Parallel chemistry investigations of ortho-directed hydrogen isotope exchange between substituted aromatics and isotopic water: Novel catalysis by cyclooctadienyliridium(I)pentan-1,3-dionates

Article abstract of DOI:10.1016/S0040-4039(00)00244-6

Novel iridium-based catalysts which promote ortho-directed hydrogen isotope exchange between substituted aromatics and isotopic water have been identified via a combination of screening and subsequent ligand optimisation. The catalysts are more active, op

Full text of DOI:10.1016/S0040-4039(00)00244-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 2705–2708
Parallel chemistry investigations of ortho-directed
hydrogen isotope exchange between substituted aromatics
and isotopic water: novel catalysis by
cyclooctadienyliridium(I)pentan-1,3-dionates
∗
Lee P. Kingston, William J. S. Lockley, Andrew N. Mather, Edward Spink,
Stewart P. Thompson and David J. Wilkinson
Department of Medicinal Chemistry, AstraZeneca R&D Charnwood, Bakewell Rd, Loughborough, LE11 5RH, UK
Received 22 November 1999; accepted 9 February 2000
Abstract
Novel iridium-based catalysts which promote ortho-directed hydrogen isotope exchange between substituted
aromatics and isotopic water have been identified via a combination of screening and subsequent ligand optimisa-
tion. The catalysts are more active, operate at lower temperature and are applicable to a wider variety of substrates
than previously known systems. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: iridium catalysts; complexes; exchange; deuterium; labelling; metalation; diones.
Compounds labelled with tritium and deuterium are fundamental research tools for the pharmaceutical
industry. Consequently a range of methodologies have been developed for labelling organic substrates
with these isotopes. Amongst the most powerful of these approaches is ortho-directed hydrogen isotope
exchange. In this approach, substituted aromatics are induced to undergo isotopic exchange in positions
ortho to a directing substituent under the influence of a transition metal catalyst. The isotope donor can
be either isotopic hydrogen gas1 or isotopic water.3
,2
Recently, we initiated a parallel chemistry programme to discover improved catalysts for the latter
process. The discovery and optimisation of suitable catalysts by rational solid phase combinatorial
4
techniques still awaits technological developments. Hence we resorted to screening of a wide range of
salts and complexes of transition metals for the ability to promote isotopic exchange between deuterium
oxide and a selection of model substrates (Fig. 1).
∗
Corresponding author. Tel: +44 (0)1509 644367; fax: +44 (0)1509 645571; e-mail: bill.lockley@astrazeneca.com (W. J. S.
Lockley)
0
040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00244-6
tetl 16524

2
706
Fig. 1.
The substrates were selected to contain known ortho-directing groups, to yield unambiguous
HPLC/MS pseudomolecular ions, well-resolved from solvent related peaks, and to possess simple NMR
spectra.
The initial screen showed catalytic activity to reside exclusively with the Group VIII metals. Moreover,
it was most evident for salts and complexes of Ru, Rh and Ir. Subsequent iterative screening of an
5
increased range of complexes of these latter metals eventually identified CODIrAcac as the compound
displaying the best catalytic activity. This compound therefore provided a starting point for further
optimisation of activity by variations in the ligand structure.
The dionate ligand was selected as the ligand for optimisation in preference to the more labile
cyclooctadiene group. Catalysts were therefore prepared from a range of commercial and freshly
synthesised dionates via a previously described approach.9
6–8
The Table 1 shows the result of changing the terminal and central substituents on the dionate ligand.
Notable improvements in activity were seen for the fluorinated pentanedionates and for all dionates
bearing a 3-substituent.
Table 1
Comparision of active catalysts using the three model substrates

2
707
The resulting novel catalysts promote deuteriation at rates which allow complete ortho-deuteriation
in as little as 1 h at 60°C, whilst labelling under even milder conditions is possible at room temperature
over several days. In addition ortho-deuteriation is now possible for substituents which previously did
10
not function, or functioned only weakly, as directors in the original RhCl /deuterium oxide system,
3
including primary sulphonamides and indoles.
To demonstrate the simplicity and efficacy of the method a large scale deuteriation of a typical substrate
was carried out using a commercially-available catalyst of intermediate activity. Thus, 4-phenylbenzoic
acid (1, 100 mg) was heated with cyclooctadienyliridium(I)pentan-1,3-dionate (20 mg) in a mixture
of DMF (6.6 ml) and deuterium oxide (3.3 ml) at 90°C for 2 h. The resulting solution was cooled,
partitioned between ethyl acetate (30 ml) and 5% w/v aqueous sodium hydrogen carbonate solution
(10 ml). The aqueous layer was separated, acidified with dilute hydrochloric acid to pH <3, and the
precipitated product re-extracted into ethyl acetate (10 ml). After removal of the solvent under reduced
pressure, crystallisation of the resulting solid from hot methanol (2.0 ml) yielded 75 mg (74%) of [2,6-
2
2
1
H ]4-phenylbenzoic acid (m.p. 223–225°C). The H NMR (61.4 MHz in [ H ]DMSO) showed a single
2
6
resonance at 8.1 ppm (2,6-positions) with no other resonances detectable. The atom% deuterium by MS
was 97%, calculated for the two exchangeable positions.
The high ortho-regiospecificity of the isotope incorporation demonstrated above was also observed for
the other substrates studied. Such specific ortho-directed labelling is expected for a mechanism involving
a cyclic ortho-metallated intermediate. Many examples of such ortho-metallation, mostly involving the
formation of five-membered metallocycles, are known for iridium. Moreover, the ease with which late-
series transition metals undergo oxidative–addition reactions, together with their tendency to form stable
hydrides, even in aqueous systems, could help to explain the catalytic activity observed.
Acknowledgements
The authors would like to thank the Physical Sciences Group at AstraZeneca R&D Charnwood, in
particular Dr. J. M. Dixon, for support with the automated NMR analyses, the Lead Generation Chemistry
teams for support with parallel synthesis methodology, and Dr. P. A. Spencer and Miss E. Proudfoot for
help with high-throughput HPLC/MS analysis.
References
1
2
3
4
5
. Heys, J. R.; Shu, A. Y. L.; Senderoff, S. G.; Phillips, N. M. J. Lab. Comp. Radiopharm. 1993, 33, 431–438.
. Hesk, D.; Das, P. R.; Evans, B. J. Lab. Comp. Radiopharm. 1995, 36, 497–502.
. Lockley, W. J. S. Tetrahedron Lett. 1982, 23, 3819–3822.
. Crabtree, R. H. Chemtech. 1999, 29, 21–26.
. Typical screening procedure.The potential catalysts to be screened (2 mg of each) were dispensed into deep-well reaction
boxes each holding 96 polypropylene tubes and then the appropriate substrate (5 mg/tube) dissolved in DMF (400 µl/tube)
and deuterium oxide (200 µl/tube) was added. The tubes were then sealed and heated at the appropriate temperature (95°C
in the initial screens, 50°C for the more active catalysts prepared later in the programme). After an appropriate time (18
h in the initial screens, 1 h later in the programme) 20 µl aliquots from each the wells were taken, diluted 10-fold with
1
2 8
:1 DMSO/H O and analysed by HPLC/MS (5 micron C silica stationary phase using acetonitrile/aqueous ammonium
acetate gradients with positive and negative ion modes) to determine the degree of deuteriation achieved. Later in the
programme the ortho-regioselectivity of the labelling was confirmed, for the active catalysts, using automated multi-sample
NMR techniques.
6
7
. Okuro, K.; Furuune, M.; Miura, M.; Nomura, M. J. Org. Chem. 1993, 58, 7606–7607.
. Sakai, T.; Miyata, K.; Tsuboi, S.; Utaka, M. Bull. Chem. Soc. Jpn. 1989, 62, 4072–4074.

2
708
8
. Hawks, G. H.; McKillop, A.; Taylor, E. C. J. Am. Chem. Soc. 1968, 90, 2421–2422.
9
. Whitmore, B. C.; Eisenberg, R. J. Inorg. Chem. 1984, 23, 1679–1703. The catalysts were prepared using a procedure derived
from the above publication. Thus in a typical preparation, commercial (CODIrCl)
degassed diethyl ether (1.7 ml) and 1,1,1,5,5,5-hexafluoropentan-2,4-dione (40 µl) added. Potassium hydroxide solution
0.5 ml of 1.0 M) was then added and the biphasic mixture stirred for 45 min under N . Removal of the ether under reduced
pressure resulted in separation of the desired deep orange complex, which was filtered and dried under vacuum. (107 mg,
1%). Yields varied, according to the dionate used, from 60–95%.
0. Hesk, D.; Jones, J. R.; Lockley, W. J. S. J. Pharm. Sci. 1991, 80, 887–890.
2
(0.1 g) was stirred under nitrogen in
(
2
7
1