D. Tilly et al. / Tetrahedron Letters 47 (2006) 1121–1123
1123
3. (a) Gohier, F.; Mortier, J. J. Org. Chem. 2003, 68, 2030–
2033; (b) Gohier, F.; Castanet, A.-S.; Mortier, J. Org.
Lett. 2003, 5, 1919–1922; (c) Nguyen, T. H.; Chau, N. T.
T.; Castanet, A.-S.; Nguyen, K. P. P.; Mortier, J. Org.
Lett. 2005, 7, 2445–2448.
4. (a) Tilly, T.; Samanta, S. S.; Faigl, F.; Mortier, J.
Tetrahedron Lett. 2002, 43, 8347–8350; (b) Tilly, D.;
Samanta, S. S.; De, A.; Castanet, A.-S.; Mortier, J. Org.
Lett. 2005, 7, 827–830; (c) Tilly, D.; Samanta, S. S.;
Castanet, A.-S.; De, A.; Mortier, J. Eur. J. Org. Chem., in
press.
leading to the metalated species 11. Since the CO2Et
group is a weaker binder than CO2Li and CONi-Pr2,
its acidifying effect in remote (C20) position is moderate.
However, CO2Et is highly electrophilic. The kinetic
anion 11c, which is destabilized by electronic repulsion
between the carbanion and the pair of the azine nitro-
gen, has not a long-enough lifetime to isomerize the
thermodynamically more stable (less basic) 40-pyridyl-
lithium 12c; it cyclizes instantaneously with the highly
electrophilic centre to give monolithium salt 13c. Hydro-
lysis of the latter gave azafluorenone 7 as a sole product.
Since CONi-Pr2 is a stronger director but a much weak-
er electrophile, isomerization (11b!12b) has time to
take place before cyclization (11b!13b) and azafluore-
none 8 is formed exclusively after acidic workup via
the stable tetrahedral gem-aminoalkoxide 14b. Isomeri-
zation 11b!12b most probably occurs by an intermole-
cular path.4
5. Cai, D.; Larsen, R. D.; Reider, P. J. Tetrahedron Lett.
2002, 43, 4285–4287.
6. LTMP (5.1 mmol) in THF (5 mL) was added dropwise to
a solution of 2-(pyridin-3-yl)benzoic acid (3a) (333 mg,
1.67 mmol) in dry THF (5 mL) at 0 ꢁC. After 18 h at rt,
D2O (620 lL, 31 mmol) was added. Stirring was main-
tained for 15 min, water (10 mL) was added and pH was
adjusted to 12 with aq 2 M NaOH. The aqueous layer was
washed with ethyl acetate (3 · 15 mL), and aq 2 M HCl
was added until pH reached an approximate value of 7.
The organic layer was dried (MgSO4), concentrated in
vacuo and the residue was chromatographed (cyclohex-
ane/ethyl acetate 9:1). 7 (79 mg, 0.43 mmol, 26%) and 8
(27 mg, 1.50 mmol, 9%) were isolated as yellow solids.
9H-indeno[2,1-b]pyridin-9-one (7). Mp 125–127 ꢁC (lit.
126–127 ꢁC: Mayor, C.; Wentrup, C. J. Am. Chem. Soc.
1975, 97, 7467–7480). 1H NMR (CDCl3, 400 MHz) d: 8.59
(dd, 1H, H1, J = 4.9 Hz, J = 1.0 Hz), 7.86 (dd, 1H, H3,
J = 7.4 Hz, J = 1.0 Hz), 7.73 (d, 1H, H7, J = 7.4 Hz),
7.56–7.54 (m, 2H, H4 and H5), 7.37 (m, H6), 7.34 (dd, 1H,
H2, J = 7.4 Hz, J = 4.9 Hz). 13C NMR (CDCl3, 100 MHz)
d: 191.3, 151.7, 149.1, 140.2, 138.6, 134.3, 130.9, 128.8,
126.6, 125.8, 123.4, 120.0.
In contrast to strong directors such as amides and
oxazolines, the CO2Li group activates moderately neigh-
bouring positions thus conferring maximum regioflexi-
bility in the metalation of the aromatic ring.3,4
Compound 3a displays a reactivity pattern intermediate
between those of the two previous limiting cases. Both
azafluorenones 7 and 8 are formed via the doubly
charged geminal dilithio dialkoxides 13a and 14a. Di-
anions 11a and 12a have a too short lifetime to be
trapped by D2O. Removal of H20 is rate determining10
whereas cyclization is fast and irreversible. Contrary to
what was observed for 2-biphenyl carboxylic acid,4 the
deprotonation of 3a by LTMP is site-selective (the ortho
position C3 is not lithiated) and the doubly charged
geminal dimetallo dialkoxides 13a and 14a are not meta-
lated further.
5H-indeno[1,2-c]pyridin-5-one (8). Mp 118–119 ꢁC. 1H
NMR (CDCl3, 400 MHz) d: 8.91 (s, 1H, H4), 8.71 (d, 1H,
H2, J = 4.5 Hz), 7.74 (d, 1H, H1, J = 7.4 Hz), 7.65 (d, 1H,
H5, J = 7.4 Hz), 7.6 (td, 1H, H6, J = 7.4 Hz, J = 8.4 Hz),
7.51 (m, 1H, H8), 7.4 (dd, 1H, H7, J = 7.4 Hz, J = 7.9 Hz).
HRMS, calcd for C12H7NO: 181.0523, found: 181.0520.
7. Zoltewicz, J. A.; Grahe, G.; Smith, C. L. J. Am. Chem.
Soc. 1969, 91, 5501–5505.
8. Complexation between Li and the pyridine nitrogen is
believed to play a role in the metalation only at a former
stage. The COX group and the pyridine nitrogen seem too
far to ensure double complexation. See: Gros, P.; Chop-
pin, S.; Fort, Y. J. Org. Chem. 2003, 68, 2243–2247, and
Ref. 4b.
The results reported herein by no means diminish the
value of the otherwise remarkable and meritorious work
´
of Mongin and Queguiner, but should be corrected for
reasons of mechanistic importance.
References and notes
9. Beak, P.; Meyers, A. I. Acc. Chem. Res. 1986, 19, 356–
363.
´
´
1. Rebstock, A.-S.; Mongin, A.-S.; Trecourt, F.; Queguiner,
G. Tetrahedron 2003, 59, 4973–4977.
2. Fu, J.-P.; Zhao, B.-P.; Sharp, M. J.; Snieckus, V. J. Org.
Chem. 1991, 56, 1683–1685.
10. Roberts, J. D.; Semonev, D. A.; Simmons, H. E.;
Carlsmith, L. A. J. Am. Chem. Soc. 1956, 22, 601–
611.