The Birch reduction-dialkylation reaction. Scope and limitations

Article abstract of DOI:10.1016/j.tetlet.2005.11.138

The Birch reduction-dialkylation reaction has been studied with different aromatic carboxylic acids and primary and secondary bromides. From these results it can be concluded that it is a reasonably general reaction. However, we found important exceptions with the primary bromides: allyl bromide, β-phenethyl bromide, and 5-bromo-2-methyl-2-pentene. The unusual behavior of allyl bromide is briefly discussed.

Full text of DOI:10.1016/j.tetlet.2005.11.138

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 973–976
The Birch reduction–dialkylation reaction. Scope and limitations
*
´
´
Ricardo Castanedo, Adrian Covarrubias-Zuniga and Luis A. Maldonado
˜
Instituto de Quı´mica, Universidad Nacional Auto´noma de Me´xico, Circuito Exterior, Ciudad Universitaria,
Coyoaca´n 04510, Me´xico D.F., Mexico
Received 21 October 2005; revised 23 November 2005; accepted 28 November 2005
Available online 19 December 2005
Abstract—The Birch reduction–dialkylation reaction has been studied with different aromatic carboxylic acids and primary and sec-
ondary bromides. From these results it can be concluded that it is a reasonably general reaction. However, we found important
exceptions with the primary bromides: allyl bromide, b-phenethyl bromide, and 5-bromo-2-methyl-2-pentene. The unusual behavior
of allyl bromide is briefly discussed.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
acid 1 was submitted to the Birch reduction–dialkyla-
tion reaction with different primary and secondary alkyl
bromides.3 Sodium was used as the reducing alkali metal
and analysis of the crude reaction mixtures were done by
1H NMR spectroscopy of the crude methyl esters as
reported previously for the ethyl case 2a.4 Results are
presented in Table 1.
The Birch reduction–dialkylation of aromatic carboxylic
acids is a potentially useful reaction since after the usual
Birch reduction of the aromatic ring, two C–C bonds are
formed sequentially at the C-1 and C-4 positions of the
intermediate 1,4-dihydrobenzoic acid.1 The reaction was
first observed with 3,5-dimethoxybenzoic acid and
excesses of Na and ethyl bromide in liquid NH3 although
Li but not K, also gives similar results (Scheme 1).2
We have now extended our studies to other aromatic
carboxylic acids and alkyl bromides with the purpose
to determine its generality.
With 6.2 molar equiv of Na, dialkylation with primary
alkyl bromides (entries 1 and 2) proceeded smoothly
to give high yields of products, whilst the secondary iso-
propyl bromide afforded the dialkylated products con-
taminated with 20% of monoalkylated product. Since
dehydrobromination of secondary alkyl bromides was
expected to be a competitive base-consuming side reac-
tion, in experiments with these bromides the molar
equivalent of Na and alkyl bromides were increased
up to 10.2, but even so some monoalkylated compound
still remained in the product (10% for isopropyl bro-
mide, entry 3 and 12% for cyclohexyl bromide, entry 5).
2. Birch reduction–dialkylation of 3,5-dimethoxybenzoic
acid with alkyl bromides
Following the general procedure described in Ref. 1, in
the first group of experiments 3,5-dimethoxybenzoic
2a, R1=R2=C2H5
2b, R1=R2=n-C3H7
2c, R1=R2=n-C5H11
2d, R1=R2=i-C3H7
2e, R1=R2=(CH2)4CH
2f, R1=R2=(CH2)5CH
2g, R1=H, R2=Me2C=CH(CH2)2
2h, R1=H, R2=Ph(CH2)2
2i, R1=H, R2=CH2CH=CH2
CO2H
HO2C
R2
Na (excess)
RBr (excess)
(NH3)
MeO
OMe
MeO
OMe
R1
1
Scheme 1.
Keywords: Birch reduction; Dialkylation; Primary and secondary alkyl bromides; Substituted benzoic acids.
*
Corresponding author. Tel.: +1 52 5622 4428; fax: +1 52 5616 2217; e-mail: lammg@servidor.unam.mx
0040-4039/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.11.138

974
R. Castanedo et al. / Tetrahedron Letters 47 (2006) 973–976
Table 1. Birch reduction–dialkylation of 3,5-dimethoxybenzoic acid
Entry
RBr (molar equiv)
Na (molar equiv)
Main product (see text)
Yield (%)
1
2
3
4
5
6
7
8
n-C3H7– (8.2)
n-C5H11– (8.2)
i-C3H7– (8.2)
(CH2)4CH– (10.2)
(CH2)5CH– (10.2)
Me2C@CH(CH2)2– (8.2)
Ph(CH2)2– (8.2)
CH2@CHCH2– (8.2)
6.2
6.2
10.2
10.2
10.2
8.2
2b
2c
2d
2e
2f
95
84
98a
91
76b
55
2g
2h
2i
8.2
8.2
75
81c
a The product contained 12% of mono iso-propyl compound.
b The product contained 10% of monocyclohexyl compound.
c The product contained 40% of starting material (see text).
Surprisingly, cyclopentyl bromide (entry 4) gave cleanly
only dialkylated products in high yield and it is also
noteworthy that a reasonably good yield was obtained
with the easily dehydrohalogenated cyclohexyl bromide
(entry 5).
HO2C
HO2C
1) n-BuLi (THF)
1
MeO
OMe
2) CH2=CHCH2Br
MeO
OMe
2i
Even more easily dehydrohalogenated primary alkyl
bromides such as the homoprenyl and homobenzyl bro-
mides failed to give the dialkylation reaction and affor-
ded only monoalkylated products in fair yields (entries
6 and 7). Probably, after alkylation at C-1 NaNH2
and/or the sodium carbanion at C-4 are too basic and
dehydrohalogenate the excess alkyl bromide before
alkylation at C-4 takes place. Fortunately, in these and
other related cases dialkylated products can be obtained
successfully in separate experiments by C-4 metalation
(n-BuLi, THF, À78 ꢁC)–alkylation of the corresponding
monoalkylated dihydrobenzoic acids (Scheme 2).5 Inci-
dentally, we have used this two-step method for the
synthesis of several 1,4-dialkyl-2,5-cyclohexadiene car-
boxylic acids containing two different groups at C-1
and C-4.
(not formed)
Scheme 3.
CO2Li
LiO2C
MeO
Li
+
OMe
MeO
OMe
Li
Scheme 4.
This unusual aromatization, which involves elimination
of a poor leaving group has been previously observed
with the somewhat related benzyl anion.6
Finally, our reaction conditions were also tested with the
highly reactive allyl bromide but unexpectedly, the 1,4-
diallyl compound was not obtained. A 3:2 mixture of
monoallyl compound 2i and 3,5-dimethoxybenzoic acid
1 were produced instead. We could demonstrate that the
3,5-dimethoxybenzoic acid present in the crude product
was not due to an incomplete Birch reduction when we
attempted to prepare 1,4-diallyl compound from 2i by
the two-step method cited before (n-BuLi, THF,
À78 ꢁC; then allyl bromide). The starting material was
totally consumed and a quantitative yield of crude 3,5-
dimethoxybenzoic acid was obtained (Scheme 3).
3. Birch reduction–diethylation of substituted benzoic
acids
The second group of experiments we performed involved
the use of different benzoic acids as substrates in the
Birch reduction–diethylation reaction. Ethyl bromide
(8.2 molar equiv) was used both as the ethylating agent
as well as reagent for NaNH2 formation. The experi-
ments were carried out following the procedure
reported for the Birch reduction–diethylation of 3,5-
dimethoxybenzoic acid with 7.2 molar equiv of Na.3
The crude products were analyzed as the methyl esters
1
Presumably, the intermediate lithium carbanion at C-4
aromatizes by expulsion of the allyl anion (Scheme 4).
by H NMR spectroscopy and results are presented in
Scheme 5 and Table 2.
HO2C
HO2C
2.2 Na (NH3)
1) n-BuLi (THF)
1
Me2C=CH(CH2)2I
MeO
OMe
MeO
OMe
2) Me2C=CH(CH2)2I
70 %
87 %
2g
Scheme 2.

R. Castanedo et al. / Tetrahedron Letters 47 (2006) 973–976
975
CO2H
R1
a, R1=R2=H
HO2C
R2
R1
b, R1=Me, R2=H
c, R1=H, R2=Me
d, R1=H, R2=OMe
e, R1=R2=OMe
1) 7.2 Na (NH3)
2) 8.2 EtBr
R2
3
4
Scheme 5.
Table 2. Birch reduction–diethylation of substituted benzoic acids
Entry
Benzoic acid
Diethylated product
Yield (%)
1
2
3
4
5
R1 = R2 = H
R1 = R2 = H
85
86
91
89
91
R1 = CH3, R2 = H
R1 = H, R2 = CH3
R1 = H, R2 = OCH3
R1 = R2 = OCH3
R1 = CH3, R2 = H
R1 = H, R2 = CH3
R1 = H, R2 = OCH3
R1 = R2 = OCH3
Very high yields of diethylated products were obtained
with the five benzoic acids listed in Table 2. The only
substituted benzoic acid tested, which failed to give a
clean reaction was o-anisic acid (not included in the
Table). In this case, a complex mixture of starting
material (10%), mono- (20%) and diethylated (20%)
products, and diethylated 1,4-dihydrobenzoic acid
(50%) was obtained in 86% yield (Scheme 6). Loss of
the o-methoxy group during the Birch reduction of this
acid has been previously observed.7
product was exclusively formed in 74% yield. Interest-
ingly, with reduced amounts of Na (4.2 molar equiv) a
1:1 mixture of mono- and triethylated products were
obtained, but diethylated products were not detected
(Scheme 7).8
In conclusion, the results presented in this letter show
that the Birch reduction–dialkylation reaction is a quite
general transformation with benzoic acids and primary
and secondary bromides. Easily dehydrohalogenated
alkyl bromides (e.g., homoallyl and homobenzyl
bromides) and the highly reactive allyl bromide are
important exceptions in this reaction.
We have also submitted 1-naphthoic acid to the
Birch reduction–diethylation reaction but triethylated
MeO2C
MeO2C
OMe
OMe
+
+
CO2H
20 %
20 %
OMe
1) 7.2 Na (NH3)
2) 8.2 EtBr
3) CH2N2
CO2Me
MeO2C
OMe
+
50 %
10 %
Scheme 6.
CO2H
CO2H
CO2H
1) n Na (NH3)
2) 8.2 EtBr
+
0 %
74 %
n = 7.2
n = 4.2
87 % (1:1)
Scheme 7.

976
R. Castanedo et al. / Tetrahedron Letters 47 (2006) 973–976
Note: In the case of methoxy substituted benzoic acids,
Acknowledgement
careful acidification with cold acid and rapid extraction is
crucial to avoid hydrolysis of the acid sensitive methoxy enol
ethers present in the products. For alkyl substituted benzoic
acids, the usual acidification and workup can be followed.
4. Furthermore, values thus obtained were checked in every
case by an alternative method involving the two-step
conversion (acid-catalyzed hydrolysis and ethereal
CH2N2) of the alkylated dihydrobenzoic acid methyl esters
into the corresponding cyclic vinylogous methyl esters 5,
which were separated and purified by preparative TLC. If
present, the more polar 2-unsubstituted vinylogous methyl
ester 5a (from monoalkylated product) was easily identified
by the olefin broad singlet signal at ꢀd 5.20 in its 1H NMR
spectrum. This second method of analysis was especially
useful when important amounts of monoalkylated product
were obtained.
´
We thank Q. Alejandrina Acosta, Q. Angeles Pena, Q.
˜
´
Rocıo Patino, and I. Q. Luis Velasco for running spectra.
˜
References and notes
´
1. Guzman, J. A.; Castanedo, R.; Maldonado, L. A. Synth.
Commun. 1991, 21, 1001–1012.
2. Unpublished results.
3. General procedure: In a two-necked 250 mL round bottom
flask, fitted with a dry ice condenser and a stopper, were
placed 1 g (5–8 mmol) of the appropriate benzoic acid and a
stirring bar. Dry liquid NH3 (125 mL, from Na) was
condensed in the flask, and with stirring, the required molar
equivalent of Na (6.2–10.2) was added in small pieces (10–
15 min depending of the amount of metal). The initial
almost colorless solution gradually turns yellow, orange,
and deep red with the first 2 equiv of metal and beyond this
number, the solution becomes deep blue. After 15 min, the
stopper was replaced by an addition funnel containing cold,
neat alkyl bromide (8.2–10.2 molar equiv) and added drop-
wise in about 10 min (CAUTION! Exothermic reaction).
The sequence of colors in the solution reverses, changing
from deep blue to deep red, orange, and finally pale yellow.
NH3 was left to evaporate overnight, the residue dissolved in
H2O (30 mL) and washed with Et2O to remove any neutral
material. The aqueous layer was covered with fresh Et2O
(20 mL) and carefully acidified (pH 5) with small portions of
cold 5% aqueous HCl, shaking vigorously after each
addition. The aqueous solution was extracted again with
20 mL of fresh Et2O and the combined organic extracts
rapidly washed with H2O (2 · 10 mL) and dried over
anhydrous Na2SO4. Removal of solvent gave the crude
product as a white or yellow crystalline mass in the indicated
yields. About 100 mg of the crude acids were quantitatively
esterified with ethereal CH2N2 and the crude methyl esters
MeO2C
O
R2
MeO2C
MeO
R2
1) H3O
2) CH2N2
OMe
OMe
R1
R1
5a, R1 = H
5b, R1 = R2
5. Piers, E.; Grierson, J. R. J. Org. Chem. 1977, 42, 3755–
3757.
6. Collins, C. J.; Hombach, H. P.; Maxwell, B. E.; Benjamin,
B. M.; McKamey, D. J. Am. Chem. Soc. 1981, 103, 1213–
1214.
7. (a) House, H. O.; Strickland, R. C.; Zaiko, E. J. J. Org.
Chem. 1976, 41, 2401–2408; (b) Birch, A. J.; Slobbe, J. Aust.
J. Chem. 1977, 30, 1045–1049.
8. The reductive methylation of 1-naphthoic acid with Li in
liquid NH3–MeI has been previously reported: (a) Slobbe,
J. Aust. J. Chem. 1978, 31, 1157–1159; (b) Murthy, A. R.;
Sundar, N. S.; Subba Rao, G. S. R. Tetrahedron 1982, 38,
2831–2836.
1
were analyzed by H NMR and/or GC.