Rapid carbonylative coupling reactions using palladium(i) dimers: Applications to 11CO-radiolabelling for the synthesis of PET tracers

Article abstract of DOI:10.1039/c1ob05268c

Palladium dimers with sterically hindered phosphines have been shown to be excellent pre-catalysts for the aminocarbonylation of aryl halides to yield amides and one of them has been successfully employed as a pre-catalyst for the synthesis of ¹¹C-radiolabelled amides for PET imaging.

Full text of DOI:10.1039/c1ob05268c

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 3499
www.rsc.org/obc
PAPER
Rapid carbonylative coupling reactions using palladium(I) dimers: applications
1
1
to CO-radiolabelling for the synthesis of PET tracers†
Gabriella Buscemi, Philip W. Miller, Steven Kealey, Antony D. Gee,a,d Nicholas J. Long, Jan Passchierb
a
a
b,c
a
and Ramon Vilar*a
Received 14th January 2011, Accepted 24th February 2011
DOI: 10.1039/c1ob05268c
Palladium dimers with sterically hindered phosphines have been shown to be excellent pre-catalysts for
the aminocarbonylation of aryl halides to yield amides and one of them has been successfully employed
1
1
as a pre-catalyst for the synthesis of C-radiolabelled amides for PET imaging.
Introduction
metal mediated reactions using either palladium or rhodium
and involves the reaction of aryl, heteroaryl or vinyl halides
Positron emission tomography (PET) is a non-invasive imaging
technique that is primarily used as a research and clinical
diagnostic tool for generating an understanding of diseases such
11
2
with a nucleophile in the presence of CO. However, besides
the conventional difficulties of processing carbon-11 reactions
within the necessary short timeframes (typically within 2 half-
1
as cancer, Alzheimer’s and Parkinson’s. Fundamentally, all PET
11
lives, i.e. 40 min), [ C]carbon monoxide reaction processes are
procedures rely on the synthesis of a positron emitting tracer and
its administration to a subject prior to scanning. The preparation
of PET tracers is however an extremely challenging area of
synthetic chemistry mainly because of the time restraints imposed
by short half-lives of the commonly used positron emitting
isotopes nitrogen-13 (t1/2 = 9.96 min), oxygen-15 (t1/2 = 2.04 min),
further hampered by its poor reactivity. This poor reactivity is a
consequence of both the inherent low solubility of CO in common
organic solvents and the unavoidable high molecular dilution
11
of CO in inert carrier gases which results in very low partial
11
pressures of CO. A number of methods, including high pressure
4
5
6
reactors, microfluidics and solution trapping have now been
2
carbon-11 (t1/2 = 20.4 min) and fluorine-18 (t1/2 = 110 min). The
11
reported to enhance CO labelling procedures.
rapid growth of PET imaging and its applicability for the study of
a wide range of diseases and conditions has stimulated the need to
develop novel approaches for labelling a new generation of PET
tracers to probe underlying biological processes. Of the common
PET radioisotopes C is particularly attractive for radiotracer
synthesis since carbon is the main constituent of all biological
molecules which are likely to be of interest for PET studies.
The type of catalyst used for carbonylation reactions is well
known to have a marked effect on the rate of reaction, product
yields and selectivities. Most conventional carbonylation studies
7
are conducted over longer time periods, typically hours to days,
where there is a desire to design robust and long lived catalytic
1
1
8
species in order to maximise product yields. Transition metal-
11
mediated [ C]carbonylation reactions on the other hand require
1
1
Although CH I has traditionally been the precursor of choice
3
catalysts that are highly active over the short timeframes typical
1
1
for carbon-11 labelling, more recently [ C]carbon monoxide
has been demonstrated to be an attractive precursor because
of its ease of production and synthetic versatility. [ C]carbon
monoxide has been used to prepare a wide range of C labelled
molecules, such as amides, esters and carboxylic acids, that would
be inaccessible using traditional CH
11
of C labelling reactions in order to achieve high radiochemical
11
yields. The activity of the catalyst used for CO labelling protocols
3
11
11
may therefore be crucial to enhancing C labelling methods
1
1
for PET. There are indeed a very limited number of reported
catalysts that can perform carbonylative coupling reactions with
1
1
I labelling techniques.
9
3
amines and alcohols at atmospheric pressure of CO and, more
1
1
Synthesis with [ C]carbon monoxide is generally achieved via
10
importantly, over short periods of time (it should be noted that
most reported reactions times are over 1 h which would be too
long for applications into C labelling with CO).
a
Department of Chemistry, Imperial College London, South Kensington,
11
11
London, UK SW7 2AZ. E-mail: r.vilar@imperial.ac.uk; Fax: +44 2075
Previously our group has reported the synthesis of palladium
9
41139; Tel: +44 2075941967
11
b
dimeric complexes 1–4 which have been demonstrated to be very
GlaxoSmithKline, Clinical Imaging Centre, Imperial College, Hammer-
smith Hospital, London, UK W12 0NN
effective pre-catalysts for amination reactions of aryl chlorides
c
11d
12
Department of Pharmacy and Pharmacology, University of Bath, Bath, UK
and bromides and for Suzuki cross-coupling reactions. Impor-
tantly, extremely high conversions were observed in these reactions
over short reaction times (under 10 min) which are typically
BA2 7AY
d
Division of Imaging Sciences and Biomedical Engineering, King’s College
London, St Thomas’ Hospital, London, UK SE1 7EH
1
1
desired for C labelling protocols. Based on these observations, we
hypothesised that these palladium dimers could also be effective
†
Electronic supplementary information (ESI) available: Experimental
procedures and characterisation data. See DOI: 10.1039/c1ob05268c
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3499–3503 | 3499

View Article Online
a
Table 1 Comparison of catalysts in the aminocarbonylation of iodobenzene with benzylamine
◦
◦
◦
THF (60 C)
Toluene (100 C)
Mesitylene (150 C)
b
b
b
b
b
Complex
Yield [%]
Conversion [%]
Yield [%]
Conversion [%]
Yield [%]
1
2
3
4
4
5
5
6
5
5
45
45
76
98
—
38
—
89
33
22
64
88
—
27
—
64
57
56
99
99
45
43
87
86
35
30
—
13
—
29
d
d
99
84
94
52
80
42
d
d
91
74
a
Main product N-benzylbenzamide. Reaction conditions: iodobenzene (0.45 mmol), benzylamine (4.6 mmol), solvent (0.5 mL), complex (2.2 mol%),
b
1
0 min. Average of two or more reactions. For details see Supplementary Information. Isolated yields and conversions determined by GC using diphenyl
ether as internal standard and based on iodobenzene. A control experiment demonstrated that about 7% of iodobenzene is lost during the cannulation,
c
thus generally leading to higher conversions and lower yields. 2–3% a-ketoamide formed by double carbonylation was observed as by-product with all
d
the complexes. Reaction performed within 5 min.
0
pre-catalysts for carbonylative coupling reactions. Thus, herein we
present our investigations on the catalytic activity of palladium
dimeric complexes 1–4 (Fig. 1) for the aminocarbonylation
this model reaction along with the benchmark Pd complexes
t
[Pd(PPh
3
)
4
] (5) and [Pd(P Bu
3
)
2
] (6) for comparative purposes.
All reactions were performed over 10 min reaction periods,
to reflect the timescale of a typical C labelling reaction, under
1
1
11
reaction of aryl halides and present initial results on CO labelling
reactions using one of these dimers (4).
1 atm of carbon monoxide. Since solvent and temperature are
often key parameters in catalytic reactions that greatly influence
reaction rate, yield and catalyst decomposition, a series of
different solvents were screened (THF, toluene and mesitylene)
◦
over a range of temperatures (60, 100 and 150 C). The results
from these screening reactions are summarised in Table 1. This
aminocarbonylation reaction has previously been reported to
7b,13
be highly dependent on the reaction temperature;
the results
shown in Table 1 further confirm this with all catalysts giving low
◦
yields in THF at 60 C and significantly higher yields at higher
◦
◦
temperatures (100 C and 150 C using toluene and mesitylene
respectively). The type of catalyst used is also observed to have
a pronounced effect on product yields. The palladium bi-aryl
phosphine dimeric complexes 1 and 2 were shown to have only
modest catalytic activity for the aminocarbonylation reaction
of iodobenzene even under the higher temperature conditions
I
using toluene or mesitylene. In contrast, Pd dimers 3 and 4
◦
exhibited excellent catalytic activities at 100 C using toluene as
Fig. 1 Palladium complexes investigated as catalyst for carbonylative
coupling reactions.
solvent and were superior to the benchmark catalyst [Pd(PPh
3
)
4
]
t
(
5) but comparable in activity to [Pd(P Bu
3
)
2
] (6). Remarkably
t
[
Pd
2
(m-I)
2
(P Bu
3
)
2
] (4) gave almost quantitative conversion of
iodobenzene and exceptionally high isolated yields of the N-
benzylbenzamide product (88%) at 100 C in toluene within
◦
Results and discussion
1
0 min. Even when 1.1 mol% of 4 was used so that the Pd mol%
was equal to that of the monometallic compounds (2.2 mol%), a
4% isolated yield of the amide (cf . 88%) was obtained. Although
Aminocarbonylation reactions mediated by Pd dimers
7
The well characterised aminocarbonylation of iodobenzene and
benzylamine to form N-benzylbenzamide (7) (Scheme 1) was
selected as an initial model reaction for our studies. Pd dimeric
complexes 1–4 were investigated for their catalytic activity in
this is ca. 15% lower than the yield obtained when using 2.2 mol%
of 4, it suggests that only one of the two Pd centres from 4 is
actively involved in the catalysis.
◦
At higher temperatures of 150 C in mesitylene over a 10 min
reaction time quantitative conversions of iodobenzene were again
achieved by complexes 3 and 4. When the reaction time was
reduced further to only five minutes complex 4 again gave quan-
titative conversions of iodobenzene (99%) and very high isolated
yields (84%) of N-benzylbenzamide. In comparison, [Pd(PPh ) ]
3
4
Scheme 1 Model reaction used to study the catalytic activity of complexes
gave only modest conversions (52%) and yields (42%) under
identical conditions. We believe this is a particularly interesting
1
–6.
3
500 | Org. Biomol. Chem., 2011, 9, 3499–3503
This journal is © The Royal Society of Chemistry 2011

View Article Online
Table 2 Screening of iodo- and bromo- substrates with complex 4
Aryl halide Amine Product
a
a
Conditions
Conv [%]
55
Yield [%]
◦
1
50 C Mesitylene 10% (4)
46
◦
1
50 C Mesitylene 10% (4)
54
48
83
46
◦
1
1
50 C Mesitylene 10% (4)
46
82
◦
00 C Toluene 2.2% (4)
◦
b
b
1
00 C Toluene 2.2% (5)
63
59
Reaction conditions: Aryl, heteroaryl halide (0.45 mmol), amine (4.6 mmol), solvent (0.5 mL), complex (4) (10 or 2.2 mol%), 10 min. Results shown are
a
the average of two or more reactions. For details see Supplementary Information. Yields and conversions determined by GC using diphenyl ether as
b
internal standard and calculated on the aryl- or heteroaryl-halide. [Pd(PPh
3
)
4
] (5) employed as pre-catalyst.
finding as it demonstrates that carbonylation reactions can be very
efficiently performed at low carbon monoxide pressures within
remarkably short timeframes.
the more challenging 5-bromoindole substrates were used in these
cases.
Complex 4 was selected for further catalytic evaluation because
of the excellent yields obtained within short timeframes and also
because of its air and moisture stability compared to complex
1
1
C-labelling via carbonylative cross coupling
Once we had established that dimer 4 acts as an excellent pre-
catalyst for carbonylative cross-couplings under 10 min reac-
tion times, it was of interest to investigate its applicability for
radiolabelling experiments with CO. The carbonylative coupling
of iodobenzene with benzylamine to form N-benzylbenzamide,
and of 5-iodoindole with piperidine to form 10 were selected as
model reactions to test CO labelling efficiency.
In addition to carrying out the reactions with dimer 4, we also
investigated the catalysis using [Pd(PPh ) ] (5) for comparison.
To overcome the problems associated with CO radiolabelling
reactions described above (low solubility and low partial pres-
sures) a CO-trapping complex, recently reported by our group,
was used to perform CO syntheses. The copper(I) tris(3,5-
dimethylpyrazolyl)-borate [CuTp*] complex was found to be a
highly efficient reagent for both trapping CO and performing
3. A selection of more challenging aryl-bromides containing
deactivating electron donating methoxy groups were therefore
tested for the aminocarbonylation reaction with benzylamine
1
1
(
Table 2). Unfortunately using 2 mol% loading of complex 4 did
not give satisfactory conversions within the set 10 min timeframe.
Increasing the catalyst loading to 10 mol% did, however, result
in a significant improvement in conversions. Bromobenzene, 4-
bromoanisole and 2-bromoanisole were successfully converted
1
1
3
4
◦
11
at 150 C to their corresponding N-benzylamides in 10 min
with modest conversions ranging from 48–55%. Interestingly
the deactivating methoxy groups had no significant effect on
conversions or yields compared to bromobenzene over this
timeframe. The scope of catalyst 4 was further investigated by
performing the aminocarbonylative coupling of 5-iodoindole with
6b
1
1
1
1
◦
11
piperidine at 100 C in an attempt to synthesize the known
palladium mediated C carbonylative coupling reactions. In a
typical radiolabelling procedure a stream of CO in helium
14
11
central nervous system 5-HT2A receptor ligand 10. Complex 4
was demonstrated to give high yields (82%, Table 2) of the indole
amide 10 within 10 min at 1 atm of CO. In comparison, under the
carrier gas was passed through a solution of [CuTp*] in toluene.
On average >95% of radioactivity was found to be retained
in the [CuTp*] in toluene solution. The palladium complex (4
or 5) and the corresponding aryl iodide (previously dissolved
in anhydrous and degassed toluene) were then injected to the
same reaction conditions, [Pd(PPh
Previously reported carbonylation methods for the preparation of
0 only gave high yields when CO pressures were 25 bar and
3
)
4
] (5) gave a modest 59% yield.
1
15
11
◦
reaction times exceeded 12 h, however it must be noted that
[CuTp* CO] solution and the vial heated to 120 C for 2 min
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3499–3503 | 3501

View Article Online
11
Table 3 Results from [ C]carbonylation reactions using complex 4 and 5 as catalyst
a
11
11
Entry
1
Aryl halide
Amine
Conditions/catalyst
Trapped radioact. [%]
C Amide RCP [%]
C Amide RCY [%]
50 eq. amine
Catalyst: 4
87
81
68
17
88
70
47
9
2
3
4
5 eq. amine
Catalyst: 4
72
60
78
5 eq. amine
Catalyst: 5
5 eq. amine
69
Catalyst 4
5
5 eq. amine
Catalyst: 5
61
36
22
Reaction conditions: Aryl/heteroaryl iodide (50 mM), amine (see table for equivalents used with respect to the aryl iodide), toluene (100 mL), palladium
complex (6.4 mol% of Pd with respect to aryl iodide), 10 min reaction time. The results shown are the average of two or more reactions. For details see
a
Supplementary Information. Decay corrected radioactivity trapped in the reaction vial expressed as a fraction of total of activity (i.e. combination of
b
activity in reaction and unreacted gases in the bag). Decay corrected RCY (radiochemical yield) based on total radioactivity delivered to the reaction
vial and corrected for RCP (radiochemical purity).
to initiate the reaction. After this time, the corresponding amine
i.e. benzylamine or piperidine) was injected and the reaction
mixture heated for a further 8 min. The reaction was then purged
with argon to remove unreacted CO which was measured and
collected in a bag in a dose calibrator. The radioactivity remaining
in the reaction vial was then measured. A 20 mL aliquot was
removed and analysed by analytical radio-HPLC to determine
the distribution of radiolabelled products (RCP). As can be
seen from Table 3, the main product from the reaction between
iodobenzene and a large excess of benzylamine (50 eq. – entry
Conclusions
(
In conclusion, it has been demonstrated that palladium dimer
4 is an excellent pre-catalyst for the carbonylative coupling of
amines and aryl iodides (and bromides) giving, in some cases,
practically quantitative yields of the corresponding amide. A
particularly important feature of this complex in comparison to
most previously reported catalytic systems is that it gives high
yields within very short reactions times (under 10 min) and
under atmospheric pressure of CO. These features allowed us to
demonstrate the applicability of complex 4 for the synthesis of
1
1
1
1
1
) was the expected [carbonyl- C]N-benzylbenzamide obtained in
1
1
1
1
radiolabelled amides using CO gas. The Pd dimer complex 4
(which can be readily prepared and is air stable) proved to be a
clearly superior catalyst compared to the widely used [Pd(PPh ) ]
high radiochemical yield (70%). A small amount of C by-product
<16%) identified as [carbonyl- C]dibenzylurea, was also formed.
The radiochemical yields of [carbonyl- C]N-benzylbenzamide
1
1
(
1
1
3 4
1
1
benchmark both for our model [ CO] aminocarbonylation and
using complex 4 as a catalyst show a marked improvement
compared to other phosphine or N-heterocyclic carbene based
1
1
for the preparation of a biologically relevant [carbonyl- C]indole
amide. Importantly, this was achieved even when the amine
is present in only a slight excess – conditions that might be
important in radiotracer synthesis where the amine may not be
readily available and hence large excess quantities cannot be easily
used.
6b,16
catalysts used for this reaction.
We were then interested to
evaluate whether the carbonylative coupling reaction with 4
as a catalyst would still proceed with only a slight excess of
benzylamine (5 eq. with respect to iodobenzene – see entry 2).
Although a decreased radiochemical yield (47%) is observed for
this reaction (entry 2 Table 3), dimer 4 still proved far superior
to the benchmark [Pd(PPh
3
)
4
] (5) catalyst (9% – entry 3) at these
lower concentrations. To explore further the applicability of 4
in radiolabelling experiments with CO, the aminocarbonylative
Acknowledgements
1
1
coupling reaction of 5-iodoindole with piperidine to yield 10
EPSRC is thanked for funding (EP/G027749/1). PWM and
RV are grateful to the EPSRC for the award of a Life Sci-
ences Interface Fellowship (EP/E039278/1) and a Leadership
Fellowship (EP/H005285/1) respectively. NJL is grateful to
the Leverhulme Trust for the award of a Research Fellowship
(RF/4/RFG/2009/0493). Mr James Hall and Dr Roberta Bom-
parola are acknowledged for their assistance with GC analysis
(
which is, as indicated above, a ligand for the 5-HT2A receptor)
was investigated. As can be seen in Table 3, entry 4, this reaction
1
1
gave a high decay-corrected radiochemical yield (RCY) of the C-
labelled amide (69%) even though only 5 eq. of amine were used.
This is considerably higher than the yield obtained (22%) for the
same reaction when using the benchmark catalyst [Pd(PPh
3
4
) ].
3
502 | Org. Biomol. Chem., 2011, 9, 3499–3503
This journal is © The Royal Society of Chemistry 2011

View Article Online
Notes and references
Int. Ed., 2001, 40, 2856; (c) W. M a¨ gerlein, A. F. Indolese and M. Beller,
J. Organomet. Chem., 2002, 641, 30; (d) J. R. Martinelli, D. M. M.
Freckmann and S. L. Buchwald, Org. Lett., 2006, 8, 4843; (e) J. R.
Martinelli, D. A. Watson, D. M. M. Freckmann, T. E. Barder and S. L.
Buchwald, J. Org. Chem., 2008, 73, 7102; (f) J. McNulty, J. J. Nair and
A. Robertson, Org. Lett., 2007, 9, 4575.
1
(a) W. A. Weber, J. Clin. Oncol., 2006, 24, 3282; (b) V. L. Cropley,
M. Fujita, R. B. Innis and P. J. Nathan, Biol. Psychiatry, 2006, 59,
8
2
98; (c) K. Herholz and W. D. Heiss, Mol. Imaging Biol., 2004, 6,
39.
2
3
4
P. W. Miller, N. J. Long, R. Vilar and A. D. Gee, Angew. Chem., Int.
10 (a) M. Iizuka and Y. Kondo, Chem. Commun., 2006, 1739; (b) S. W.
Lee, K. Lee, D. Seomoon, S. Kim, H. Kim, H. Kim, E. Shim, M. Lee,
S. Lee, M. Kim and P. H. Lee, J. Org. Chem., 2004, 69, 4852.
11 (a) U. Christmann, R. Vilar, A. J. P. White and D. J. Williams, Chem.
Commun., 2004, 1294; (b) R. Vilar, D. M. P. Mingos and C. J. Cardin,
J. Chem. Soc., Dalton Trans., 1996, 4313; (c) V. Dura-Vila, D. M. P.
Mingos, R. Vilar, A. J. P. White and D. J. Williams, J. Organomet.
Chem., 2000, 600, 198; (d) U. Christmann, D. A. Pantazis, J. Benet-
Buchholz, J. E. McGrady, F. Maseras and R. Vilar, J. Am. Chem. Soc.,
2006, 128, 6376.
12 (a) T. J. Colacot, Platinum Met. Rev., 2009, 53, 183; (b) J. P. Stambuli,
R. Kuwano and J. F. Hartwig, Angew. Chem., Int. Ed., 2002, 41, 4746.
13 P. W. Miller, N. J. Long, A. J. de Mello, R. Vilar, J. Passchier and A.
Gee, Chem. Commun., 2006, 546.
14 H. Bottcher, J. Marz, H. Greiner, J. Harting, G. Bartoszyk, C. Seyfried,
C. V. Amsterdam, (Merck Patent GmbH, Germany), WO 2001007434,
2001.
15 (a) K. Kumar, A. Zapf, D. Michalik, A. Tillack, T. Heinrich, H.
Bottcher, M. Arlt and M. Beller, Org. Lett., 2004, 6, 7; (b) I. J. S.
Fairlamb, S. Grant, P. McCormack and J. Whittall, Dalton Trans.,
2007, 859.
Ed., 2008, 47, 8998.
B. Langstom, O. Itsenko and O. Rahman, J. Labelled Compd. Radio-
pharm., 2007, 50, 794.
(a) T. Kihlberg, B. Langstrom, T. Ferm, J. Eriksson, Patent number:
US2008095693, 2005; (b) J. Eriksson, O. Aberg and B. Langstrom,
Eur. J. Org. Chem., 2007, 455; (c) E. D. Hostetler and H. D. Burns,
Nucl. Med. Biol., 2002, 29, 845.
5
(a) P. W. Miller, N. J. Long, A. J. de Mello, R. Vilar, H. Audrain, D.
Bender, J. Passchier and A. Gee, Angew. Chem., Int. Ed., 2007, 46, 2875;
(
b) P. W. Miller, H. Audrain, D. Bender, A. J. deMello, A. D. Gee, N. J.
Long and R. Vilar, Chem.–Eur. J., 2011, 17, 460.
6
7
(a) H. Audrain, L. Martarello, A. Gee and D. Bender, Chem. Commun.,
2
004, 558; (b) S. Kealey, P. W. Miller, N. J. Long, C. Plisson, L.
Martarello and A. D. Gee, Chem. Commun., 2009, 3696.
(a) C. F. J. Barnard, Organometallics, 2008, 27, 5402; (b) P. W. Miller,
L. E. Jennings, A. J. deMello, A. D. Gee, N. J. Long and R. Vilar, Adv.
Synth. Catal., 2009, 351, 3260.
8
9
A. Brennfuehrer, H. Neumann and M. Beller, Angew. Chem., Int. Ed.,
2
009, 48, 4114.
(a) Joseph R. Martinelli, Thomas P. Clark, Donald A. Watson, Rachel
H. Munday and Stephen L. Buchwald, Angew. Chem., Int. Ed., 2007,
16 L. E. Jennings, S. Kealey, P. W. Miller, N. J. Long and A. D. Gee, Journal
4
6, 8460; (b) W. M a¨ gerlein, A. F. Indolese and M. Beller, Angew. Chem.,
of Labelled Compounds and Radiopharmaceuticals, 2011, 54, 135.
This journal is © The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 3499–3503 | 3503