. A. J. Blacker, S. Brown, B. Clique, B. Gourlay, C. E. Headley, S. Ingham,
D. Ritson, T. Screen, M. J. Stirling and D. Taylor, Org. Proc. Res. Dev.,
2009, 13 (6), 1370-1378.
General procedure for deuterium/ proton exchange of chiral
amines by d8-iso-propyl alcohol. Catalyst (1) (29.1 mg, 25.0
µmol) was added to a microwave tube with a magnetic stirrer.
The chiral amine (1 mmol) was dissolved into d8-iso-propyl
18. A. J Blacker and C. E. Headley in Green Chemistry in the Pharmaceutical
Industry (Eds. P. J. Dunn, A. S. Wells, M. T. Williams) 2010, 269-288
Wiley VCH Verlag GmbH & Co. KGaA.
1
alcohol (766 µL) and d8-toluene (306 µL), analyzed by HNMR
(300 MHz) then added to the catalyst and heated to 110°C by
microwave. After each hour of heating, the reaction mixture was
cooled down to room temperature and sampled for 1HNMR
analysis, then returned to the microwave vial and reheated to
110°C by microwave. The heating, cooling and sampling process
was repeated for 4 to 6 hours. For amines (2) and (6), a d6-DMSO
insert (which was prepared by filling a melting point capillary
tube with d6-DMSO and sealed at both ends) was placed into the
NMR sample for analysis.
19. A. J. Blacker, M. J. Stirling and M. I. Page, Org. Proc. Res. Dev. 2007, 11
(3), 642-648.
20. M. J. Stirling, A. J. Blacker and M. I. Page, Tetrahedron Letts. 2007, 48,
1247-1250.
21. Manuscript submitted for publication
22. M. J. Stirling, J. M. Mwansa, G. Sweeney, A. J. Blacker and M. I. Page,
Org. Biomol. Chem. 2016, 14 (29), 7092-7098.
23. A. M. R. Hall, P. Dong, A. Codina, J. P. Lowe and U. Hintermair, ACS
Catalysis 2019 9 (3), 2079-2090
24. R. Kawahara, F.-i. Fujita and R. Yamaguchi, J. Am. Chem. Soc., 2010,
132 (43), 15108-15111.
Declaration of competing interest
25. R. Kawahara, K.-i. Fujita and R. Yamaguchi, Adv. Syn. Cat., 2011, 353
(7), 1161-1168.
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
26. J. Vicente, M. T. Chicote, I. Vicente-Hernández and D. Bautista, Inorg.
Chem., 2008, 47 (20), 9592-9605.
Acknowledgements
We gratefully acknowledge AstraZeneca, Apex Molecular and
EPSRC for the financial support of M.H.T.K. and N.P.B.P.
through CASE awards EP/L50550X/1, EP/R51200X1
respectively. Dr Catherine Lyall and Dr John Lowe at the
Dynamic Reaction Monitoring (DReaM) Facility at the
University of Bath are kindly acknowledged for their assistance
in continuous flow NMR and MS experiments.
Notes and references
1. J. Wouters and L. Quéré, Pharmaceutical Salts and Co-crystals. RSC
Publishing, 2011.

Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests: