Toward a better understanding on the mechanism of ortholithiation. Tuning of selectivities in the metalation of meta-anisic acid by an appropriate choice of base

Article abstract of DOI:10.1021/ol050761c

(Chemical Equation Presented) If employed in THF at 0°C, LTMP metalates meta-anisic acid at the doubly activated position. In contrast, n-BuLi/t-BuOK deprotonates position C-4 preferentially at low temperature. Functionalization at C-6 requires protection of the C-2 site beforehand. As a result of these findings, a new mechanism is proposed for the heteroatom-directed ortholithiation of aromatic compounds.

Full text of DOI:10.1021/ol050761c

ORGANIC
LETTERS
2005
Vol. 7, No. 12
2445-2448
Toward a Better Understanding on the
Mechanism of Ortholithiation. Tuning of
Selectivities in the Metalation of
meta-Anisic Acid by an Appropriate
Choice of Base
Thi-Huu Nguyen,† Nguyet Trang Thanh Chau,† Anne-Sophie Castanet,†
Kim Phi Phung Nguyen,‡ and Jacques Mortier*,†
Unite´ de Chimie Organique Mole´culaire et Macromole´culaire (UMR 6011),
Faculte´ des Sciences, UniVersite´ du Maine and CNRS, AVenue OliVier Messiaen,
72085 Le Mans Cedex 9, France, and Laboratoire de Chimie Organique,
EÄ cole des Sciences Naturelles, UniVersity of Ho Chi Minh City,
283/2 Nguyen Van Cu. Arrondissement 5, Ho Chi Minh City, Vietnam
jacques.mortier@uniV-lemans.fr
Received April 8, 2005
ABSTRACT
If employed in THF at 0 °C, LTMP metalates meta-anisic acid at the doubly activated position. In contrast, n-BuLi/t-BuOK deprotonates position
C-4 preferentially at low temperature. Functionalization at C-6 requires protection of the C-2 site beforehand. As a result of these findings, a
new mechanism is proposed for the heteroatom-directed ortholithiation of aromatic compounds.
There are a number of studies dealing with lithiations of
arenes carrying 1,2- and 1,4-interrelated directed metalation
groups (DMGs). The most dramatic effects are observed in
cases where the selection of either of two ortho positions
may be controlled by the metalating conditions.1 Few
examples of this kind of regioselective control for aromatic
metalation are known. The best examples are perhaps the
metalation of 2- and 4-fluoroanisole (1 and 2, respectively),
which occurs primarily at the oxygen-adjacent position
with n-BuLi and ortho to the fluorine with n-BuLi/t-BuOK
or n-BuLi/N,N,N′,N′′,N′′-pentamethyldiethylenetriamine
(PMDTA),2 and the metalation of 4-fluoro- and 4-chloroben-
zoic acids (3), which with s-BuLi, s-BuLi/ TMEDA, or
t-BuLi results in exclusive reaction ortho to the carboxylate
but with LTMP selects the C-3 positions.3
Literature furnishes much less information regarding
lithiations of 1,3-interrelated systems, which offer selection
of either of three possible ortho substitutions (C-2, C-4, and
C-6). It is usually considered as a rule of thumb that the
two DMGs function in concert to direct introduction of the
metal between them. Even though this is true in most cases,
there are a few exceptions. Thus, while meta-methoxy phenyl
† Universite´ du Maine and CNRS.
‡ University of Ho Chi Minh City.
(1) Reviews: (a) Hartung, C. G.; Snieckus, V. Modern Arene Chemistry;
Astruc, D., Ed.; Wiley-VCH: New York, 2002; p 330. (b) Schlosser, M.
Angew. Chem., Int. Ed. 2005, 44, 376.
(2) Maggi, R.; Schlosser, M. J. Org. Chem. 1996, 61, 5430.
(3) Gohier, F.; Castanet, A.-S.; Mortier, J. J. Org. Chem. 2005, 70, 1501.
10.1021/ol050761c CCC: $30.25
© 2005 American Chemical Society
Published on Web 05/17/2005

1
oxazoline (4) and secondary and tertiary benzamides (5) are
deprotonated at the C-2 position by n-BuLi or s-BuLi/
TMEDA,4 Meyers has described the unusual behavior of
R-ethoxyvinyllithium-HMPA at -83 °C, which metalates
4 and 5 exclusively at C-4.5
proton in the H NMR spectra.7 The products of lithiation
ortho to the methoxy group and para to the carboxylate
exhibited a doublet for the H-5 proton.
After considerable experiment beforehand, we found that
treatment of 6 with lithium 2,2,6,6-tetramethylpiperidide
(LTMP, 5 equiv) in THF for 1.5-2 h at 0 °C provided
quantitatively the dianion 7 (M ) Li). Quenching with D2O
under external quench (EQ) conditions afforded a 92% yield
of deuterated product 10 in which >95% of the deuterium
was incorporated in the C-2 position (entry 1). Treatment of
6 with LTMP (3 equiv) and Me3SiCl at -78 °C under in
situ quench (ISQ) conditions,8 followed by warming to room
temperature provided a 75% yield of 2-trimethylsilyl-3-meth-
oxybenzoic acid (11) after chromatography (entry 2). In
addition, one other fraction was obtained in 7% yield. This
was 3-methoxy-2,6-bis(trimethylsilyl)benzoic acid (15), re-
sulting from the partial deprotonation of lithium salt of the
primary product 11 by the excess of base at the carbon C-6.9
The acid 15 was formed predominantly (63%, entry 3) when
LTMP was used in large excess (5 equiv) at 0 °C (ISQ con-
ditions). A sequential process involving a rapid intraaggregate
lithiation through a quasi dianion complex “QUADAC”
presumably proceeds during the formation of 15.3
This letter discusses features of the CO2H(Li) group that
make it unique among ortho-directing groups.6 We have been
able to identify conditions that ensure good to excellent
selectivities when CO2H(Li) and MeO groups are located
meta to one another. meta-Anisic acid (6) (Scheme 1) in THF
Scheme 1. Site-Selective Metalation of meta-Methoxybenzoic
6 by Strong Bases
s-BuLi/TMEDA was found to be more reactive than
LTMP. Although the selectivity obtained was not quite so
favorable, the direction of the selectivity was maintained.
After quenching with iodomethane at -78 °C, a mixture of
three isomeric acids 12-14 was obtained in a 63:17:20 ratio
with 70% overall yield (entry 4). The major C-2 isomer 12
was readily separated by fractional crystallization (31%).
Whereas lithium 2-lithio-3-methoxybenzoate monomer (7)
prepared by metalation with LTMP (according to entry 1 of
Table 1) was found to be indefinitely stable in the interval
of temperature 0-60 °C,9 it degraded somewhat at 0 °C
when generated from s-BuLi/TMEDA under the conditions
reported in entry 5.10
When meta-anisic acid (6) was treated with n-BuLi/
t-BuOK (1:1 ratio, 4 equiv) in THF (-78 f -50 °C) (entry
6), the dianion 8 arising from metalation in C-4 formed
preferentially. Quenching with iodomethane followed by
hydrolysis with dilute HCl resulted in a 84% yield of the
was subjected to a series of strong bases under the conditions
depicted in Table 1. All compounds obtained as the result
of lithiation at the position C-2, ortho to both substituents,
showed a characteristic triplet corresponding to the H-5
Table 1. Metalations of meta-Anisic Acid (6)
regioselectivity (%)
entry
conditionsa
yield (%)b
92
90 (75* [11])
29
C-2
C-4
C-6
other
7* [15]
1
2
3
4
(1) LTMP (5 equiv), 0 °C; (2) D2O, 0 °C
LTMP (3 equiv), Me3SiCl, -78 °C f rt
LTMP (5 equiv), Me3SiCl, 0 °C
(1) s-BuLi/TMEDA (2.2 equiv), -78 °C;
(2) MeI, -78 °C
100 [10]
100 [11]
100 [11]
63 [12]
0
0
0
0
0
0
63 (51*) [15]
70 (31* [12])
17 [13]
20 [14]
5
6
(1) s-BuLi/TMEDA (2.2 equiv), -78 f 0 °C;
(2) MeI, 0 °C
(1) n-BuLi/t-BuOK (4 equiv), -78 f -50 °C;
(2) MeI, -50 °C
34
79 [12]
9 [12]
15 [13]
80 [13]
14 [14]
11 [14]
c
84 (59* [13])
a External quench (EQ) technique for D2O and MeI. In situ quench (ISQ) method for Me3SiCl. Hydrolysis was carried out at rt. b Overall yield (%).
Isolated yields (recrystallized or chromatographed) are followed by an asterisk (*). c Degradation products.
2446
Org. Lett., Vol. 7, No. 12, 2005

acids 12, 13, and 14 in a ratio of 9:80:11, respectively.11
The major C-4 isomer 13 was readily isolated by fractional
crystallization in ethyl acetate (59%).
Table 2. Regioselective Preparation of 2-, 4-, and
6-Substituted 3-Methoxybenzoic Acidsa
By employing the optimized conditions found, we were
able to synthesize a variety of 2- (12, 16a-f) and 4-substi-
tuted (13, 17a-d,f) 3-methoxybenzoic acids (Scheme 2). The
route
EX
MeI
C2Cl6
C2Br2Cl4
I2
Me2S2
DMF
PhCHO
E
i
ii
iii
iii + iv
Me
Cl
Br
I
50
47
60
53
46
27b
65c
59
39
65
20
51
57
57
66
53
61
54
55
63
50
61
Scheme 2
MeS
CHO
PhCH(OH)
54
20c
a Isolated yields (recrystallized or chromatographed). For general pro-
cedures, see Supporting Information. b Product cyclized into hydroxy-
phthalide 21. c Products cyclized into lactones 22 and 23.
It does not seem feasible to direct metalation regioselec-
tively at the C-6 carbon to give the dianion 9. Then, to
prepare 6-substituted 3-methoxybenzoic acids, one has to
block the C-2 site by introducing a trimethylsilyl group to
give 11 (see entry 2 of Table 1), lithiate again, deliver the
electrophile, and remove the protective group.12
The C-6 position of 11 was lithiated regiospecifically by
the s-BuLi/TMEDA complex (there was no trace of the C-4
isomer 3-methoxy-2,4-bis(trimethylsilyl)benzoic acid) and
treated with electrophiles to give acids 18 and 19a-d,f after
careful hydrolysis with 2 M HCl. These acids were easily
deprotected by 6 M HCl to afford 14 and 20a-d. Condensa-
tion with DMF followed by acid-catalyzed cyclization gave
the hydroxyphthalide 21 via the ortho formyl product 16e.
Clean hydroxy alkylation was achieved with benzaldehyde
to give 16f and 19f, which were directly transformed into
lactones 22 and 23.
The mechanism of heteroatom-directed ortholithiation of
aromatics still inspires very active debate and ongoing
controversy in the scientific community. An early hypothesis,
advanced by Beak and Meyers to account for regioselective
lithiation of aromatic compounds bearing a Lewis basic
heteroatom, is that the lithium coordinates with the lone pairs
of the heteroatom of the directing group to form a prelithia-
tion complex (CIPE effect).13
An alternative hypothesis claims that heteroatom-directed
ortholithiation of aromatics should be strictly considered as
a kinetically controlled reaction, i.e., a one-step reaction that
Schleyer has named “kinetically enhanced metalation”.14 For
this author, “this is not precomplexation that is important
but the existence of a stabilizing metal-substituent interaction
at the rate-limiting transition structures”.14 In this model,
there is not an intermediate with a finite lifetime on the
reaction pathway prior to proton transfer.
results with diverse electrophiles are summarized in Table
2. Although yields in this preliminary study are only
moderate, they are, even at this stage, usable, since no
protection and deprotection steps of the reactive carboxylic
acid group are needed.
(4) (a) De Silva, S. O.; Reed, J. N.; Snieckus, V. Tetrahedron Lett. 1978,
19, 1823. (b) Beak, P.; Brown, R. A. J. Org. Chem. 1981, 46, 34.
(5) Shimano, M.; Meyers, A. I. J. Am. Chem. Soc. 1994, 116, 10815.
(6) For recent work on ortholithiation reactions of unprotected benzoic
acid derivatives, see: (a) Gohier, F.; Castanet, A.-S.; Mortier, J. Org. Lett.
2003, 5, 1919. (b) Gohier, F.; Mortier, J. J. Org. Chem. 2003, 68, 2030. (c)
Tilly, D.; Samanta, S. S.; De, A.; Castanet, A.-S.; Mortier, J. Org. Lett.
2005, 7, 827. (d) Tilly, D.; Castanet, A.-S.; Mortier, J. Chem. Lett. 2005,
34, 446.
(7) The term “the H-5 proton” used refers to the 5-position of the aromatic
ring of the starting anisic acid and to the meta position with respect to both
the methoxy and the carboxylate groups in the obtained products.
(8) (a) Marsais, F.; Laperdrix, B.; Gu¨ngo¨r, T.; Mallet, M.; Que´guiner,
G. J. Chem. Res., Miniprint 1982, 2863. (b) Krizan, T. D.; Martin, J. C. J.
Am. Chem. Soc. 1983, 105, 6155.
(9) By comparison, lithium ortho-lithio benzoate 2-LiC6H4CO2Li con-
denses instantaneously with itself as the temperature is increased to -20
°C to give benzophenone derivatives. See: Bennetau, B.; Mortier, J.;
Moyroud, J.; Guesnet, J.-L. J. Chem. Soc., Perkin Trans. 1 1995, 1265.
(10) Since an excess (5 equiv) of base was required for optimal
conversion with LTMP using EQ conditions, the stability of 7:LTMP might
be due to steric effects created by a large cluster of yet undetermined
stoichiometry involving both the reactive monomer and LTMP. Attempts
of crystallization for X-ray structure determination in ether or THF failed.
1H NMR spectrum of the residue in THF-d8 displayed broad signals with
poor resolution in the aromatic region, which are probably due to the
presence of aggregates that prevent a simple structural elucidation.
(11) Minor isomers 12 and 14 were not detected by other investigators:
Sinha, S.; Mandal, B.; Chandrasekaran, S. Tetrahedron Lett. 2000, 41, 3157.
(12) Mills, R. J.; Taylor, N. J.; Snieckus, V. J. Org. Chem. 1987, 52,
448.
(13) (a) Beak, P.; Meyers, A. I. Acc. Chem. Res. 1986, 19, 356. (b)
Whisler, M. N.; MacNeil, S.; Snieckus, V.; Beak, P. Angew. Chem., Int.
Ed. 2004, 43, 2206.
(14) Van Eikema Hommes, N. J. R.; Schleyer, P. v. R. Tetrahedron 1994,
50, 5903.
Org. Lett., Vol. 7, No. 12, 2005
2447

In our opinion, both theories furnish a complete picture if
adequately put together. Since there is abundant evidence
in the literature for coordination of lithium to heteroatoms
in ground states,15 it is very probable that the alkyllithium
(i.e., the 1:1 s-BuLi/TMEDA complex in this work) ap-
proaches the benzoate 6 in the initial step by strong coordi-
nation with the highly electron-rich π-system of the carbox-
ylate and the p-electrons of the methoxy group, leading to a
prelithiation complex PLC1 (kinetic control) (Figure 1).
Figure 2. Directed metalation reaction. Qualitative energy diagram.
The directing and accelerating effect of substituents is due
to the stabilization of both the initial complex and the
transition structure. The metal is inVolVed in partial bonds,
and coordination by the substituent becomes stronger in the
transition state than in the initial complex. As a result,
complexation increases the rate of reaction by proViding a
new mechanism that has a smaller actiVation energy (Ea).17
The geometries of the precursor complex and the transition
structure could be radically different.
The results obtained with LTMP (entry 1 of Table 1)
firmly suggest that the regiochemistry of the lithiation of 6
is truly thermodynamically controlled. Resonance and induc-
tive effects favor removal of the H-2 proton.
Figure 1. Prelithiation complexes PLC1-3 and transition states
TS1-3 (S: solvent, TMEDA, or RLi aggregate).
Since we have checked that bromine-lithium exchange
of 2-bromo-3-methoxybenzoic acid (16b) (preparation vide
supra) with s-BuLi (2 equiv) alone or chelated to TMEDA
in ether (-78 °C f rt) followed by trapping with iodo-
methane provides 3-methoxy-2-methylbenzoic acid (12) as
a single regioisomer, it is reasonable to assume that s-BuLi/
TMEDA metalates 6 randomly at the C-2, C-4, and C-6 sites
and that there is no interconversion of the resulting organo-
metallic species 7-9. Consequently, equilibration occurs be-
tween the prelithiation complex PLC1 and the less stabilized
forms PLC2 and PLC3 coordinated by only one substituent.
The probable reaction pathway for the conversion of 6 to
the dianions 7-9 is illustrated by the qualitative energy
diagram in Figure 2. The directed lithiation is suggested to
proceed by a two-step mechanism in which initial complex-
ation is reversible, whereas the second step is rate determin-
ing.16 Not only must the reactants be brought together by
chelation of s-BuLi by the substituents (to form the pre-
lithiation complexes PCL1-3) but they also have to be held
in exactly the right orientation relative to each other in the
transition states TS1-3 (represented in Figure 1) to ensure
that deprotonation can occur. Both of these factors raise the
free energy of the system by lowering the entropy. Some
energy also must be invested to begin breaking the C-H
bond so that the C-Li bond can form.
The major product 13 obtained (entry 6 of Table 1) arises
from a noncoordinating-deprotonation mechanism at what
would appear to be the least activated of the methoxy ortho
centers. Superbases are not significantly influenced by ortho-
directing groups and preferentially attack the inductively
activated aromatic position next to the most electronegative
heteroatom and/or the most acidic position available.18
Extensions of the manipulation of carboxylic functional
group are ongoing in our laboratories and will be reported
upon in due course.
Acknowledgment. This research was supported by the
CNRS, Universite´ du Maine, and Institut Universitaire de
France.
Note Added after ASAP Publication. Errors were
detected in Scheme 2 and Table 2 in the version published
ASAP May 17, 2005; the revised version was published
ASAP May 18, 2005.
Supporting Information Available: Details of compound
characterization. This material is available free of charge via
the Internet at http://pubs.acs.org.
OL050761C
(16) Anderson, D. R.; Faibish, N. C.; Beak, P. J. Am. Chem. Soc. 1999,
121, 7553.
(17) There are two transition states, each with its own activation energy
(Ea1 and Ea2). The overall activation energy is the difference in energy
between the reactant state and the highest energy transition state (Ea).
(18) See: (a) Bauer, W.; Lochmann, L. J. Am. Chem. Soc. 1992, 114,
7482. (b) Shi, G.; Takagishi, S.; Schlosser, M. Tetrahedron 1994, 50, 129.
(15) See, inter alia: (a) Streitwieser, A. Jr.; Williams, J. E.; Alexandralos,
S.; McKelveg, J. M. J. Am. Chem. Soc. 1987, 111, 4778. (b) Hay, D. R.;
Gallagher, D. J.; Du, H.; Long, S. A.; Beak, P. J. Am. Chem. Soc. 1996,
118, 11391.
2448
Org. Lett., Vol. 7, No. 12, 2005