Ammonia-free Birch reductions with sodium stabilized in silica gel, Na-SG(I)

Article abstract of DOI:10.1016/j.tetlet.2009.07.040

Ammonia-free Birch reduction conditions were developed based upon sodium stabilized in silica gel for a variety of substrates. In general, the yields were similar to those reported for lump sodium in liquid ammonia.

Full text of DOI:10.1016/j.tetlet.2009.07.040

Tetrahedron Letters 50 (2009) 5463–5466
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Ammonia-free Birch reductions with sodium stabilized in silica gel, Na–SG(I)
Michael J. Costanzo , Mitul N. Patel, Kathryn A. Petersen, Paul F. Vogt *
*
Process Development Services, SiGNa Chemistry, 1 Deer Park Drive, Suite C, Monmouth Junction, NJ 08852, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 June 2009
Revised 7 July 2009
Accepted 7 July 2009
Available online 5 August 2009
Ammonia-free Birch reduction conditions were developed based upon sodium stabilized in silica gel for a
variety of substrates. In general, the yields were similar to those reported for lump sodium in liquid
ammonia.
Ó 2009 Elsevier Ltd. All rights reserved.
The reduction of aromatic compounds by alkali metals in liquid
ammonia is a synthetically powerful and versatile method that has
been in use since its discovery in 1937 by Wooster and Godfrey.
pounds with lithium metal.7a,b This reduction is performed in a
mixture of THF and bis-(methoxyethyl)-amine at ꢀ78 °C with
either di-tert-butyl-biphenyl or naphthalene serving as an electron
carrier. One major advantage of this ammonia-free method over
the classic Birch reduction is the ability to subsequently react the
intermediate anions with electrophiles (e.g., alkyl halides, acid
chlorides, and aldehydes) that may otherwise react with ammonia
under the standard Birch conditions.
1
The full scope and limitations of this method were expanded
through the efforts of Birch, hence it now bears his name (Eq.
2
a–d
1).
H
H
M / NH3
Recently, we have developed a technology for encapsulating al-
kali metals into nano-structured porous oxides, such as silica gel
ð1Þ
ROH, co-solvent
8
a,b
-
33 °C or less
M = Na or Li)
and alumina.
Encapsulation reduces the dangers associated
H
H
(
with the handling of alkali metals while still retaining the reducing
power of the metal. Stage I sodium and sodium-potassium alloys in
2 2
silica gel (Na–SG, Na K–SG, and K Na–SG) are free-flowing black
The Birch reduction is one of only a few methods that can read-
ily convert aromatic synthons into alicyclic structures and has been
used in the synthesis of drugs and complex natural products.
powders that are typically loaded with 35–40 wt % of the neat al-
9
3
a–d
kali metal. These reagents have demonstrated applications in es-
1
0
11
8a
ter reductions,
detosylations,
desulfurizations,
and the
Despite its great utility, the classic Birch reduction has several
undesirable attributes that have limited its use, particularly on a
1
2
cleavage of arylphosphines to metal phosphides.
4
a,b
Herein we describe a safer and more practical modification of the
classic Birch reduction that avoids the use of liquid ammonia and
cryogenic temperatures (Eq. 2). Central to this modification is the
use of Stage I sodium in silica gel, Na–SG(I), as the reducing agent.
Na–SG(I) is a more convenient and safer form of metallic sodium than
eitherlumpsodiumorsodiumsandbecauseitcanbeeasilyhandledin
the open air without loss of activity and it is safer to quench.
large scale.
Primary among these are the inherent dangers of
handling elemental alkali metals, the practicality and hazards of
5
a,b
running cryogenic reactions, and the high toxicity of ammonia.
Consequently, researchers have sought alkali-metal reduction
conditions that avoid the use of liquid ammonia. Benkeser et al.
have successfully reduced aromatic substrates with lithium metal
in neat low molecular weight amines. These include primary
amines, primary and secondary amine mixtures, and ethylenedia-
mine (EDA) without a cosolvent or an alcohol present to serve as
H
H
Na-SG (I)
t-BuOH or t-AmOH
6
a,b
a proton donor.
This method, known as the Benkeser reduction,
ð2Þ
is generally a more powerful reducing system than the Birch
reduction and frequently results in over-reduced product mixtures.
However, greater selectivity can be achieved in neat mixtures of
primary and secondary amines. In a second method, Donohoe
et al. have reported on the ammonia-free reductions of electron-
deficient (i.e., activated) hetero- and carbocyclic aromatic com-
with or without EDA
solvent, 0 °C or RT
H
H
The Birch reductions of 16 different substrates with Na–SG(I) un-
der ammonia-free conditions were examined and compared with
literature values for sodium in liquid ammonia (Table 1). These sub-
strates fall into 4 compound classes, including polycyclic aromatic
hydrocarbons (1a–4a), aromatic hydrocarbons (5a–7a), aryl ethers
*
Corresponding authors. Tel.: +1 732 438 0700x11; fax: +1 732 438 1341 (P.F.V.).
(8a–12a), and polycyclic aromatic heterocycles (13a–16a).
E-mail address: pfvogt@signachem.com (P.F. Vogt).
0
040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.07.040

5
464
M. J. Costanzo et al. / Tetrahedron Letters 50 (2009) 5463–5466
Table 1
Comparison of the Birch reduction of aromatic substrates with Na–SG(I) under ammonia-free conditions with lump sodium in liquid ammonia
Na–SG(I) methoda
Rxn equivb
Time (h)
Na–SG(I) Prod. /Yieldd (%)
c
Na/NH
(lit.)e Prod./Yieldd (%)
3
Entry
Substrate (a)
Major Product (b)
1
A
3.5
2.0
96/70
—/85f
2
3
4
A
B
C
2.0
3.5
2.7
1.0
2.5
4.0
87/83
99/94
63/34
—/97g
—/96g
52/20h
Ph
Ph
—/80–90i
—/80–90i
5
6
Ph
Ph
Ph
A
A
3.5
3.5
2.0
2.0
100/83
100/82
5b
Ph
Me
Me
7
8
9
D
D
A
A
D
E
14
24
24
24
24
24
24
24
48
2.0
9/—
—/63j
—/74k
—/19l
—/—m
—/74n
—/—p
—/69q
—/15r
—/—s
OMe
OMe
OMe
14
30/—
12/—
68/43
85/68
—/56o
—/58o
94/62
89/38
OMe
OMe
3.5
3.5
14
OMe
MeO
MeO
MeO
1
1
1
1
1
1
0
1
2
3
4
5
OMe
OMe
MeO
Et
5.6
5.6
14
O
O
OH
12b
E
D
F
N
N
Me
Me
3.5
N
H
N
1
1
6
7
F
3.5
3.5
2.0
2.5
77/40
92/80
—/89t
—/74k
N
NH
8a
8b
G
a
Methods A–F are ammonia-free conditions and method G (entry 17) was run in liquid ammonia; all methods are described in detail in Ref. 14.
Number of Na–SG(I) reaction equivalents (2 moles of Na are required per mole of substrate).
Percent product in the crude reaction mixture as estimated by GC/MS (either EI or CI detection).
Isolated yield after purification.
Literature values.
Ref. 15.
2
Ref. 16; performed in either Et O or THF at room temperature instead of liquid ammonia.
Ref. 17; percent conversion as estimated by GC (FID).
b
c
d
e
f
g
h
i
Ref. 18.
Ref. 19.
Ref. 20.
j
k
l
2
,5-Dihydroanisole obtained through demethoxylation followed by Birch reduction was reported as the major product in Ref. 20 (37% by GC).
m
n
o
A crude yield of 90% of 1,4-dimethoxycyclohexa-1,4-diene was reported in Ref. 21.
Ref. 22.
Unidentified products, presumably a mixture of Birch reduction isomers by NMR (m/z 122) were also isolated for entries 12 and 13, however they were inseparable from
2
-ethylphenol by GC/MS.
p
3 3
No literature values exist for Na/NH , however a 95% yield of 2-ethylphenol was reported for Li/NH in Ref. 23.
Ref. 24.
The major products as reported in Ref. 25 were 1-methylindoline (15%) and 4,7-dihydro-l-methylindole (20%).
q
r
s
t
No literature values exist for the preparation of 1,2,3,4-tetrahydroquinoline with Na/NH
3 3
, however a 36% yield was reported for Li/NH in Ref. 26.
Ref. 27.

M. J. Costanzo et al. / Tetrahedron Letters 50 (2009) 5463–5466
5465
Several sets of ammonia-free reaction conditions (methods A–F,
Table 1) were developed and optimized for each compound. Like
the Benkeser reduction, most of our conditions use an amine, eth-
ylenediamine (EDA), to facilitate the reduction. However unlike the
classic Benkeser conditions, our reaction conditions use only
reaction conditions.13 For example, the reductions of 2,3-dihydro-
benzofuran (12a) and benzofuran (13a) with excess Na–SG(I) un-
der ammonia-free conditions both provided 2-ethylphenol (12b)
as the major product (56% and 58% yields, respectively). In general,
we found that increasing the amount of alcohol and EDA mini-
mized the amount of dealkylation observed with aryl ethers.
The reduction of certain aromatic heterocycles with Na–SG(I)
under ammonia-free conditions was also investigated. While the
reduction of N-methylindole (15a) was somewhat sluggish
(48 h), it provided 15b in 62% yield, which is a dramatic improve-
ment over the 15% yield reported in the literature. Quinoline (15a)
and isoquinoline (16a) were quickly reduced with excess Na–SG(I)
to 15b and 16b, respectively, although the yield for 16b (40%) was
lower than the literature yield (89%).
In conclusion, we have developed ammonia-free modifications
of the classic Birch reduction based upon sodium encapsulated in
silica gel, Na–SG(I). Na–SG(I) is a more convenient and safer form
of metallic sodium than either lump sodium or sodium sand be-
cause it can be easily handled in the open air without loss of activ-
ity. In general, the yields for a variety of substrates using Na–SG(I)
under ammonia-free conditions were similar to those reported for
lump sodium in liquid ammonia.
1
mole of amine per mole of sodium rather than a large excess of
amine as the solvent. In addition, our conditions employ an alcohol
as the proton source instead of an amine, as in the Benkeser condi-
tions. Although some of the substrates studied can be quickly and
cleanly reduced with Na–SG(I) without EDA, the presence of EDA
generally shortened the reaction times and increased the product
yields. In addition, we found t-amyl alcohol (2-methyl-2-butanol)
to be more convenient to use than t-butanol because of its lower
melting point (ꢀ12 °C vs 26 °C), although there were some in-
stances were t-butanol gave better results. All of the Na–SG(I)
reductions shown in Table 1 were run either in THF at 5 to 5 °C
or in 1,4-dioxane at room temperature. However, depending upon
the solubility of the substrate, other aprotic solvents can be used.
These include heptanes, cyclohexane, toluene, methyl t-butyl ether
(
MTBE), 1,2-dimethoxyethane (DME), and 2-methyltetrahydrofu-
ran. Usually 2.0–3.5 reaction equivalents of sodium as Na–SG(I)
were needed to completely consume the starting substrate, which
is typical for classical Birch conditions with sodium in liquid
ammonia. Some slower reactions required more than 3.5 reaction
equivalents of sodium to proceed to completion because of the com-
petitive decomposition of sodium from reaction with the alcohol.
The results given in Table 1 show that the yields using Na–SG(I)
under ammonia-free conditions are generally similar to those re-
ported in the literature for lump sodium in liquid ammonia,
although there are some notable exceptions. The polycyclic aro-
matic hydrocarbons indene (1a) and naphthalene (2a) were cleanly
and quickly reduced in the presence of EDA to 1b and 2b in 71%
and 83% yields, which are similar to the literature yields of 85%
and 97%, respectively. Anthracene (3a) was reduced with Na–
SG(I) in the absence of EDA to furnish 3b in 94% yield. In contrast,
reduction of anthracene (3a) with lump sodium metal under other-
wise identical conditions provided 3b in only 14% conversion. The
Na–SG(I) reduction of phenanthrene (4a) to 4b under optimal con-
ditions required the formation of the intermediate dianion at low
temperature (ꢀ74 °C) before addition of the proton source (t-buta-
nol), otherwise a substantial amount of over-reduced side products
were obtained. Still, the 34% yield of 4b was comparable with the
literature yield (20%).
Acknowledgments
We would like to thank Brian S. Bodnar, Partha Nandi, James L.
Dye, James E. Jackson, and Michael Lefenfeld for many helpful
discussions.
Supplementary data
Supplementary data (Experimentals and NMR data) associated
with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2009.07.040.
References and notes
1
2
.
.
Wooster, C. B.; Godfrey, K. L. J. Am. Chem. Soc. 1937, 59, 596–597.
(a) Birch, A. J. J. Chem. Soc. 1944, 430–436; (b) Birch, A. J.; Subba-Rao, G. S. R.
Reductions by Metal-Ammonia Solutions and Related Reagents. In Advances In
Organic Chemistry, Methods and Results; Taylor, E. C., Ed.; Wiley-Interscience:
New York, 1972; pp 1–65; (c) Rabideau, P. W.; Marcinow, Z. Org. React. 1992,
4
2, 1–334; (d) Birch, A. J. Pure Appl. Chem. 1996, 68, 553–556.
3. (a) Subba-Rao, G. S. R. Pure Appl. Chem. 2003, 75, 1443–1451; (b) Schultz, A. G.
Chem. Commun. 1999, 14, 1263–1271; (c) Menzek, A.; Altundas, A.; Glültekin,
D. J. Chem. Res. (S) 2003, 752–753; (d) Hook, J. M.; Mander, L. N. Nat. Prod. Rep.
The reductions of diphenylacetylene (5a) and trans-stilbene
(6a) with excess Na–SG(I) under the ammonia-free conditions
1986, 3, 35–85.
cleanly provided 5b in yields of 83% and 82%, respectively, which
are analogous to the literature yields (80–90%). In sharp contrast,
the reductions of toluene (7a) and anisole (8a) gave low conver-
sions of 7b (9% vs 63% lit.) and 8b (30% vs 74% lit.), even when a
large excess (14 reaction equiv) of Na–SG(I) was used. A control
reduction of anisole (8a) was performed with Na–SG(I) in liquid
ammonia according to the literature procedure for lump sodium
4
.
(a) Joshi, D. K.; Sutton, J. W.; Carver, S.; Blanchard, J. P. Org. Process Res. Dev.
2005, 9, 997–1002; (b) Gala, D.; Dahanukar, V. H.; Eckert, J. M.; Lucas, B. S.;
Schumacher, D. P.; Zavialov, I. A.; Buholzer, P.; Kubisch, P.; Mergelsberg, I.;
Scherer, D. Org. Prep. Proced. Int. 2004, 8, 754–768.
(a) The maximum permissible exposure limit (PEL) for NH over 8 h is 50 ppm
3
for OSHA and is 25 ppm for NIOSH and ACGIH.; (b) Gangopadhyay, R. K.; Das, S.
K. Process Saf. Prog. 2007, 27, 15–20.
(a) Kaiser, E. M. Synthesis 1972, 8, 391–415; (b) Garst, M. E.; Dolby, L. J.;
Esfandiari, S.; Fedoruk, N. A.; Chamberlain, N. C.; Avey, A. A. J. Org. Chem. 2000,
65, 7098–7104.
(a) Donohoe, T. J.; House, D. J. Org. Chem. 2002, 67, 5015–5018; (b) Donohoe, T.
J.; Thomas, R. E. Nat. Protocol. 2007, 2, 1888–1895.
(a) Dye, J. L.; Cram, K. D.; Urbin, S. A.; Redko, M. Y.; Jackson, J. E.; Lefenfeld, M. J.
Am. Chem. Soc. 2005, 127, 9338–9339; (b) Shatnawi, M.; Paglia, G.; Dye, J. L.;
Cram, K. C.; Lefenfeld, M.; Billinge, S. J. L. J. Am. Chem. Soc. 2007, 129, 1386–
5
.
.
6
(
entry 18). Under these conditions, compound 8b was produced
in 92% yield, which is somewhat improved over the literature yield
74%) obtained with lump sodium. Hence, the poor conversion of
7
.
.
(
8
anisole (8a) to 8b is due to the ammonia-free solvent conditions
rather than from using Na–SG(I) instead of lump sodium. Although
the reduction of 1,2-dimethoxybenzene (9a) gave 9b in only 12%
conversion, it was still similar to the literature yield (19%). 1,4-
Dimethoxybenzene (10a) and 6-methoxy-1,2,3,4-tetra-hydro-
naphthalene (11a) gave reasonable yields of 10b (43%) and 11b
1392.
9.
SiGNa Chemistry Inc. (http://signachem.com/) has developed three categories
of alkali metals in silica gel (M-SG): Stage 0 materials are strongly reducing
pyrophoric powders; stage I materials are synthetically useful reducing agents;
stage II materials have the least reducing capability, but react with water to
form H
2
. All three stages are commercially available.
(
68%), respectively. The results for 8a–11a suggest that the yields
10. Bodnar, B. S.; Vogt, P. F. J. Org. Chem. 2009, 74, 2598–2600.
for anisole derivatives are highly dependent upon the substitution
pattern.
The reductive dealkylation of aromatic ethers with alkali metals
is well known in the literature and is highly dependent upon the
11. Nandi, P.; Redko, M. Y.; Petersen, K.; Dye, J. L.; Lefenfeld, M.; Vogt, P. F.; Jackson,
J. E. Org. Lett. 2008, 10, 5441–5444.
1
1
2. Nandi, P.; Dye, J. L.; Bentley, P.; Jackson, J. E. Org. Lett. 2009, 11, 1689–1692.
3. Casado, F.; Pisano, L.; Farriol, M.; Gallardeo, I.; Marquet, J.; Melloni, G. J. Org.
Chem. 2000, 65, 321–322.

5
466
M. J. Costanzo et al. / Tetrahedron Letters 50 (2009) 5463–5466
1
4. Method A: 1.0 mmol of substrate, 7.0 mmol of Na–SG(I), 7.0 mmol of EDA,
15. Hückel, W.; Schwen, R. Chem. Ber. 1956, 89, 150–155.
16. Menzek, A.; Altundas, A.; Glültekin, D. J. Chem. Res., Synop. 2003, 11, 752–
753.
17. Caamaño, O.; Eirin, A.; Fernandez, F.; Gomez, G.; Lopez, M. C.; Santos, A. An.
Quím., Serie C 1988, 84, 76–81.
18. Streltsova, S. G.; Shilov, E. A. Ukrainskii Khemichnii Zhurnal 1956, 22, 489–492.
19. Gassman, P. G.; Bonser, S. M.; Mlinaric-Majerski, K. J. Am. Chem. Soc. 1989, 111,
2652–2662.
8.4 mmol of t-AmOH in THF at 5–5 °C; Method B: 1.0 mmol of substrate,
7.0 mmol of Na–SG(I), 7.6 mmol of t-BuOH in 1,4-dioxane at rt; Method C:
1.0 mmol of phenanthrene and 7.0 mmol of Na–SG(I) in THF at rt for 15 min,
cool to ꢀ74 °C, add 7.0 mmol of EDA, stir at ꢀ74 °C for 45 min, add 4.2 equiv of
t-BuOH and stir at ꢀ74 °C for 2–3 h; Method D: 1.0 mmol of substrate, 28 mmol
of Na–SG(I), 30 mmol of EDA, 36 mmol of t-AmOH in THF at 5–5 °C. Method E:
1
.0 mmol of substrate, 11.2 mmol of Na–SG(I), 11.2 mmol of EDA, 8.4 mmol of
t-AmOH in THF at 5–5 °C. Method F: Same as Method B except run at 5 to 5 °C
in THF. Method G: 1.0 mmol of anisole, 7.0 mmol of Na–SG(I), 3.2 mmol of EtOH
in liquid ammonia at ꢀ33 °C.General procedure: Na–SG(I) was suspended in
solvent (1.75 mL), cooled to 0 °C, treated with EDA and stirred for 15 min. The
alcohol was added dropwise at <5 °C and stirred for 5 min. A solution of the
substrate (1 mmol) in solvent (0.25 mL) was added and the reaction mixture
was stirred for the specified time. The reaction mixture was carefully quenched
by treatment with 0.6 mL of MeOH/H O (9:1) followed by stirring for 2 h. The
2
mixture was filtered through paper and the filtrate was diluted with water and
extracted twice with EtOAc. The combined organic extracts were washed with
20. Pernemalm, P. A.; Dence, C. W. Acta Chem. Scand. B 1974, 28, 453–464.
21. Kanakam, C. C.; Mani, N. S.; Ramanathan, H.; Subba Rao, G. S. R. J. Chem. Soc.,
Perkin. 1 1989, 11, 1907–1913.
22. Dryden, H. L., Jr.; Webber, G. M.; Burtner, R. R.; Cella, J. A. J. Org. Chem. 1961, 26,
3237–3245.
23. Bachki, A.; Foubelo, F.; Yus, M. Tetrahedron Lett. 1998, 39, 7759–7762.
24. Hurd, C. D.; Oliver, G. L. J. Am. Chem. Soc. 1959, 81, 2795–2798.
25. O’Brien, S.; Smith, D. C. C. J. Chem. Soc. 1960, 4609–4612.
26. Remers, W. A.; Gibs, G. J.; Pidacks, C.; Weiss, M. J. J. Org. Chem. 1971, 36, 279–
284.
brine, dried (Na
2
SO
4
), filtered and concentrated in vacuo.
27. Birch, A. J.; Nasipuri, D. Tetrahedron 1959, 6, 148–153.