Ring selective lithiation of 3,5-lutidine. A new route to heterocyclic building blocks

Article abstract of DOI:10.1055/s-2002-22702

The first direct ring lithiation of 3,5-lutidine has been performed by the BuLi-LiDMAE reagent. Successful sequential metallations and electrophilic condensations have been performed giving access to new useful polysubstituted pyridines.

Full text of DOI:10.1055/s-2002-22702

628
LETTER
Ring Selective Lithiation of 3,5-Lutidine. A New Route to Heterocyclic
Building Blocks.
R
ing Selective
h
Lithiation o
i
f
3,5-L
l
utidineippe Gros, Cédric Viney, Yves Fort*
Equipe Synthèse Organique et Réactivité, UMR CNRS 7565, Université Henri Poincaré, Faculté des Sciences, Boulevard des Aiguillettes,
54506 Vandoeuvre-Lès-Nancy, France
Fax +33(383)912104; E-mail: Yves.Fort@sor.uhp-nancy.fr
Received 15 January 2002
Table 1 BuLi–LiDMAE Mediated Lithiation of 1a
Abstract: The first direct ring lithiation of 3,5-lutidine has been
Temp Conv. 2a %b
3%b
2a:3
performed by the BuLi–LiDMAE reagent. Successful sequential
metallations and electrophilic condensations have been performed
giving access to new useful polysubstituted pyridines.
Entry
BuLi–
LiDMAE (°C)
(equiv)
(%)
Key words: pyridines, lithiation, carbanions, chemoselectivity, re-
gioselectivity
1
2
3
4
5
3
3
4
6
8
0
–78
–78
–78
–78
96
97
83
12
10
6
87:13
90:10
93:7
96:4
–
86
98
92
The growing demand for polyfunctional heterocyclic
compounds encourages organic chemists to develop se-
lective processes allowing fast and clean access to target
compounds. In this context, the selective lithiation of het-
eroaromatic rings with retention of reactive substituents is
a particularly attractive method. Indeed, tedious protec-
tion-deprotection sequences are thus avoided allowing
immediate subsequent functionalisation. Our laboratory is
working actively on the development of new selective
lithiating agents. For example, we have reported that the
BuLi–Me2N(CH2)2OLi superbase1 (noted BuLi–LiD-
MAE) completely tolerated the C-Cl bond of 2-, 3- and 4-
chloropyridines while inducing a regioselective lithiation
at C-2.2 BuLi–LiDMAE also accomplished the selective
para-lithiation of 3-picoline without affecting the acidic
methyl group.3 This latter unusual selectivity prompted us
to attempt the metallation of more substituted pyridines
and we chose to investigate the functionalisation of 3,5-
lutidine 1 (Scheme 1). From our knowledge, unlike side-
chain functionalisation which has been efficiently realised
with a LDA–HMPA mixture,4 direct ring metallation of 1
has never been reported. Moreover, the sequential selec-
tive ring metallation of pyridine ring of 1 is of great syn-
thetic value since it gives access to new heterocyclic
building blocks thus providing an alternative to classical
Knoevenagel–Hantzsch sequences used for construction
of polysubstituted pyridines.5
99
94
4
100
97 (93)c
–
a Reactions performed on 2 mmoles of 1.
b GC yields.
c Isolated yield in parentheses.
From Table 1, it appeared that the metallation was direct-
ed exclusively towards the pyridine ring and no product
arising from side-chain lithiation was detected. Despite
formation of the nucleophilic addition product 3 in low
amount (6–12%), the reaction was found particularly
clean and compound 2a was generally obtained in high
yields. The 2a:3 ratio was slightly improved performing
the condensation step at –78 °C and especially when the
amount of base was increased from 3 to 8 equivalents. In
the latter case, the formation of 3 was totally inhibited.6
Although these conditions led to an addition-free reaction,
the use of 8 equivalents of basic reagent was replaced by
alternative conditions detailed in entry 3 (Table 1) to illus-
trate the scope of our new lithiation process (Scheme 2,
Table 2).7
Scheme 2
As shown, the reaction allowed the selective introduction
of various functionalities at C-2 on 3,5-lutidine in good to
very good yields. Deuterium was incorporated quantita-
tively (> 98%) as a probe for complete formation of the C-
2 lithiated intermediate after metallation (entry 1). As ex-
pected, addition product 3 was always obtained in 2–5%
yield and easily separated from the desired products by
column chromatography. Our process is synthetically
Scheme 1
Synlett 2002, No. 4, 29 03 2002. Article Identifier:
1437-2096,E;2002,0,04,0628,0630,ftx,en;D01102ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

LETTER
Ring Selective Lithiation of 3,5-Lutidine
629
valuable since metallation with BuLi–LiDMAE provides
a previously unavailable unequivocal route to useful8 het-
erocyclic building blocks 2e–h.
Table 2 Preparation of 2-Substituted 3,5-Dimethyl Pyridinesa
Entry Electrophile Product
Yield (%)b
98c
Scheme 3
1
2
3
CH3OD
2b
2c
2d
We first performed control experiments with 2e and
MeSSMe as electrophile and subsequently examplified
the reaction with other electrophiles and substrates
(Table 3).9
ClSiMe3
t-BuCHO
70
65
Generally, clean reactions were obtained and the 2,6-di-
substituted derivatives were produced in high yields.
However, notable differences were observed in the reac-
tivity of 1 and C-2 monosubstituted derivatives 2,3. In-
deed, with latter substrates, the use of 8 equivalents of
basic reagent was needed to improve conversion while the
condensation temperature had to be raised to –30 °C with
2a and 2e.
4
5
6
7
C2Cl6
CBr4
I2
2e
2f
75
88
74
65
The exclusive ring metallation obtained with 1 and deriv-
atives was in agreement with those previously obtained
with 3-picoline.3 We had already postulated an initial
deprotonation of the far more acidic side chain followed
by a lithium migration to the C-6 position favoured by
strong complexation of BuLi–LiDMAE aggregates in the
proximity of the pyridine nitrogen atom. We assume that
the same pathway will be involved with 3,5-lutidine and
its derivatives, subsequent electrophilic quenching and
prototropy giving the 2-substituted product (Scheme 4).
2g
2h
ClSnBu3
a Reactions performed on 2 mmoles of 1.
b Isolated yields after column chromatography.
c Deuterium content determined by 1H NMR.
With C-2 substituted compounds, the increase of base up
to 8 equivalents could result in deprotonation of the two
methyl groups offering additional stabilisation of the in-
termediate pyridyllithium. With compounds 2a and 2e an
intramolecular lithium complexation by heteroatom at
C-2 could occur at –78 °C partially preventing lithium mi-
Our attention then focused on the sequential lithiation of
1 in order to obtain the 2,3,5,6-tetrasubstituted derivatives
4–7 (Scheme 3).
Table 3 Lithiation of 2a and 2ea
Substrate
Cl
Base (equiv)
Electrophile
MeSSMe
MeSSMe
MeSSMe
MeSSMe
C2Cl6
Temp (°C)
–78
Product E
MeS
MeS
MeS
MeS
Cl
Yield (%)
35b
2e
2e
2e
2e
2e
2e
2a
2b
3
4
6
8
8
8
8
8
8
8
4a
4a
4a
4a
4b
4c
5
Cl
–78
42b
Cl
–78
60b
Cl
–30
83c
Cl
–30
82c
Cl
CBr4
–30
Br
85c
SMe
D
MeSSMe
MeSSMe
MeSSMe
–30
MeS
MeS
MeS
73c
–78
6
90c
Bu
–78
7
93c
a Reactions performed on 2 mmoles of 2e.
b GC yields, the remaining part was unreacted 2e.
c Isolated yields, starting material (1–2%).
Synlett 2002, No. 4, 628–630 ISSN 0936-5214 © Thieme Stuttgart · New York

630
P. Gros et al.
LETTER
Scheme 4
gration leading to incomplete reaction. The increase of the References
temperature from –78 °C to –30 °C was here assumed to
(1) (a) Gros, P.; Fort, Y.; Quéguiner, G.; Caubère, P.
Tetrahedron Lett. 1995, 36, 4791. (b) Gros, P.; Fort, Y.;
promote disruption of such a complexation allowing com-
plete lithiation at C-6 (Scheme 5). Such a potential in-
volvement of internal heteroatom–lithium interaction was
supported by reactions of 2b and 3. Indeed with these
compounds, which did not bear complexing substituents,
complete functionalisation occurred at –78 °C and tem-
perature increase was not necessary.
Caubère, P. J. Chem. Soc., Perkin Trans. 1 1997, 20, 3071.
(c) Gros, P.; Fort, Y.; Caubère, P. J. Chem. Soc., Perkin
Trans. 1 1997, 24, 3597. (d) Gros, P.; Ben Younès-Millot,
C.; Fort, Y. Tetrahedron Lett. 2000, 41, 303.
(2) (a) Choppin, S.; Gros, P.; Fort, Y. Org. Lett. 2000, 2, 803.
(b) Choppin, S.; Gros, P.; Fort, Y. Eur. J. Org. Chem. 2001,
3, 603.
(3) Mathieu, J.; Gros, P.; Fort, Y. Chem. Commun. 2000, 11,
951.
(4) Mehdi-Baradarani, M.; Joule, J. A. J. Chem. Soc., Perkin
Trans. 1 1980, 72.
(5) (a) Bhandari, A.; Gallop, M. A. Synthesis 1999, 1951.
(b) Vanden, E. Tetrahedron 1999, 55, 2687. (c)Tadesse, S.;
Bhandari, A.; Gallop, M. A. J. Comb. Chem. 1999, 1, 184.
(6) This result was in agreement with observations that we made
during metallation of 2-methoxypyridine see ref.1b
(7) Preparation of 2e is given as a typical procedure. To a
solution of 2-dimethylaminoethanol (0.8 mL; 8 mmol) in
hexane (5 mL) cooled at 0 °C, was added dropwise BuLi (16
mmol; 10 mL of a 1.6 M solution in hexanes). After 15 min,
a solution of 1 (214 mg; 2 mmol) in hexane (5 mL) was
added dropwise and the orange solution stirred for 1 h at
0 °C. After cooling at –78 °C, a solution of C2Cl6 (0.95 g; 10
mmol) in hexane (10 mL) was added dropwise. The reaction
mixture was maintained at –78 °C for 0.5 h then allowed to
warm to r.t. Hydrolysis at 0 °C with water (15 mL) was
followed by extraction with diethyl ether (20 mL) and drying
over MgSO4. After evaporation of solvents, the crude
product was purified by column chromatography using
hexane–EtOAc (70:30). 2e was obtained as an oil (212 mg;
75%). 1H NMR (CDCl3, TMS): = 2.13 (s, 3 H), 2.17 (s, 3
H), 7.23 (d, 0.8 Hz, 1 H), 7.86 (d, 0.5 Hz, 1 H). 13C NMR
(CDCl3, TMS): = 17, 18.9, 131.4, 132.0, 140.0, 146.5,
148.1. MS (EI): m/z (%) = 141(100) [M+], 106(37), 77(19).
Anal. Calcd for C7H8ClN: C, 59.38%; H, 5.69%; N, 9.89%.
Found: C, 59.52%; H, 5.48%; N, 9.72%.
Scheme 5
Finally we decided to further examine polyfunctionalisa-
tion of 1 and we attempted the C-4 lithiation of dichloro-
pyridine 4b (Scheme 6). We were pleased to observe that
the BuLi–LiDMAE reagent led to the 2,4,6-trihalogenat-
ed compound 8 in 65% yield (40% yield from 1).
Scheme 6
In summary, we have developed a new selective lithiation
process to functionalise exclusively the pyridine ring of
3,5-lutidine. Furthermore, the BuLi–LiDMAE superbase
allowed an efficient sequential metallation leading to new
polysubstituted reactive pyridines as potential building
blocks for heterocyclic chemistry. Work is now in
progress to gain more insight into the lithiation pathway to
assess the scope of this reaction.
(8) (a) Fernando, S.; Maharoof, U.; Deshayes, K.; Kinstle, T.;
Ogawa, M. J. Am. Chem. Soc. 1996, 118, 5783. (b) In this
paper, 2f was prepared from 3,5-lutidine by a two-step
procedure in 44% overall yield.
(9) The procedure was identical as described in ref.7 excepted
that 1.6 mL (16 mmol) of 2-dimethylaminoethanol and 20
mL (32 mmol) of BuLi were used and that the electrophile
(20 mmol) was added at –30 °C with 2a and 2e and at –78 °C
with 2b and 3. Data given for 4b: 1H NMR (CDCl3, TMS):
= 2.31 (s, 6 H), 7.42 (s, 1 H). 13C NMR (CDCl3, TMS):
= 18.4, 130.9, 142.2, 146.7. MS (EI) m/z (%) = 177(48) [M+
+ 1], 175(76) [M+ – 1], 140(71), 104(68), 77(88), 62(63),
51(100). Anal. Calcd for C7H8ClN: C, 47.76%; H, 4.01%; N,
7.96%. Found: C, 47.94%; H, 3.87%; N, 8.02%.
Synlett 2002, No. 4, 628–630 ISSN 0936-5214 © Thieme Stuttgart · New York