B. Roberts et al. / Tetrahedron Letters 52 (2011) 3793–3796
3795
Mo(CO)6
Cl.NEt4
Cl-Mo(CO)5.NEt4
I
D
C
I
O
[Mo]
O
[Mo]
R
R
X
N
R
N
N
G
N
H
X
reductive
elimination
CO insertion
N
H
N
H
X
[Mo] = Mo(CO)5
I
or intermediate F
X
[Mo] = Mo(CO)4
R
reductive
elimination
N
H
N
H
oxidative
addition
X = O , NR1
[Mo] = Mo(CO)4
[Mo]
I
[Mo]
R
N
O
[Mo]
I
O
I
R
[Mo]
oxidative
addition
HN
I
Cl
R
N
R
N
N
H
X
N
H
X
CO insertion
N
H
X
N
X
[Mo] = Mo(CO)5
[Mo] = Mo(CO)5
H
E
F
B
A
Scheme 2. Speculative mechanism for the synthesis of quinazolin-4(3H)-ones via molybdenum-mediated cyclocarbonylation.
formed with sub-stoichiometric amounts of Mo(CO)5ClÁNEt4
although the yields of product are reduced and longer reaction
times are required for complete consumption of the starting mate-
rials (Table 2, entries 12 and 13). Further work to identify the opti-
mal ratio of [Mo]-complex and reaction concentration is on-going.
This methodology was applicable to aryl bromides, although the
yields were reduced and longer reaction times were required
(Table 2, entries 14 and 15). In all these reactions we saw no evi-
dence of intramolecular cyclisation via the urea oxygen to give
4H-benzo[d][1,3]oxazin-4-one products which we have observed
when performing this urea cyclisation under standard palladium-
catalysed carbonylation conditions using carbon monoxide gas.17
In both of the outlined reactions, we speculate that urea and
guanidine decomposition is a competing reaction at high tempera-
tures and the lower yields obtained using Mo(CO)6 compared to
the Mo(CO)5ClÁNEt4 complex are due to the fact that the chloride
ligand in Mo(CO)5ClÁNEt4 is more readily displaced by the urea or
guanidine nitrogen than a CO ligand in Mo(CO)6 (see proposed
mechanism discussed below).
The mechanism of these reactions is unclear at present, but one
can speculate on possible pathways (Scheme 2).18 In our previous
work we demonstrated that Et4NÁCl readily displaces a CO ligand
from Mo(CO)6 to give Mo(CO)5ClÁNEt4 and this complex reacts
readily with nitrogen nucleophiles.10 Thus we postulate that inter-
mediate B could be generated in the reaction. This could then
undergo oxidative-addition followed by CO insertion to give D.
Reductive elimination furnishes product G. Alternatively interme-
diate B could undergo CO insertion, to give E, followed by oxida-
tive-addition to give intermediate F. Once again reductive
elimination would give the product.
The methods are ideally suited for parallel synthesis and automa-
tion often required in modern drug discovery. Further develop-
ment of the sub-stoichiometric versions of these reactions is in
progress.
Acknowledgements
We are grateful to Hema Pancholi of AstraZeneca Charnwood
Chemistry Department for assistance with MS studies. We also
thank Tony Shephard of AstraZeneca Charnwood PARD Depart-
ment for palladium analyses.
Supplementary data
Supplementary data associated with this article can be found, in
References and notes
1. Armarego, W. L. F. In Quinazolines; Brown, D. J., Ed.; John Wiley & Sons Ltd.:
London, 1967; (b) Ellis, G. P. Synthesis of Fused Heterocycles; John Wiley & Sons,
Ltd.: Chichester, 1987. Vol. 47.
2. Pendergast, W.; Johnson, J. V.; Dickerson, S. H.; Dev, I. K.; Duch, D. S.; Ferone, R.;
Hall, W. R.; Humphrey, J.; Kelly, J. M.; Wilson, D. C. J. Med. Chem. 1993, 36, 2279.
3. Alargarsamy, V.; Dhanabal, K.; Parthiban, P.; Anjana, G.; Deepa, G.; Murugesan,
B.; Rajkumar, S.; Beev, A. J. J. Pharm. Pharmacol. 2007, 59, 669.
4. Chao, Q.; Deng, L.; Shih, H.; Leoni, L. M.; Genini, D.; Carson, D. A.; Cottam, H. B. J.
Med. Chem. 1999, 42, 3860.
5. Chern, J.-W.; Tao, P.-L.; Wang, K.-C.; Gutcait, A.; Liu, S.-W.; Yen, M.-H.; Chien,
S.-L.; Rong, J.-K. J. Med. Chem. 1998, 41, 3128.
6. Zappala, M.; Grasso, S.; Micale, N.; Zuccala, G.; Menniti, F. S.; Ferreri, G.; Sarroc,
G. D.; Michelid, C. D. Bioorg. Med. Chem. Lett. 2003, 13, 4427.
7. (a) Daidone, G.; Plescia, S.; Raffa, D.; Schillaci, D.; Maggio, B.; Benetollo, F.;
Bombieri, G. Heterocycles 1996, 43, 2385; (b) Wang, H.; Genesan, A. J. Org. Chem.
1998, 63, 2432; (c) He, F.; Snider, B. B. J. Org. Chem. 1999, 64, 1397.
8. For examples, see: (a) Alper, H.; Larksarp, C. J. Org. Chem. 2000, 65, 2773; (b)
Alper, H.; Zeng, F. Org. Lett. 2010, 12, 1188. and references cited therein.
9. Liang, Y.; Ding, M.-W.; Liu, Z.-J.; Liu, X.-P. Synth. Commun. 2003, 33, 2843.
10. Roberts, B.; Liptrot, D.; Alcaraz, L.; Luker, T.; Stocks, M. J. Org. Lett. 2010, 12,
4280.
It should be noted that we cannot rule out direct oxidative-
addition to the aryl halide of a molybdenum species followed by
CO insertion to generate an acyl molybdenum species in an analo-
gous manner to palladium-catalysed cyclocarbonylation.8 How-
ever, when we heated
a 1:1 mixture of iodobenzene and
11. Ren, W.; Yamane, M. J. Org. Chem. 2010, 75, 8410.
12. Volonterio, A.; Zanda, M. J. Org. Chem. 2008, 73, 7486.
Mo(CO)6 in 1,4-dioxane in a microwave at 150 °C for 1 h we were
unable to detect any reaction and iodobenzene remained intact.
In summary, we have developed an efficient and practical strat-
egy for the synthesis of quinazolin-4(3H)-ones via molybdenum-
mediated cyclocarbonylation. These methods allow access to a
wide range of quinazolin-4(3H)-ones in reasonable yield without
the need for gaseous carbon monoxide and palladium catalysts.
13. All reactions were performed in a CEM Discover single mode microwave
reactor equipped with a 300 W power source set either to 100 or 150 W. The
reactions were performed in a CEM 10 mL microwave reaction vial and new
rare earth Teflon coated magnetic stir bars were used to stir the reaction
mixtures. All temperature measurements were performed with an IR probe.
Caution: Mo(CO)6 and its derivatives are toxic and these sealed reactions
should only be carried out using specialised microwave equipment.