Fluorine as an ortho-directing group in aromatic metalation: Generality of the reaction and the high position of fluorine in the Dir-Met potency scale

Article abstract of DOI:10.1016/S0040-4039(00)60805-5

Many para-substituted fluorobenzenes can be lithiated ortho to fluorine in moderate to good yields, often with one of the dialkylamide bases, lithium diisopropylamide (LDA) or lithium 2,2,6,6-tetramethylpiperidide (LiTMP). Intramolecular competition experiments reveal that fluorine is one of the most potent Dir-Met activating groups under these conditions.

Full text of DOI:10.1016/S0040-4039(00)60805-5

Vol. 33, No. 49. pp.
in Great Britain
+
Fluorine as an ortho-Directing Group in Aromatic Metalation: Generality of
the Reaction and the High Position of Fluorine in the Dir-Met Potency Scale
Alexander J .
Arthur Lee, Emmanuel C. Maduakor and C. Eric Schwartz
Research Institute, 4
Drive. Andover MA 01810
Many
can be
ortho to fluorine in moderate to good yields,
or lithium
one of the
bases. lithium
Intramolecular competition experiments reveal
groups under these conditions.
fluorine is one of the most potent Dir-Met activating
In recent years, much of the advance in aromatic synthesis has come in the non-classical area of aromatic
A fluorine atom considerably acidifies the neighboring ortho hydrogen atom(s) of an aromatic
ring.2 and should be a
the use of this strategy is the perceived low position of fluorine on the directed-metalation (Dir-Met) potency
scale.3 where most other functional groups are expected to control the metalation
participant in this type of chemistry (Scheme 1, path 2).
An apparent drawback in
in preference to fluorine. This perception has come, in part, from early studies on the metalation of
fluoroanisole.4 where only the product of lithiation ortho to oxygen was observed.5 A recent preparation of
revealed that both the
fluorine atom in a Dir-Met reaction, rather
and pivaloylamido groups both completely outcompete a
the view of fluorine as a poor ortho-directing group.
Accordingly, the general opinion might be that, for many if not most useful metalations, path 1 (X-directed
should dominate over path 2 (fluorine-dimcted metalation).
F
PATH
F
PATH 2
F-control
X
X
Litbiatioa-Carboxylation of
Derivatives
Scheme 1.
In this communication we wish to report that, in contrast to the low Dir-Met activity of fluorine implied
by the literature, conditions can be readily found where a single fluorine atom can be one of the best ortho-
directing substituents available.
Furthermore, many aryl fluorides can be cleanly and rapidly
temperatures by lithium dialkylamide bases, and many useful functional groups are compatible with these
metalating conditions.
at low
Carboxylation of the resulting aryl
3 provides
acids 5 in useful
Address: Parke-Davis
Research Division,
Plymouth
Ml 48105.
7495

7496
synthetic yields. Similar results have also been obtained in our laboratory by trapping the intermediates 3 with
other (such as DMF, and
We chose a standard protocol for all of our
using
as the electrophile.8 and varying
only the base and reaction
our results, we used in
In those cases, where we suspected that anion
might be distorting
chlorotrimethylsilanc as the electrophile.
None of these reactions have been
optimized, and we believe that many yields may be improved by examining other variables which we ignored
(eg. solvent, concentration, stoichiometry, temperature).
Table 1 shows the results of our carboxylation experiments. Entries
fluorobenzene itself is not straightforward. n-Butyllithium proved inefficacious for metalation at
the presence of TMEDA, whereas was quite effective, and s-butyllithium (Entry 3) was excellent. A
wide variety of are tolerated in these metalations, as can be seen in the remainder of Table 1.
demonstrate that the lithiation of
even in
Although several substituents have been described in the literature as ortho directors for aromatic metalations
a fluorine atom is an overall superior ortho director under the
conditions described. Furthermore, several entries (Entries
demonstrate that even the most mildly
Table 1.
of Aryl Fluorides (1)
by
unstable to workup, yield quoted for
acid.
DMF quench.
used.
acid.

acidifying substituent shows additivity with fluorine in acidification of its ortho proton, making high yield
metalations by LDA possible at This allows for the very reactive but versatile iodo (or group to
without the problems of competing halogen-metal exchange
Even can be metalated in moderate yield with
be tolerated during lithiition (Entries 9 and
reactions associated with the use of alkyl
although the yield is considerably improved using
of benzylic metalation were isolated in these reactions. The lithiation of
No products
(Entry 12) is not
in sharp contrast to
the metalation
(Entries 5 and
only completely tegiospecific, but produces an anion which is reasonably stable at
simple ortho lithiated
regiochemistry of
Additionally, in contrast to an older literature
can be completely controlled by fluorine when LiTMP is used as the base
(Entry 15). Recently, this complete reversal of regiochemistry has also been reported using
In contrast to these earlier reports. under our conditions, we found no bases which metalated exclusively ortho to
(Entries 16-19). and
carbamate was metalated (Entries 20 and
was again surprisingly inefficacious. However, when
no ortho to fluorine metalation was seen with
and
As Entry 22 shows, the directing effect of fluorine versus methoxy was enhanced
less than 20% with LiTMP.
when the two groups were ortho rather than
Table 1 does not include the strong ortho directors
and
as the
in competition with fluorine. The
of these
substrates was examined, and gave very low yields of carboxylic acids and surprisingly poor material balances.
For example,
did show exclusive ortho to fluorine metalation in 26% yield
2 h), but under other conditions three different acids were obtained in even lower yields, none
The
of the other two aryl fluorides was even
of which could be unequivocally identified.
less promising. In order to obtain a relative potency order against fluorine for these three groups, the anions
generated by LiTMP were trapped in situ by with the results shown in Table 2.
Under these conditions, diethylcarboxamide (Entry 23) appears approximately equal in
However, fluorine was a much stronger ortho director than oxaxoline (Entry
ability when
and no
compared to fluorine.
trace of metalation ortho to the pivaloylamido group was observed (Entry 25). This last result is in contrast to
the
of Muchowski aud coworkers6 who reported exclusive ortho to nitrogen lithiation with
(2.5
Apparently, the N- (or 0-) lithioamide produced with
is a much stronger ortho
equiv base,
director than the (presumed) 0-silylimidate produced under our conditions. A possible explanation for the
difference between our silylations and carboxylations is that ortho-lithio aryl fluorides were always
generated in the latter reactions, and that these
protection from self-condensation (in the absence of a trapping agent) that the
lithium species do not share the same
analogues possess.
Table 2. In Situ Trapping with
LiTMP
3-Si
to F di-Si
2,5-
di-Si
a-F/
recovered
3
22
23
24
0
0
15
39
54
34
15
19
9.5
6
0
0.54-1.2
23
24
25
1 .O
1 .O equiv
2.0
Ratio varies, depending on where
compound is assumed to have silylated first.
In summary, we have demonstrated that most para-substituted fluorobenzenes can be lithiated ortho to
fluorine in good yields and with high mgioselectivity.
From our internal competition results it is clear that a
fluorine atom can be one of the most potent ortho directing groups for aromatic metalation, given that the
appropriate base (often LDA) is used in conjunction with a low reaction temperature.* In addition, many useful

7498
functional groups can be tolerated in these lithiations, especially if LDA is the metalating agent, and substitution
of other electrophiles for
should allow for the preparation of a wide variety of highly functionalized
REFERENCES AND NOTES
For reviews of directed metalation see: a) Beak, P.; Meyers, A. I. Ace. Chem. Res.
b)
1.
Beak, P.; Snieckus V.
Chem. Res.
306. c) Snieckus, V. Pure App. Chem. 1990.62,
2047. d). Snieckus. V. Chem. Rev.
2.
3.
a) Streitweiser, A; Mares, F. J. Am. Chem.
Niemeyer, H. M. ibid 1975.97, 7936.
90, 644. b) Streitweiser, A.;
P. J.;
a) Fraser, R. R.; Bresse, M.,
Ronald, R. C. J. Org. Chem.
T. S. J. Am. Chem.
2101.
b) Winkle, M. R.;
Slocum, D. W.; Jennings; G. A. J. Org. Chem.
3653.
4.
5.
For the opposite results see: a) Furlano, D. C.; Calderon, S. N.; Chen, G.; Kirk, K. L. J. Org. Chem.
1988.53, 3145. b) Katsoulos, G.; Takagishi, S.; Schlosser, M. Synlett 1991, 731.
Cho, L.-S.; Gong, L.; Muchowski, J. M. J. Org. Chem. 1990.56, 7289.
Bridges, A. J.; Patt, W. C.; Stickney, T. M. J. Org. Chem
6.
7.
8.
A typical experimental
mmol) was added
6) is as follows:
over 5 min to a
(1.06 M in cyclohexane, 1.88
solution of 4-fluorotoluene (220.4 mg. 2.0 mmol) and
stirred under After 2 h, the slurry was blown rapidly
When the reaction warmed up to it was
separated, the organic phase was washed with
2.0
(231.5 mg, 2.0 mmol) in THF (5
through a catheter onto solid
quenched with dilute
in ether (50
M, 10
The phases
solution (0.2 M, 2x10
acidified with dilute (1 M, 10
The combined ether extracts were washed with water
After filtration, the solvent was removed under reduced pressure to give
methylbenzoic acid (238.7 mg 77%) as white needles, mp NMR
s), 7.65 sl br dd, J = 1.9, 7.3 Hz), 7.41 sl br ddd, J = 1.9, 4.2, 8.3 Hz), 7.18
8.3, 11.0 Hz), 2.31 s).
Grtho-fluoro aryllithiums are known to decompose to benzyne intermediates at ca.
available temperature range for metalation. See: a) A.; Miller, D. D. Tetrahedron
5597. b) Huisgen, R.; Rist. H. Liebigs Ann. Chem. 1955,594, 147.
a) Block, E.; Eswarakrishnan, V.; M.; G.; Saha, C.; Tang, K.; Zubieta, J.
water (10
and extracted with dilute
The combined base extracts
and extracted with ether (3x10
were washed with ether (10
saturated brine (10
and dried
12.9
br
dd, J =
9.
limiting the
10.
Am. Chem.
1989,111, 658. b) -CN: Krizan, T. D.; Martin, J. C. ibid 1983,105, 6155. c)
M. P.; Corran, P. J.; Zuckerman, J. J. ibid 1981,103, 7643.
Takagishi, S; Schlosser, M. Synlett 1991, 119.
11.
12. a) Beak, P.; Brown, R. A. J.Org. Chem.
34. b) Edgar, K. J.; Bradsher, C. K. ibid 1982,
47, 1585. c) Fuher, W.; Gschwend, H. W. ibid. 1979.44, 1133.
13. Bridges, A. J.; Lee, A.; Maduakor, E. C.; Schwartz, C. E.; following communication this journal.
(Received in USA 14 May 1992; accepted 2 September 1992)