Labelling of anilines, benzylamines and some N-heterocyclics using cycloocta-1,5-dienyliridium(I) -1,1,1,5,5,5-hexafluoro-pentan-2,4-dionate and isotopic hydrogen gas in DMF or DMA

Article abstract of DOI:10.1002/jlcr.898

A wide range of anilines, benzylamines and some N-heterocyclics can be ortho-deuterated at room temperature using deuterium gas and cycloocta-1,5-dienyliridium(I)-1,1,1,5,5,5-hexafluoropentan-2,4-dionate in DMF or DMA. The method is applicable to labellin

Full text of DOI:10.1002/jlcr.898

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2005; 48: 75–84.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.898
Research Article
Labelling of anilines, benzylamines and some
N-heterocyclics using cycloocta-1,5-dienyliridium(I)
-1,1,1,5,5,5-hexafluoro-pentan-2,4-dionate and isotopic
hydrogen gas in DMF or DMA
1
2
2
2
1
R. N. Garman , M. J. Hickey , L. P. Kingston , B. McAuley , J. R. Jones ,
1,
2
W. J. S. Lockley *, A. N. Mather and D. J. Wilkinson
2
1
School of Biological and Molecular Sciences, University of Surrey, Guildford
GU2 7XH, UK
2
AstraZeneca R&D Charnwood, Bakewell Rd, Loughborough, Leics LE11 5RH, UK
Summary
A wide range of anilines, benzylamines and some N-heterocyclics can be ortho-
deuterated at room temperature using deuterium gas and cycloocta-1,5-dienyliridiu-
m(I)-1,1,1,5,5,5-hexafluoropentan-2,4-dionate in DMF or DMA. The method is
applicable to labelling with tritium. Copyright # 2004 John Wiley & Sons, Ltd.
Key Words: anilines; benzylamines; ortho-labelling; cycloocta-1,5-dienyliridium(I)-
1
,1,1,5,5,5-hexafluoropentan-2,4-dionate; deuteration; CH-activation; isotope-ex-
change; COD.Ir.F Acac
6
Introduction
The preparation of compounds labelled with isotopes of hydrogen is of great
importance in the biological and life sciences. Amongst the most powerful of
the methods used is ortho-directed exchange between an aromatic substrate
and isotopic hydrogen gas (Figure 1). The most commonly employed catalysts
1
for this process are Crabtree’s catalyst or variants thereof.
Such catalysts are essentially limited to compounds soluble in non-polar
2
solvents, or which are easily recovered from ionic liquid matrices and a
consequence of this is that it excludes many polar and ionic compounds of
interest.
If the compound to be labelled is not amenable to the above techniques,
recourse can still be made to ortho-exchange using isotopic water as the
*Correspondence to: W. J. S. Lockley, Department of Chemistry, School of Biological and Molecular
Sciences, University of Surrey, Guildford GU2 7XH, UK. E-mail: w.lockley@surrey.ac.uk
Copyright # 2004 John Wiley & Sons, Ltd.
Received 1 September 2004
Revised 8 September 2004
Accepted 9 September 2004

7
6
R. N. GARMAN ET AL.
R
R
Suitable catalyst
Solvent, *H2
H*
H*
Figure 1. Ortho-exchange with isotopic hydrogen gas
isotope donor. This method is compatible with several dipolar aprotic solvents
and, since deuterium oxide may easily be employed in high molar
concentrations, it is the preferred option for labelling with deuterium at high
isotopic abundance. The approach is also applicable to the preparation of
3
4
tritium-labelled compounds at low and high specific activities, though
tritium gas is usually preferred in the latter case, since it is easier to handle and
generally produces less radiolysis than high specific activity tritium oxide.
Recently we developed several new and efficient catalysts for this isotopic
5
water exchange and in the current paper we report the application of the most
effective of these catalysts, (cycloocta-1,5-dienyliridium(I)-1,1,1,5,5,5-hexa-
fluoropentan-2,4-dionate, 1), to the labelling of several classes of aromatic
substrates using isotopic hydrogen gas. A preliminary account of some of
6
these studies has been published.
Results and discussion
7
In a previous paper in this journal, we identified a wide selection of cycloocta-
1,5-dienyliridium(I)-2,4-pentandionates as catalysts for the ortho-labelling of a
range of aromatic substrates using isotopic water as the isotope donor. Several
of these catalysts also proved capable of promoting isotopic exchange between
deuterium gas and suitable substrates in dichloromethane.
When the solvent was changed to DMF or DMA, only three of the catalysts
showed good activity. Of these three compounds, catalyst 2 could not be
prepared routinely in high purity and was unstable upon storage. Further-
more, when compared directly across a range of substrates, 2 proved less
active than 1. Although easily prepared, catalyst 3 was less active and yielded
lower regiospecificity for the ortho-exchange process than 1. Consequently, all
further studies concentrated upon 1 alone (Figure 2).
When screened in DMA against a range of substrate classes, 1 proved most
active with anilines, benzylamines and with those N-heterocyclics which could
promote ortho-exchange into an adjacent ring (Table 1). The activity of the
catalyst with the two classes of amines, is particularly useful since these
functions are not efficient directing groups for catalysts of the Crabtree type.
It should be noted that a sub-optimal catalyst loading (25%) was used
throughout these studies to enable a good differentiation between the extent of
labelling obtained for the various substrates. Labelling at higher abundance is
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

LABELLING OF
77
R1
O
Ir
R2
O
R3
1
2
3
, R1=CF , R2=H, R3=CF
3
3
, R1=CF , R2=F, R3=CF
3
3
, R1=CH , R2=H,R3=CO Et
3
2
Figure 2. Cycloocta-1,5-dienyliridium(I) dionate catalysts active with deuterium
gas in DMF or DMA
easily achieved by the use of either higher catalyst loading or longer reaction
times, or both.
No significant labelling was seen with acids, amides or esters when the solvent
was DMA or DMF, however, the catalyst was active with a wider range of
directing groups in both acetone and acetonitrile. Interestingly, the latter is a
solvent in which the Crabtree catalyst shows little activity with any substrate.
Table 1. Deuteration of various substrates by 1 and D gas in DMA
2
Substrate
Atom%D (No. of atoms)
Labelling assigned by
1
4
4
4
3
7
-Aminoacetophenone
-Aminobenzoic acid
-Aminotoluene
-Aminoquinoline
,8-Benzoquinoline
72% (2)
80% (2)
77% (2)
30% (2)a
36% (1)
70% (2)
38% (1)
49% (2)
11% (2)
35% (2)
57% (4)
50% (4)
66% (2)
55% (2)b
42% (2)
48% (2)
55% (2)
53% (2)
69% (2)
94% (2)
60% (2)
45% (2)
34% (2)
51% (2)
69% (2)
H-NMR/MS
H-NMR/ H-NMR/MS
1
1
1
1
2
H-NMR/MS
H-NMR/MS
H-NMR/ H-NMR/MS
2
Benzylamine
Biphenyl-2-ylmethylamine
Biphenyl-3-ylmethylamine
4-Dimethylaminobenzoic acid
N,N-Dimethylbenzylamine
MS only
MS only
MS only
MS only
1
H-NMR/MS
1
1
1
4
4
3
4
4
4
,2-Diphenylethylamine
,1-Diphenylmethylamine
-(4-Fluorophenyl)ethylamine
-Amino-2-hydroxybenzoic acid
-Iodobenzylamine
-Methoxybenzylamine
-Methoxybenzylamine
-Methylaminobenzoic acid
-Methylbenzylamine
MS only
1
1
1
2
H-NMR/ H-NMR/MS
2
H-NMR/ H-NMR/MS
H-NMR/MS
MS only
1
2
H-NMR/ H-NMR/MS
1
1
1
1
1
2
H-NMR/ H-NMR/MS
H-NMR/MS
H-NMR/ H-NMR/MS
2
N-Methylbenzylamine
-Methyl-1-phenylethylamine
Naphthalen-2-ylmethylamine
H-NMR/MS
H-NMR/ H-NMR/MS
2
1
MS only
1
2
H-NMR/ H-NMR/MS
2
2
4
-Phenyl-1H-imidazole
-Phenylpyridine
-Trifluoromethylbenzylamine
1
1
2
H-NMR/ H-NMR/MS
2
H-NMR/ H-NMR/MS
Reaction conditions: the substrate (0.04 mmol) and catalyst (0.01 mmol) in DMA (250 ml) stirred under D
2
gas for 4 h at room temperature.
ca. 30% at each of positions 2 & 4.
a
b
65% at position 5, 45% at position 3.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

7
8
R. N. GARMAN ET AL.
The labelling reaction shows little sensitivity to the electronic activation or
deactivation state of the substrate, a behaviour previously noted for other
1
(a),8,10
metal-catalysed isotope exchanges of this type.
Interestingly, mono-
substitution on nitrogen is well tolerated whilst di-substitution results in some
inhibition of labelling. Such inhibition has been observed in a number of
1
g,3a,9
analogous homogeneous systems,
to display this behaviour, at least for N-substituted benzamides.
though the Crabtree catalyst seems not
1
(e),10
Some
evidence of steric inhibition of labelling was observed in the case of 4-amino-2-
hydroxybenzoic acid, whilst additional examples are also included in Table 3.
Unlike the Crabtree catalyst, 1 did not catalyse the exchange between
deuterium gas and the aldehyde proton of DMF and hence could be used
effectively in this solvent. However, in common with the Crabtree catalyst, it
did yield reduction of double bonds. Nitro groups were also reduced and in
this case, the corresponding ortho-deuterated anilines were recovered.
Aromatic bromides and iodides proved stable to the exchange conditions.
The nature of the effective catalytic species is uncertain. The reaction often
shows an induction period (Figure 3), which varies with the substrate,
presumably whilst the active catalytic species is formed.
Deuteration vsTime
2
4
4
-Aminobenzoic acid
-Methoxybenzylamine
1
0
.5
1
.5
0
0
3
10
15
20
30
60
90
120 180
Time (minute)
Figure 3. Catalyst induction demonstrated with two substrates
The delay could represent the removal of the COD ligand by hydrogenation
or by substrate substitution. Figure 4 shows that the catalyst can be activated
by treatment with hydrogen gas, a useful property for the efficient use of the
tritium isotope. An additional example is provided by the deuteration/
tritiation of 4-aminoacetophenone described later.
Since iridium metal is precipitated during the reaction with some substrates,
the possibility exists that the reaction is catalysed by a heterogeneous process.
1
1
However, when the effect of added liquid mercury on the labelling of four
substrates (7,8-benzoquinoline, 2-phenylpyridine, 1-methyl-1-phenylethyla-
mine and 4-trifluoromethylbenzylamine) was examined, no definitive
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

LABELLING OF
79
Hydrogen then deuterium
Deuterium only
30
25
20
15
10
5
0
0
15
60
75
120
180
210
Minute
Figure 4. Comparison of 7,8-benzoquinoline deuteration with and without 1 h
pre-incubation with hydrogen gas
conclusions could be drawn. The first two substrates showed no significant
effect of mercury whilst the third showed a 40% increase and the last, a 58%
decrease in the extent of labelling under the non-equilibrium conditions used.
Further studies to investigate the catalytic process are shown in Table 2.
Here the catalytic activity of various iridium metal preparations was
examined using three representative substrates. None of the metal prepara-
tions yielded significant labelling in comparison with 1, though this does not
eliminate the possibility of catalysis via a species on the reductive pathway to
iridium metal, such as an iridium cluster or colloid. The differentiation
between homogeneous and heterogeneous catalysis has always proved
1
2
problematic as excellent recent reviews make clear.
Several other Group VIII metal blacks also proved inactive. Although the
analogous cycloocta-1,5-dienylrhodium(I)-1,1,1,5,5,5-hexafluoropentanedio-
nate did show some activity with the 1-methyl-1-phenylethylamine substrate
(67% ortho-D), it showed no activity with several other anilines and
benzylamines.
Table 2. Labelling vs catalyst employed for three substrates (as % D2)
Catalyst
4-Aminobenzoic
acid
4-Amino-
acetophenone
1-Methyl-1-
phenylethylamine
CODIrF Acac, 1
Iridium from CODIrF
Ir black (Commercial)
6
91
92
53
52
53
Decomposed
52
52
52
}
68
6
Acac/D
2
53
52
}
}
}
}
}
52
54
52
52
}
}
}
Iridium from IrO2/NaBH
Palladium from PdO/NaBH
Platinum from PtO /NaBH
4
)
4
)
2
4
)
Rhodium from RhCl /NaBH )
3
Ruthenium from RuCl
Crabtree’s catalyst
4
3
/NaBH
4
)
}
}
}Indicates that the experiment was not carried out.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

8
0
R. N. GARMAN ET AL.
All the above studies were carried out with deuterium gas, however the most
frequent use of the method is likely to be with tritium. Hence a tracer study,
employing 4-aminoacetophenone as substrate, was carried out to investigate
the use of this isotope. In this study, the reaction was initiated with deuterium
gas, to demonstrate the isotope-sparing option shown in Figure 4, and then a
tracer amount of tritium gas added and the reaction continued. The product of
1
3
the reaction was then studied by H and H-NMR. The results are shown in
Figure 5.
Figure 5. NMR spectra of the product from deuteration/tritiation of 4-
1
aminoacetophenone catalysed by 1 Top spectrum: H-NMR of the product in
CDCl showing reduction in the proton intensity at d 6.65 ppm and partial
3
collapse of the 2 and 6 doublet resonance at d 7.78 ppm to a singlet, confirming
deuteration at the 3 and 5 positions. Middle spectrum: H-NMR spectrum of the
3
3
same sample showing H-labelling at the 3 and 5 positions. Bottom spectrum: as
above but showing full range and absence of labelling at the methyl group.
The results presented in this paper show a quite different selectivity for
directing groups when the same catalyst is employed with a D O or D isotope
2
2
donor. This raises the possibility of controlling the regiochemistry of labelling
for some substrates by selection of the donor. Table 3 shows the control which
can be exercised with various aminobenzoic acids. Moreover, given that the
acetophenone grouping is an excellent director for catalysts of the Crabtree
type, the results with the aminoacetophenones bodes well for the control of
labelling regioselectivity for these substrates also.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

LABELLING OF
81
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

8
2
R. N. GARMAN ET AL.
Experimental
Di-m-chlorobis(1,2,5,6-Z)-1,5-cyclooctadiene]diiridium (Fluka item 14691),
deuterium oxide (99.8%D), DMF and DMA (water content 50.03 and
0
.05%, respectively) were obtained from Sigma-Aldrich (Poole, Dorset, UK).
Deuterium gas (99 atom%) was purchased from CK Gas Products Ltd (Hook,
Hampshire, UK) and tritium gas from RC Tritec (Teufen, Switzerland). All
other solvents and reagents were obtained from recognised chemical suppliers,
including the above, and were used as received.
Cycloocta-1,5-dienyliridium(I)-1,1,1,5,5,5-hexafluoropentan-2,4-dionate (1)
Commercial [di-m-chlorobis(1,2,5,6-Z)-1,5-cyclooctadiene]diiridium (300 mg) is
stirred under nitrogen in degassed ether (6 ml) at which point it is only partly
dissolved. Commercial 1,1,1,5,5,5-hexafluoro-2,4-pentanedione (0.3 ml) in
ether (0.6 ml) is then added via a syringe and the reaction stirred for 10 min.
Sodium hydroxide solution (1 M, 1.25 ml) is added, dropwise via a syringe.
During this phase the remaining crystals of precursor dissolve and the mixture
becomes a deep claret colour. Water (3 ml) is added, again via a syringe, and
the biphasic mixture stirred under nitrogen for a further 20 min. The ether
layer is evaporated under a slow stream of nitrogen leaving suspended clumps
of fine claret-coloured crystals. The aqueous supernatant is decanted and the
crystals washed twice more by decantation with portions of water (2 ml).
Finally, the crystals are filtered and dried over silicagel overnight to
leave cycloocta-1,5-dienyliridium(I)-1,1,1,5,5,5-hexafluoropentan-2,4-dionate
1
5
1
(
412.5 mg, 91%) as a fine free-flowing stable red-purple solid. H-NMR: d
(
CDCl ) 1.75 (4 H, q, J=7.5 Hz), 2.30 (4 H, m), 4.31 (4 H, s), 6.32 (1 H, s)
3
1
3
p.p.m. C-NMR: d (CDCl ) 31.1, 62.6, 93.1, 118.5 (quartet), 174.3 (quartet)
3
p.p.m. MS (EI mode): m/z 508/506 (mol. ion), 301-292 (cluster) a.m.u. Found
C 30.55%, H 2.91%, required C 30.67%, H 2.91%.
Typical deuteration protocol for anilines and benzylamines
The substrate amine (0.04 mmol) and catalyst (0.01 mmol) were dissolved in
DMF or DMA (250 ml) and stirred under deuterium gas for 3–4 h at room
temperature. The labelled substrate was isolated by a diethyl ether/
hydrochloric acid (2 M) partition, followed by basification of the aqueous
extract with NaOH solution (1 M), extraction with diethyl ether, drying with
anhydrous magnesium sulphate and removal of the solvent under a stream of
dry carbon dioxide free nitrogen. At this point the labelled substrates were
1
2
usually sufficiently pure for reliable analysis by H- and H-NMR and LC-MS.
2
3
[
3,5- H, 3,5- H]-4-Aminoacetophenone
Cycloocta-1,5-dienyliridium(I)-1,1,1,5,5,5-hexafluoropentan-2,4-dionate (1,
mg) and 4-aminoacetophenone (5.4 mg) were dissolved in N,N-dimethylace-
5
tamide (0.3 ml) and stirred under an atmosphere of deuterium gas for 1 h. A
tracer quantity of tritium gas was then admitted and the reaction allowed to
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

LABELLING OF
83
proceed for a further 3 h. The reaction was partitioned between aqueous
sodium hydrogen carbonate solution (5% v/v, 3 ml) and chloroform (6 ml) and
the chloroform layer separated and evaporated under a stream of dry nitrogen
2
3
1
to yield [3,5- H,3,5- H]-4-aminoacetophenone. H-NMR of the product
2
showed it to be >90% pure with an abundance of >60% H at the 3 and
3
5
6
positions. The H-NMR of the product showed the expected resonance at d
.7 ppm for the 3- and 5-positions. Moreover, no labelling was detectable at
the positions of the acetyl methyl group or the 2- and 6-positions. An
unidentified impurity was observable at the 510% level in both the triton and
proton NMR spectra (see text).
Acknowledgements
The authors would like to thank the Physical Sciences Group and Mr. J.
Gardiner of AstraZeneca R&D Charnwood for microanalyses and the LC
analysis of labelled reaction mixtures. We are grateful to Mr. J. Bloxsidge and
Mr. R. Chaudry, both of the University of Surrey, for help with the
spectroscopic analysis and characterisation of both catalysts and labelled
substrates.
References
1
. See for example, but not exclusively (a) Heys JR. J Chem Soc Chem Commun
992; 680; (b) Heys JR, Shu AYL, Senderoff SG, Phillips NM. J Label Compd
1
Radiopharm 1993; 33: 431–438; (c) Shu AYL, Heys JR. J Label Compd
Radiopharm 1994; 34: 587; (d) Ellames GJ, Gibson JS, Herbert JM, Kerr WJ,
McNeil AH. Tetrahedron Lett 2001; 42: 6413–6416; (e) Ellames GJ, Gibson JS,
Herbert JM, McNeil AH. Tetrahedron 2001; 57: 9487–9497; (f) Salter R,
Chappelle MR, Morgan A, Moenius T, Ackerman P, Studer M, Spindler F. Synth
Appl Isot Label Compd Proceedings of the International Symposium, vol. 7, Pleiss
U, Voges R (eds). Wiley: New York, 2001; 63–67; (g) Shu AYL, Chen W, Heys
JR. J Organomet Chem 1996; 524: 87–93; (h) Cross PWC, Ellames GJ, Gibson JS,
Herbert JM, Kerr WJ, McNeil AH, Mathers TW. Tetrahedron 2003; 59:
3
349–3358; (i) Hesk D, Das PR, Evans B. J Label Compd Radiopharm 1995;
6: 497–502; (j) Valsborg JS, Sorensen L, Foged C. Synth Appl Isot Label Compd
3
Proceedings of the International Symposium, vol. 7, Pleiss U, Voges R (eds). Wiley:
New York, 72–75; (k) Cross PWC, Ellames GJ, Gibson JS, Herbert JM, Kerr,
WJ, McNeil AH, Mathers TW. Tetrahedron 2003; 59: 3349–3358; (l) Ellames GJ,
Gibson JS, Herbert JM, Kerr WJ, McNeill AH. J Label Compd Radiopharm 2004;
4
7: 1–10.
2
3
. Salter R, Bosser I. J Label Compd Radiopharm 2003; 46: 489–498.
. (a) Lockley WJS. J Chem Res (S) 1985; 178–179, J Chem Res (M) 1985;
1
976–1986; (b) Wilkinson DJ. M Phil Thesis, University of Surrey, 1995; (c) Hesk
D, Jones JR, Lockley WJS, Wilkinson DJ. J Label Compd Radiopharm 1990; 28:
309–1320; (d) Hesk D, Jones JR, Lockley WJS. J Pharm Sci 1991; 80: 887–890;
1
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84

8
4
R. N. GARMAN ET AL.
(e) Oohashi K, Seki T. J Radioan Nucl Chem Lett 1994; 187: 303–311; (f) Carroll
L, Jones JR, Lockley WJS. Synth Appl Isot Label Compd Proceedings of the
International Symposium, Muccino RR (ed.). Elsevier: Amsterdam, 1986;
3
89–393.
. (a) Wilkinson DJ, Lockley WJS. J Label Compd Radiopharm 1985; 22: 883–892;
b) Lockley WJS, Wilkinson DJ. J Label Compd Radiopharm 1984; 21: 363–373.
4
5
6
7
(
. Kingston LP, Lockley WJS, Mather AN, Spink E, Thompson SP, Wilkinson DJ,
Tetrahedron Lett 2000; 41: 2705–2708.
. Hickey MJ, Jones JR, Kingston LP, Lockley WJS, Mather AN, McAuley BM,
Wilkinson DJ. Tetrahedron Lett 2003; 44: 3959–3961.
. Hickey MJ, Jones JR, Kingston LP, Lockley WJS, Mather AN, McAuley BM,
Spink E, Thompson SP, Wilkinson DJ. J Label Compd Radiopharm 2003; 46:
1
–14.
8
. (a) Hesk D. Ph.D. Thesis, University of Surrey, 1988; (b) Lockley WJS, Hesk D,
Jones JR, Nolan K. Synth Appl Isot Label Compd Proceedings of the International
Symposium, vol. 4, Buncel E, Kabalka GW (eds). Elsevier BV: Amsterdam, 1991;
6
00–603; (c) Garnett JL, Hodges RJ, Kenyon RS, Long MA. J Chem Soc Perkin
979; 2: 885–890.
1
9
. Lockley WJS. J Label Compd Radiopharm 1984; 21: 45–57.
1
0. Valsborg JS, Sorensen L, Foged C. Synth Appl Isotop Label Compd Proceedings of
the International Symposium, vol. 7, Pleiss U, Voges R (eds), Wiley: New York,
2
001; 72–75.
1
1. Anton DR, Crabtree RH. Organometallics 1983; 2: 855–859.
1
2. Widegren JA, Finke RG. J Mol Catal A: Chem 2003; 198: 317–341; 2003; 191:
1
87–207.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2005; 48: 75–84