Harnessing asymmetric N-heterocyclic carbene ligands to optimise SABRE hyperpolarisation

Article abstract of DOI:10.1039/c8cy01214h

The catalytic signal amplification by reversible exchange process has become widely used for the hyperpolarisation of small molecules to improve their magnetic resonance detectability. It harnesses the latent polarisation of parahydrogen, and involves the

Full text of DOI:10.1039/c8cy01214h

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Harnessing asymmetric N-heterocyclic carbene
ligands to optimise SABRE hyperpolarisation†
Cite this: DOI: 10.1039/c8cy01214h
Chin Min Wong,ab Marianna Fekete, c Rhianna Nelson-Forde,c
b
c
c
Mark R. D. Gatus,
Simon B. Duckett
Peter J. Rayner,
and Barbara A. Messerle
Adrian C. Whitwood,
c
b
*
*
The catalytic signal amplification by reversible exchange process has become widely used for the
hyperpolarisation of small molecules to improve their magnetic resonance detectability. It harnesses the la-
tent polarisation of parahydrogen, and involves the formation of a labile metal complex that often contains
an N-heterocyclic carbene (NHC) ligand (e.g. [IrIJH)2IJNHC)IJpyridine)3]Cl), which act as a polarisation transfer
catalyst. Unfortunately, if the target molecule is too bulky, binding to the catalyst is poor and the
hyperpolarisation yield is therefore low. We illustrate here the behaviour of a series of asymmetric NHC
containing catalysts towards 3,4- and 3,5-lutidine in order to show how catalyst design can be used to dra-
matically improve the outcome of this catalytic process for sterically encumbered ligands.
Received 11th June 2018,
Accepted 17th July 2018
DOI: 10.1039/c8cy01214h
rsc.li/catalysis
SABRE involves the catalytic transfer of polarisation into a
hyperpolarisation target (L) from what is effectively p-H2.
Introduction
N-Heterocyclic carbenes (NHCs) are used widely as ligands in
an array of chemical transformations such as hydrogenation,
hydrogen transfer, hydroformylation, hydrosilylation, oxida-
tion, isomerisation, telomerisation and various C–C bond
forming reactions.1–4 By varying the N-bound substituents of
classic imidazole based NHCs it is possible to diversify their
electronic and steric properties to control reactivity.5 For the
free NHCs themselves, bulky substituents provide kinetic
stabilisation whilst opening the carbene bond angle favours
the triplet rather than the electronic singlet state formation.6,7
When an NHC is bound to a metal complex the steric bulk of
the NHC can be assessed via its percent buried volume8–10
while the electronic effect is based on Tolman's electronic pa-
rameter.11 Both affect catalytic reactivity.
Here we target a novel form of catalysis, that is abbrevi-
ated to SABRE, signal amplification by reversible exchange.12
This process takes the latent magnetism of parahydrogen (p-
H2) and transfers it into a new material according to the
process that is outlined in Scheme 1. Unlike a normal cata-
lytic reaction, which chemically functionalises a substrate,
The reversible addition of p-H2 is therefore important, as it
places the source of hyperpolarisation within the coordina-
tion sphere of the metal complex. Hence, the ligand ex-
change rates of the catalyst are important factors when con-
sidering how the catalyst works.12–14
When this process happens in low magnetic field, the ef-
fects of chemical shift and coupling become important. Once
a second order spin system is created, it becomes possible to
transfer the p-H2 derived hydride ligand polarisation into the
NMR active nuclei and then to a reversibly bound ligand
through the complex's scalar coupling network.15 The impor-
tance of this approach stems from the fact that the related
hyperpolarisation method, dynamic nuclear polarisation
(DNP),16 takes materials such as pyruvate and enhances their
Magnetic Resonance Imaging (MRI) detectability to the point
where they can be seen in vivo. In the case of pyruvate,17 the
a School of Chemistry, University of New South Wales, Sydney 2052, Australia
b Department of Molecular Sciences, Macquarie University, North Ryde 2109,
Australia. E-mail: barbara.messerle@mq.edu.au
c Centre for Hyperpolarisation in Magnetic Resonance, York Science Park,
University of York, Heslington, York YO10 5NY, UK.
E-mail: simon.duckett@york.ac.uk
Scheme 1 Conceptual representation of the signal amplification by
reversible exchange process which achieves the catalytic
hyperpolarisation of a substrate (L) via polarisation transfer within the
iridium catalyst from a pair of protons that were previously located in a
molecule of p-H2.
† Electronic supplementary information (ESI) available: NMR raw data can be
found at https://doi.org/10.15124/42187464-6f06-4805-a487-9b469b7a15e4. CCDC
1848485. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c8cy01214h
This journal is © The Royal Society of Chemistry 2018
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
subsequent assessment of its metabolism has proven to be
diagnostic of human cancer.17–20 Another approach using
hyperpolarised 129Xe has progressed to the point it is used
for the diagnosis of diseases of the lung in humans.21 Hence,
if SABRE catalysts can be optimised sufficiently, there is an
expectation that a simple low cost route to hyperpolarised
agents for disease diagnosis in humans would be possible.
The benefits of SABRE for analytical science are however also
likely to be substantial.22–25
It is common to use precatalysts based on the
[IrIJNHC)IJCOD)Cl] (1) motif in SABRE, and they react according
to Scheme 2 to form active polarisation transfer catalysts such
as [IrIJH)2IJNHC)IJpy)3]Cl (2). During this process cyclooctadiene
(COD) is first converted into cyclooctene (COE) and then
cyclooctane (COA) and hence whilst the reaction steps shown
are reversible, in practice 2 is stable.26 The ligand exchange
rate of the substrate, in this case pyridine (py), also influences
the efficiency of catalytic polarisation transfer because upon
ligand loss the propagating spin–spin coupling framework
within the catalyst is lost. Ideally, the lifetime of this complex
needs to be commensurate with these small scalar couplings
that are responsible for hyperpolarisation transfer.26,27
revealed by SABRE is reflected in its ability to create
hyperpolarised magnetic states with long lifetimes,47 an array
of research that is of growing importance as hyperpolarisation
methods are beginning to feature in clinical diagnosis.
One of the limitations of SABRE has proven to be associ-
ated with its ability to hyperpolarise sterically encumbered
agents due to the steric properties of the commonly used
NHC IMes complex which prevents adequate binding.48 How-
ever, without some steric interaction ligand loss will be
slow.26 Here we prepare a series of complexes using a range of
asymmetrically substituted NHC's where we maintain some
steric congestion and assess them for their SABRE activity.
We achieve this by maintaining one mesityl arm in conjunc-
tion with the introduction of a smaller substituent to the sec-
ond nitrogen centre. These substituents were selected care-
fully to prevent catalyst deactivation through well-known C-H
bond activation.49 The use of asymmetry in this way to drive
exchange is somewhat analogous to the ansa-bridged meta-
llocenes that are used to create isotactic polymers.50 We start
our studies with pyridine, which reflects the most widely used
model ligand in SABRE51 before considering 3,4- and
3,5-lutidine so as to introduce additional steric interactions
that we predict will lead to improved SABRE behaviour. It is
noteworthy that previous studies with lutidine have identified
it as a promising tool for pH imaging although steric effects
acted to limit SABRE-hyperpolarization effects52,53 and
quench it totally in ortho-substituted lutidines and picolines.54
It was found that iridium catalysts containing NHCs28,29
were more active than their mono phosphine counterparts be-
cause their lifetimes were closer to the optimum.30 Whilst
other iridium complexes with bidentate,31 pincer ligands,32
and rhodium analogues have been tested33 NHCs have still
proven to reflect the best ligand class and they have been
modified to work in aqueous solution.34 In fact SABRE trans-
fers its hyperpolarisation successfully through multiple bonds
to nitrogen containing compounds such as nicotinamide,35
isoniazid,36 pyrazole,37 acetonitrile38 and amines39 to improve
the detectability of appropriate 1H, 13C, 31P, 15N and 19F
responses.40–44 It has also proven possible to use radio fre-
quency excitation to mimic the low-field condition in order to
allow high-field transfer.15,45,46 A further exciting opportunity
Results and discussion
1a and the asymmetric complexes 1b–1e were prepared
according to Scheme 3. In the case of the yellow 1b, this ne-
cessitated the silverIJI) oxide mediated transmetalation of
[IrIJCOD)Cl]2. However, for 1c–1e a modified literature proce-
dure55 involving BPh4 salts, [IrIJCOD)Cl]2 and K2CO3 was
−
employed. In all cases the yields exceeded 60% and the prod-
ucts were characterised by 2D NMR spectroscopy, mass
spectrometry and elemental analysis as detailed in the ESI.†
X-ray data for 1e is included which map on to other reported
structures.
Scheme 2 Route to the formation of the active SABRE catalyst 2 via
the reaction of 1 with H2 and pyridine. The COD is converted first the
COE and COA during this process.
Scheme
asymmetric carbene.
3 Synthetic routes to 1 containing the specified C2-
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Catalysis Science & Technology
Paper
rates were determined. These are detailed in Table 1 and fol-
low the same trend. Hence we can conclude that the associ-
ated ligand loss rates are all too small, when compared to 2a,
for optimal behaviour. This observation is consistent with
reported literature behaviour.56,57 When these experiments
were repeated at 313 K, the corresponding signal gains in-
creased although 2b proved unstable (see ESI†) and is not
discussed further. Notably, the overall SABRE efficiency trend
changes to 2a > 2d > 2c > 2e at 313 K.
Scheme 4 Formation of products 2, 3 and 4 from reactions with 1.
(b) 3,4-Lutidine and 3,5-lutidine SABRE activity. In view of
the fact that pyridine binds too tightly in 2b–2e we tested the
bulkier 3,4- and 3,5-lutidine. These targets have pKa values of
6.28 and 5.85 (ref. 58) respectively compared to 5.3 for pyri-
dine.59 Hence they should bind more strongly if the process
is not sterically inhibited and their ligand loss rates should
be lower. When CD3OD solutions of 1 and 5-equivalents of
these lutidines were examined under 3 bar of p-H2, the for-
mation of [IrIJH)2IJlut)3IJLNHC)]Cl (3 and 4) was indicated as de-
tailed in the ESI† in all cases. Tables 2 and 3 summarise the
hydride chemical shifts of 3 and 4, the associated slower
lutidine ligand loss rate and the ortho proton and total pro-
ton signal enhancement values that were obtained after SA-
BRE transfer at 60 G and detection at 9.4 T (Fig. 1).
SABRE activity analysis of precatalysts 1a–1e
(a) Pyridine SABRE activity. The reactivity of precatalysts 1
towards pyridine and H2 was first investigated. In order to do
this, each of the complexes 1 were dissolved in CD3OD solu-
tions that contained pyridine (py, 5-fold excess) prior to
adding p-H2. In all cases, the solutions changed colour from
yellow to colourless as
2 forms, with generic formula
[IrIJLNHC)IJH)2IJpy)3]Cl, where LNHC is the corresponding mono-
dentate carbene of Scheme 4. Hydride resonances were read-
ily seen at around −22 ppm in these NMR spectra due to
these species. We note that when 1b was re-examined with
15N labelled pyridine the corresponding hydride signal ex-
The signal enhancements seen for aromatic 1H resonances
of 3,4-lutidine with 3b–3e at 298 K proved to be larger than
those seen for pyridine with 2b–2e. This suggests that the
result of the chemical structure change and which focuses
polarisation transfer into three rather than five proton sites,
as for pyridine, is beneficial; this view is consistent with the
comparatively poor methyl proton enhancements seen. Upon
warming to 313 K the resulting signal gains increased such
that 3a yields an average aromatic proton signal gain of 1700-
fold whilst 3d remains the best competitor with a value of
510-fold.
2
hibits a JH–(15)N = 19.6 Hz splitting; two 15N resonances can
be seen at 236.1 and 253.4 ppm for the corresponding axial
and equatorial ligands respectively.
The SABRE activities of these species were then compared
1
at 298 K through a series of H NMR studies using the ‘shake
and drop’ method (see ESI†). The resulting per proton ortho
phenyl 1H NMR resonance signal gains, and the total free
pyridine substrates 1H NMR signal gains, are reported in
Table 1 for observations at 9.4 T. It can be readily seen that
the order of SABRE efficiency is 2a > 2d > 2e > 2c > 2b for
both data types. These values suggest that the performance
of none of these new complexes compare favourably with that
of 2a for pyridine signal enhancement. In order to identify
the reason for this, the corresponding pyridine ligand loss
In contrast, the bulkier 3,5-lutidine shows much better
signal enhancements at 298 K than 3,4-lutidine, which is now
reflected by the total gain of 1615-fold with 4d. In both cases,
the NHC 1-mesityl-3-benzylimidazole yields the best catalyst
Table 1 Rate of ligand loss per molecule in solution, level of pyridine 1H NMR signal enhancement (fold) at 9.4 T and 298 K and 313 K for 2
Complex
2a
2b
2c
2d
2e
Pyridine ligand loss rate (298 K, s−1
Ortho pyridine proton signal gain (per proton)
)
13.35
—
0.51
−48
0.71
1.28
−1452 81
5246 262
87.82
−0.77 0.04
2
7
−95
5
−68
3
and total proton gain, 298 K (fold)
2.1 0.1
147
304 15
6.82
209 10
7.36
Pyridine ligand loss rate (313 K, s−1
)
—
—
—
3.66
−158
Ortho pyridine proton signal gain (per proton)
and total proton gain, 313 K (fold)
−414 21
8
−196 10
−130
6
3137 156
544 27
677 34
450 22
Table 2 Hydride peak positions, 3,4-lutidine loss rate, and indicated 1H NMR signal enhancements yield for 3,4-lutidine after SABRE at 298 K
Complex
3a
3b
3c
3d
3e
Ir–H chemical shift (ppm)
−22.80
−22.42
—
−22.09
−22.27
−22.32
Ligand loss rate (s−1
)
9.72
0.39
0.76
1.9
Ortho proton signal gain (per proton)
and total proton gain (fold)
−270 13
—
—
−69
3
−86
4
−33
2
1709 85
471 23
506 25
223 11
This journal is © The Royal Society of Chemistry 2018
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 3 Hydride peak positions, 3,5-lutidine loss rate, and 1H NMR signal enhancements for 3,5-lutidine after SABRE at 298 K
Complex
4a
4b
4c
4d
4e
Ir–H chemical shift (ppm)
−22.79
2.10
−22.24
−22.35
1.01
−22.36
−22.34
1.9
Ligand loss rate (s−1
)
0.34
2.14
Ortho proton signal gain (per proton) and total proton gain (fold)
−145
7
−1.7 0.1
−125
6
−258 13
−86.7
4
740 37
10.5 0.5
719 36
1615 81
504 25
(3d and 4d) and we note there are two pairs of ortho phenyl
protons in [IrIJH)2IJlut)3IJLNHC)]Cl that provides a JHH coupling
value for polarisation transfer now lies between 70 and 80 G
with the methyl groups receiving a far greater share of the
signal.
4
link to their hydride ligands which should be similar to those
for pyridine. It is not surprising that both of these positions
therefore receive strong polarisation under similar SABRE.
Furthermore, it can be seen that, as predicted, the ability of
the SABRE target to receive hyperpolarisation is controlled by
the ligand scaffold.
Continuous flow experiments. The variation in the level of
SABRE observed is therefore strongly dependent on the
catalyst and substrate combination. According to the
literature, it is also influenced by the magnetic field
experienced by the sample at the point of polarisation
transfer. This situation arises because polarisation transfer is
relayed via the scalar coupling in the catalyst when under
weak-field conditions. Hence the chemical shift difference
that exists between the parahydrogen-derived hydride ligands
and the interacting spin-½ nuclei of the ligated substrate influ-
ence the level of polarisation transfer observed.12
Effects of NMR relaxation. In order to further rationalise
this behaviour, the relaxation times of the protons of these
three substrates were measured at 9.4 T in the absence of
catalysts in methanol-d4 whilst under H2 as detailed in
Table 4. The corresponding values determined under SABRE
conditions are also detailed. It is clear that, as expected, the
presence of the catalyst consistently reduces these values.35,62
We therefore determined the corresponding relaxation times
of key protons in 3 and 4 when they were prepared from 1
with just a 3.05 fold excess based on iridium.
From all these data it can be seen that whilst 4d has a very
similar ligand loss rate to 4a, the total signal enhancement it
provides is improved by almost 50%. This change is clearly
due to less efficient relaxation within the catalyst, which im-
proves by 18% in 4d versus 4a. In support of this, 4c and 4a
actually also deliver similar signal gains with similar relaxa-
tion times even though their ligand exchange rates differ by
>50%. In view of these observations we can state that the ef-
fects of ligand exchange on SABRE activity must be less criti-
cal than those of relaxation within the catalyst.
We used an automated polarizer that has been described
previously60,61 to examine this effect by reference to pyridine
1
and 3,4-lutidine. Fig. 2 shows how the H NMR signal gain for
3,4-lutidine varies with the changing polarisation transfer field
1
when using catalyst 3d. For H transfer, the optimal value for
pyridine was 70 G whilst for the aromatic resonances of
3,4-lutidine it is 60 G. It appears, however, that optimal transfer
into the aliphatic groups is more efficient at 70 G. It is highly
noteworthy that the methyl resonances of 3,4-lutidine are sub-
stantially enhanced when compared to related reports.35
13C SABRE transfer with 3 and 4
It has also been shown that SABRE can enhance the 13C re-
sponses of molecules. When the 13C hyperpolarisation of pyr-
idine was evaluated for 2a, limited 13C signal gains are ob-
served even though the IMes derivate works extremely well
Fig. 3 shows the corresponding polarisation transfer field
profile for 3,5-lutidine achieved with 4d. The optimal field
1
for H. This suggests that the ligand exchange rate is too fast
(13.35 s−1) for substantial direct 13C transfer at this field. In
contrast, the results with slower exchanging 2b–2e prove
Fig.
1
Typically thermal and hyperpolarised 1H NMR spectra of
Fig. 2 Polarisation transfer field profile of 3,4-lutidine using precursor
1d.
3,4-lutidine achieved by 3d illustrating the hyperpolarisation effect.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Catalysis Science & Technology
Paper
Fig. 3 Polarisation transfer field profile for 3,5-lutidine based on pre-
cursor 1d.
Fig. 4 (a) 13C{1H} NMR spectrum recorded with 1024 scans for a
mixture of 3d and 3,4-lutidine without p-H2; (b) 13C NMR spectra
recorded in one scan on the same sample after SABRE showing the
corresponding hyperpolarised carbon signals (the vertical expansion
setting used to display the spectrum in Fig 4 (a) is four times that used
in Fig. 4 (b)); the relaxation delay was 3 seconds alongside a 30° excita-
tion pulse.
Table 4 Observed T1 values for the indicated protons of 3,4-lutidine and
3,5-lutidine in the free material at 9.4 T without, and in the presence of
the indicated catalyst, alongside the corresponding values when bound
as a ligand trans to hydride. The initial ratio of 1 to the indicated ligand
was 1 : 3.05 to ensure free substrate was present
seen after transfer between 90–140 G. The meta-methyl 13C
signals (position 3) generally exhibited lower polarisation
levels to that of the para-methyl resonance (position 4). In
the case of the aromatic region, the meta quaternary carbon
at position 4 on 3,4-lutidine showed an individual enhance-
ment value of 544-fold at 40 G whereas the ortho-CH value
was 560-fold. The meta-CH resonance showed minimal
polarisation enhancement at 40 G and this increase to 50-
fold at 30 G. The trend in signal intensity with polarisation
transfer field for these 13C resonances is very different to that
Sample/agent
T1 (/s) value (site)
3,4-Lutidine
Without
3a, free ligand
3a, coordinated
3,5-Lutidine
Ortho
Meta
5.14
6.45
5.09
CH3
5.75 & 5.58
2.83 & 2.14
3.20 & 2.49
3.31 & 3.83
1.80 & 1.82
1.51 & 1.20
Without
21.30
3.23
3.13
3.86
2.97
3.74
3.81
19.58
6.26
4.30
5.47
6.91
6.63
6.34
6.63
2.10
2.44
2.45
1.92
2.49
2.35
4a, free ligand
4a, coordinated
4c, free ligand
4c, coordinated
4d, free ligand
4d, coordinated
1
seen in the corresponding H NMR spectra.
In addition, the maximum signal gain seen in the 1H
NMR spectra is reflected in the ortho-CH resonances,
followed by the methyl groups, and finally the meta-CH pro-
ton. Comparatively, 3a still yields significantly lower 13C
polarisation than 3d, achieving only a 222-fold gain for the
meta-quaternary carbon at 70 G. While the ortho-CH signal
could be seen, the remaining carbon resonances exhibited
signal enhancements of 20–300 fold depending on the mag-
netic field strength tested. Hence, the total 13C signal en-
hancements seen for 3d were 1300-fold after transfer at 40 G,
which was significantly higher than the corresponding 780-
fold at 0.5 G achieved with 3a as detailed in Fig. 5.
better with 2d being the most efficient. The ortho and meta
13C positions exhibit the strongest signal gains on the mole-
cule. When the polarisation transfer field (PTF) dependence
was considered, optimum transfer was achieved at 50 G for
2d with the total carbon signal enhancement being 430-fold
at this value.
When 3,4-lutidine was examined, 3d and 3e proved to be
similarly effective. Fig. 4 shows a typical NMR spectrum that
was obtained with 3d. For context, when a normal 13C{1H}
NMR spectrum was acquired, 1024 scans were needed to see
the associated carbon signals (Fig. 4a).
The signals that are seen in these fully coupled NMR spec-
tra contain anti-phase character that is associated with the
detection of a heteronuclear longitudinal two spin order term
(IzSz). In each case, the anti-phase splitting is ∼8 Hz in size
and reflects a long-range rather than direct one bond CH
splitting. This is fully consistent with related reported obser-
vations.15,38 The SABRE 13C activity of 3d was also investi-
gated more rigorously using the flow apparatus. The two
inequivalent meta- and para-methyl 13C signals (positions 3
and 4) of 3,4-lutidine were found to show relatively large en-
hancements after transfer between 0–90 G, with no signal
Precursor 1d also achieved the optimal 13C NMR signal
enhancement in 3,5-lutidine and the results showed a
lower level of enhancement compared to 3,4-lutidine,
which is in agreement with the corresponding 1H NMR
changes. Now signals for all of the carbon sites, except
the para position, are clearly visible after transfer in fields
between 0 and 130 G. At 30 G, the optimum magnetic
field, the total carbon NMR signal gain was 800-fold, with
a 380-fold contribution coming from the quaternary meta
carbon position, 96-fold for the methyl carbons, and the
remainder being due to ortho 13C signal. The full PTF pro-
file is presented in Fig. 6.
This journal is © The Royal Society of Chemistry 2018
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
eraged, a pair of very weak antiphase hydride signals is
detected at −12.68 and −17.67 ppm that share a common JH–H
coupling of −10 Hz. Subsequently the species giving rise to
this material converts into 2b, although significant decompo-
sition is evident.
When a similar sample of 1c was exposed to p-H2 at 273 K
in methanol-d4 a pair of similar strongly enhanced antiphase
hydride signals is detected. They now occur at −12.86 and
−17.82 ppm and share a JHH coupling of −5.5 Hz. Similar ob-
servations result from 1d and 1e and the associated NMR
spectra are illustrated in Fig. 7. We note that the rate of for-
mation of final products 2d and 2e exceeds that of 2b and 2c.
Fig. 5 Graph comparing the total 13C enhancement observed for the
carbon resonances of 3,4-lutidine when using complex 3d (green) and
Experimental
3a (orange) respectively as
a function of the strength of the
polarisation transfer field.
The asymmetric monocarbene ligands of Scheme 1 were
synthesised by modifying previously published procedures.63
Phenylimidazole, or mesitylimidazole, were reacted with the
respective iodomethane, benzyl bromide or phenethyl bro-
mide in acetonitrile and stirred at 25 °C or at reflux. The sol-
vent was evaporated and crude brown-oil was recrystallised
from methanol : diethyl ether mixtures to give off-white solids
[PhIMe]I, [MesIMe]I, [MesIBn]Br or [MesIEtPh]Br. These li-
gands were dissolved in dichloromethane and stirred for 1
hour with NaBPh4 to exchange the counter-ion. The mixture
was filtered and the solvent evaporated in vacuo to yield
[MesIMe]BPh4, [MesIBn]BPh4 or [MesIEtPh]BPh4 as white
crystalline solids.
Early reaction stages
During such SABRE studies it has proven common in the early
stages of reaction to observe complexes of the type
[IrIJH)2IJCOD)IJNHC)IJsub)]Cl. These materials form transiently
as 1 converts into the final tris substituted product. They actu-
ally reflect H2 addition to the initial substitution product
[IrIJsub)IJNHC)IJCOD)]Cl which forms from the precursor
[IrIJNHC)IJCOD)Cl]. Their observation is important because If
the initial substitution is slow then inefficient H2 addition to
[IrIJNHC)IJCOD)Cl] occurs. However, while the resulting product
[IrIJH)2IJNHC)IJCOD)Cl] goes on to yield common [IrIJH)2IJsub)-
IJNHC)IJCOD)]Cl the reaction can take weeks at 298 K. For com-
pleteness, the COD ligand present in these complexes is
converted to COE and finally COA during this process.
We therefore examined the early stages of these reactions
with pyridine but observed [IrIJpy)IJNHC)IJCOD)]Cl to form only
when the NHC was IMes. When the associated sample of 1b
containing a 5-fold excess of pyridine was exposed to para-
hydrogen in methanol-d4 at 263 K, and the solution warmed
to 273 K, no reaction was initially evident in the correspond-
ing 1-scan 1H NMR spectrum. However, when 16-scans are av-
Synthesis of IrIJI) complexes containing asymmetric
monocarbene ligands 1b–1e
1b was synthesised by the silverIJI) transmetallation of
[IrIJCOD)Cl]2. The reaction mixture was stirred overnight at 25
Fig.
6
Graph comparing the 13C enhancement observed for the
carbon resonances of pyridine, 3,4-lutidine and 3,5-lutidine when
using complex 3d as a function of the strength of the polarisation
transfer field.
Fig. 7 Hydride region of a series of 1H NMR spectra recorded using a
π/4 pulse (400 MHz, CD3OD, 273 K) showing signals for: (a) 5c; (b) 5d;
(c) 5e and (d) 5a.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Catalysis Science & Technology
Paper
°C, filtered and the solvent evaporated. The crude product Conflicts of interest
was then dissolved in dichloromethane, filtered, isolated and
recrystallised from dichloromethane: hexane as a bright yel-
There are no conflicts to declare.
low solid. 1c–1e were synthesised by heating [IrIJCOD)Cl]2 and
K2CO3 in acetone with the corresponding ligand at reflux for
4 hours. After cooling, the mixture was filtered, and the sol-
Acknowledgements
vent evaporated. The crude product was redissolved in
dichloromethane, and white solid precipitates.
[IrIJCOD)IJMesIMe)Cl] (1c), [IrIJCOD)IJMesIBn)Cl] (1d) and
[IrIJCOD)IJMesIEtPh)Cl] (1e) resulted as yellow solids, and were
used without further purification. All these complexes were
air-stable as solids.
Single crystals of and 1e suitable for X-ray structure deter-
mination were obtained by slow diffusion of a concentrated
dichloromethane solution of the complex in hexane at −20 °C
(see ESI†).
This work was supported by The Wellcome Trust (Grants
092506 and 098335) and the University of York. This research
was also supported under the Australian Research Council's
Discovery Projects funding scheme (project number
DP130101838). CMW thanks the Australian Government for
the award of an International Postgraduate Research Scholar-
ship (IPRS).
a
Notes and references
1 W. A. Herrmann, K. Öfele, D. v. Preysing and S. K.
Schneider, J. Organomet. Chem., 2003, 687, 229–248.
2 G. Altenhoff, R. Goddard, C. W. Lehmann and F. Glorius,
Angew. Chem., Int. Ed., 2003, 42, 3690–3693.
Conclusions
The results presented in this work demonstrate that the
asymmetric carbene ligands of Scheme 2 can be used to pre-
pare complexes of the type [IrIJNHC)IJCOD)Cl] (1). These com-
plexes have been shown to react with hydrogen and sub-
strates pyridine, 3,4-lutidine or 3,5-lutidine to form the
corresponding IrIJIII) products [IrIJH)2IJNHC)IJsub)3]Cl (2–4) of
Scheme 3. These products were then assessed for activity in
the SABRE process, which involves the catalytic transfer of
polarisation from p-H2 into the substrate to enhance its NMR
detectability. Key parameters in this process, namely the li-
gand exchange rates, and the relaxation times of the associ-
ated NMR resonances have been assessed and compared to
those of the symmetric carbene bearing [IrIJH)2IJIMes)IJsub)3]Cl
(2–4a), the established SABRE reference catalyst. For pyridine
and 3,4-lutidine poor activity was observed as a consequence
of slow ligand loss due to the reduced steric bulk associated
with these NHC's relative to IMes and strong interactions
with the associated nitrogen centres due to their high basic-
ity. However, when 3,5-lutidine was examined, the enhanced
steric encumbrance of this substrate out ways the pKa benefit
and increases the rates of ligand loss for the asymmetric
NHC's such that they become competitive with their IMes de-
rivatives. The result of this process is that 4d, achieved an or-
tho proton signal gain of 258-fold per proton in 3,5-lutidine
in contrast to the 145-fold value achieved by 4a at 9.4 T.
One further benefit of these catalysts is reflected in their
ability to sensitise the 13C response of these agents. Now the
enhanced stability works favourably as evidenced by the spec-
tra of Fig. 4. It is therefore likely that asymmetric carbenes of-
fer a route to further improve the ability of SABRE to enhance
the MR response of sterically encumbered substrates and
hence the range of applications. For efficient polarization
transfer to 13C, the utilization of a micro-Tesla field in con-
junction with the presence of 15N to relay polarization trans-
fer has proven highly efficient thereby suggesting a route to
improve on these results.64,65
3 M. Eckhardt and G. C. Fu, J. Am. Chem. Soc., 2003, 125,
13642–13643.
4 H. Lebel, M. K. Janes, A. B. Charette and S. P. Nolan, J. Am.
Chem. Soc., 2004, 126, 5046–5047.
5 H. Jacobsen, A. Correa, A. Poater, C. Costabile and L.
Cavallo, Coord. Chem. Rev., 2009, 253, 687–703.
6 B. C. Gilbert, D. Griller and A. S. Nazran, J. Org. Chem.,
1985, 50, 4738–4742.
7 D. Bourissou, O. Guerret, F. P. Gabbaï and G. Bertrand,
Chem. Rev., 2000, 100, 39–92.
8 A. C. Hillier, W. J. Sommer, B. S. Yong, J. L. Petersen, L.
Cavallo and S. P. Nolan, Organometallics, 2003, 22,
4322–4326.
9 S. Díez-González and S. P. Nolan, Coord. Chem. Rev.,
2007, 251, 874–883.
10 H. Clavier and S. P. Nolan, Chem. Commun., 2010, 46,
841–861.
11 C. A. Tolman, Chem. Rev., 1977, 77, 313–348.
12 R. W. Adams, J. A. Aguilar, K. D. Atkinson, M. J. Cowley,
P. I. P. Elliott, S. B. Duckett, G. G. R. Green, I. G. Khazal, J.
Lopez-Serrano and D. C. Williamson, Science, 2009, 323,
1708–1711.
13 R. W. Adams, S. B. Duckett, R. A. Green, D. C. Williamson
and G. G. R. Green, J. Chem. Phys., 2009, 131, 194505.
14 K. D. Atkinson, M. J. Cowley, P. I. P. Elliott, S. B. Duckett,
G. G. R. Green, J. López-Serrano and A. C. Whitwood, J. Am.
Chem. Soc., 2009, 131, 13362–13368.
15 A. N. Pravdivtsev, A. V. Yurkovskaya, H. Zimmermann, H.-M.
Vieth and K. L. Ivanov, RSC Adv., 2015, 5, 63615–63623.
16 J.-H. Ardenkjaer-Larsen, G. S. Boebinger, A. Comment, S.
Duckett, A. S. Edison, F. Engelke, C. Griesinger, R. G.
Griffin, C. Hilty, H. Maeda, G. Parigi, T. Prisner, E. Ravera, J.
van Bentum, S. Vega, A. Webb, C. Luchinat, H. Schwalbe
and L. Frydman, Angew. Chem., Int. Ed., 2015, 54, 9162–9185.
17 K. Golman, R. i. t. Zandt, M. Lerche, R. Pehrson and J. H.
Ardenkjaer-Larsen, Cancer Res., 2006, 66, 10855–10860.
This journal is © The Royal Society of Chemistry 2018
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
18 I. Marco-Rius, M. C. D. Tayler, M. I. Kettunen, T. J. Larkin,
K. N. Timm, E. M. Serrao, T. B. Rodrigues, G. Pileio, J. H.
Ardenkjaer-Larsen, M. H. Levitt and K. M. Brindle, NMR
Biomed., 2013, 26, 1696–1704.
19 S. R. Y. Xu, C. Ø. Mariager, L. B. Bertelsen, M. Schroeder, H.
Qi, C. Laustsen and H. Stødkilde-Jørgensen, Tomography,
2017, 3, 67–73.
35 P. J. Rayner, M. J. Burns, A. M. Olaru, P. Norcott, M. Fekete,
G. G. R. Green, L. A. R. Highton, R. E. Mewis and S. B.
Duckett, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, E3188–E3194.
36 H. Zeng, J. Xu, J. Gillen, M. T. McMahon, D. Artemov, J.-M.
Tyburn, J. A. B. Lohman, R. E. Mewis, K. D. Atkinson,
G. G. R. Green, S. B. Duckett and P. C. M. van Zijl, J. Magn.
Reson., 2013, 237, 73–78.
20 C. H. Cunningham, J. Y. C. Lau, A. P. Chen, B. J. Geraghty,
W. J. Perks, I. Roifman, G. A. Wright and K. A. Connelly,
Circ. Res., 2016, 119, 1177–1182.
21 J. P. Mugler and T. A. Altes, Magn. Reson. Imaging, 2013, 37,
313–331.
37 E. B. Dücker, L. T. Kuhn, K. Münnemann and C. Griesinger,
J. Magn. Reson., 2012, 214, 159–165.
38 R. E. Mewis, R. A. Green, M. C. R. Cockett, M. J. Cowley,
S. B. Duckett, G. G. R. Green, R. O. John, P. J. Rayner and
D. C. Williamson, J. Phys. Chem. B, 2015, 119, 1416–1424.
39 W. Iali, P. J. Rayner and S. B. Duckett, Sci. Adv., 2018, 4,
eaa06250.
22 N. K. J. Hermkens, R. Aspers, M. C. Feiters, F. Rutjes and M.
Tessari, Magn. Reson. Chem., 2018, 56, 633–640.
23 I. Reile, N. Eshuis, N. K. J. Hermkens, B. J. A. van
Weerdenburg, M. C. Feiters, F. Rutjes and M. Tessari,
Analyst, 2016, 141, 4001–4005.
24 N. Eshuis, R. Aspers, B. J. A. van Weerdenburg, M. C. Feiters,
F. Rutjes, S. S. Wijmenga and M. Tessari, Angew. Chem., Int.
Ed., 2015, 54, 14527–14530.
25 L. S. Lloyd, R. W. Adams, M. Bernstein, S. Coombes, S. B.
Duckett, G. G. R. Green, R. J. Lewis, R. E. Mewis and C. J.
Sleigh, J. Am. Chem. Soc., 2012, 134, 12904–12907.
26 L. S. Lloyd, A. Asghar, M. J. Burns, A. Charlton, S. Coombes,
M. J. Cowley, G. J. Dear, S. B. Duckett, G. R. Genov, G. G. R.
Green, L. A. R. Highton, A. J. J. Hooper, M. Khan, I. G.
Khazal, R. J. Lewis, R. E. Mewis, A. D. Roberts and A. J.
Ruddlesden, Catal. Sci. Technol., 2014, 4, 3544–3554.
27 N. Eshuis, R. Aspers, B. J. A. van Weerdenburg, M. C. Feiters,
F. Rutjes, S. S. Wijmenga and M. Tessari, J. Magn. Reson.,
2016, 265, 59–66.
40 P. J. Rayner, M. J. Burns, A. M. Olaru, P. Norcott, M. Fekete,
G. G. R. Green, L. A. R. Highton, R. E. Mewis and S. B.
Duckett, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, E3188–E3194.
41 J. F. P. Colell, A. W. J. Logan, Z. Zhou, R. V. Shchepin, D. A.
Barskiy, G. X. Ortiz, Q. Wang, S. J. Malcolmson, E. Y.
Chekmenev, W. S. Warren and T. Theis, J. Phys. Chem. C,
2017, 121, 6626–6634.
42 R. V. Shchepin, B. M. Goodson, T. Theis, W. S. Warren and
E. Y. Chekmenev, ChemPhysChem, 2017, 18, 1961–1965.
43 M. J. Burns, P. J. Rayner, G. G. R. Green, L. A. R. Highton,
R. E. Mewis and S. B. Duckett, J. Phys. Chem. B, 2015, 119,
5020–5027.
44 D. A. Barskiy, R. V. Shchepin, C. P. N. Tanner, J. F. P. Colell,
B. M. Goodson, T. Theis, W. S. Warren and E. Y.
Chekmenev, ChemPhysChem, 2017, 18, 1493–1498.
45 T. Theis, M. Truong, A. M. Coffey, E. Y. Chekmenev and
W. S. Warren, J. Magn. Reson., 2014, 248, 23–26.
28 L. S. Lloyd, A. Asghar, M. J. Burns, A. Charlton, S. Coombes,
M. J. Cowley, G. J. Dear, S. B. Duckett, G. R. Genov, G. G. R.
Green, L. A. R. Highton, A. J. J. Hooper, M. Khan, I. G.
Khazal, R. J. Lewis, R. E. Mewis, A. D. Roberts and A. J.
Ruddlesden, Catal. Sci. Technol., 2014, 4, 3544–3554.
29 B. J. A. van Weerdenburg, S. Gloeggler, N. Eshuis, A. H. J.
Engwerda, J. M. M. Smits, R. de Gelder, S. Appelt, S. S.
Wymenga, M. Tessari, M. C. Feiters, B. Bluemich and
F. P. J. T. Rutjes, Chem. Commun., 2013, 49, 7388–7390.
30 R. W. Adams, S. B. Duckett, R. A. Green, D. C. Williamson
and G. G. R. Green, J. Chem. Phys., 2009, 131,
194505–194505.
46 S. S. Roy, G. Stevanato, P. J. Rayner and S. B. Duckett,
J. Magn. Reson., 2017, 285, 55–60.
47 S. S. Roy, P. J. Rayner, P. Norcott, G. G. R. Green and S. B.
Duckett, Phys. Chem. Chem. Phys., 2016, 18, 24905–24911.
48 M. J. Cowley, R. W. Adams, K. D. Atkinson, M. C. R. Cockett,
S. B. Duckett, G. G. R. Green, J. A. B. Lohman, R.
Kerssebaum, D. Kilgour and R. E. Mewis, J. Am. Chem. Soc.,
2011, 133, 6134–6137.
49 N. Phillips, J. Rowles, M. J. Kelly, I. Riddlestone, N. H. Rees,
A. Dervisi, I. A. Fallis and S. Aldridge, Organometallics,
2012, 31, 8075–8078.
50 H. H. Brintzinger and D. Fischer, in Polyolefins: 50 years
after Ziegler and Natta II: Polyolefins by Metallocenes and
Other Single-Site Catalysts, ed. W. Kaminsky, Springer Berlin
Heidelberg, Berlin, Heidelberg, 2013, pp. 29–42.
31 A. J. Ruddlesden, R. E. Mewis, G. G. R. Green, A. C.
Whitwood and S. B. Duckett, Organometallics, 2015, 34,
2997–3006.
32 A. J. Holmes, P. J. Rayner, M. J. Cowley, G. G. R. Green, A. C.
Whitwood and S. B. Duckett, Dalton Trans., 2015, 44,
1077–1083.
33 V. V. Zhivonitko, I. V. Skovpin and I. V. Koptyug, Chem.
Commun., 2015, 51, 2506–2509.
34 P. Spannring, I. Reile, M. Emondts, P. P. M. Schleker,
N. K. J. Hermkens, N. G. J. v. d. Zwaluw, B. J. A. v.
Weerdenburg, P. Tinnemans, M. Tessari, B. Blümich,
F. P. J. T. Rutjes and M. C. Feiters, Chem. – Eur. J., 2016, 22,
9277–9282.
51 W. Jiang, L. Lumata, W. Chen, S. Zhang, Z. Kovacs, C.
Khemtong and A. D. Sherry, Sci. Rep., 2015, 5, 9104.
52 W. Jiang, L. Lumata, W. Chen, S. Zhang, Z. Kovacs, A. D.
Sherry and C. Khemtong, Sci. Rep., 2015, 5, 9104.
53 R. V. Shchepin, D. A. Barskiy, A. M. Coffey, T. Theis, F. Shi,
W. S. Warren, B. M. Goodson and E. Y. Chekmenev, ACS
Sens., 2016, 1, 640–644.
54 R. V. Shchepin, M. L. Truong, T. Theis, A. M. Coffey, F. Shi,
K. W. Waddell, W. S. Warren, B. M. Goodson and E. Y.
Chekmenev, J. Phys. Chem. Lett., 2015, 6, 1961–1967.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Catalysis Science & Technology
Paper
55 R. Savka and H. Plenio, Dalton Trans., 2015, 44,
891–893.
Kerssebaum, D. Kilgour and R. E. Mewis, J. Am. Chem. Soc.,
2011, 133, 6134–6137.
56 J.-B. Hovener, N. Schwaderlapp, T. Lickert, D. Leibfritz, J.
Hennig, D. von Elverfeldt, B. Duckett Simon, E. Mewis Ryan,
A. R. Highton Louise, M. Kenny Stephen, G. R. Green Gary
and G. Korvink Jan, Nat. Commun., 2013, 4, 2946.
57 G. W. Vuister, M. Tessari, Y. Karimi-Nejad and B.
Whitehead, in Modern Techniques in Protein NMR, ed. N. R.
Krishna and L. J. Berliner, Springer US, Boston, MA, 2002,
pp. 195–257.
61 R. E. Mewis, K. D. Atkinson, M. J. Cowley, S. B. Duckett,
G. G. R. Green, R. A. Green, L. A. R. Highton, D. Kilgour,
L. S. Lloyd, J. A. B. Lohman and D. C. Williamson, Magn.
Reson. Chem., 2014, 52, 358–369.
62 R. E. Mewis, M. Fekete, G. G. R. Green, A. C. Whitwood and
S. B. Duckett, Chem. Commun., 2015, 51, 9857–9859.
63 C. Nieto-Oberhuber, S. Lopez and A. M. Echavarren, J. Am.
Chem. Soc., 2005, 127, 6178–6179.
58 C. S. Handloser, M. R. Chakrabarty and M. W. Mosher,
J. Chem. Educ., 1973, 50, 510–511.
59 C. S. Handloser, M. R. Chakrabarty and M. W. Mosher,
J. Chem. Educ., 1973, 50, 510.
60 M. J. Cowley, R. W. Adams, K. D. Atkinson, M. C. R. Cockett,
S. B. Duckett, G. G. R. Green, J. A. B. Lohman, R.
64 R. V. Shchepin, L. Jaigirdar, T. Theis, W. S. Warren, B. M.
Goodson and E. Y. Chekmenev, J. Phys. Chem. C, 2017, 121,
28425–28434.
65 T. Theis, M. L. Truong, A. M. Coffey, R. V. Shchepin, K. W.
Waddell, F. Shi, B. M. Goodson, W. S. Warren and E. Y.
Chekmenev, J. Am. Chem. Soc., 2015, 137, 1404–1407.
This journal is © The Royal Society of Chemistry 2018
Catal. Sci. Technol.