Di-2-methoxyethyl azodicarboxylate (DMEAD): An inexpensive and separation-friendly alternative reagent for the mitsunobu reaction

Article abstract of DOI:10.1246/cl.2007.566

Di-2-methoxyethyl azodicarboxylate (DMEAD) was prepared as an alternative of DEAD or DIAD for the Mitsunobu reaction. Removal of the hydrazinedicarboxylate generated from DMEAD becomes much easier owing to the polar and water-soluble property. Copyright

Full text of DOI:10.1246/cl.2007.566

566
Chemistry Letters Vol.36, No.4 (2007)
Di-2-methoxyethyl Azodicarboxylate (DMEAD):
An Inexpensive and Separation-friendly Alternative Reagent
for the Mitsunobu Reaction
Takashi Sugimuraà and Kazutake Hagiya1
Graduate School of Material Science, Himeji Institute of Technology, University of Hyogo,
Kohto, Kamigori, Ako-gun, Hyogo 678-1297
(Received February 1, 2007; CL-070124; E-mail: sugimura@sci.u-hyogo.ac.jp)
Di-2-methoxyethyl azodicarboxylate (DMEAD) was pre-
O
pared as an alternative of DEAD or DIAD for the Mitsunobu
reaction. Removal of the hydrazinedicarboxylate generated
from DMEAD becomes much easier owing to the polar and
water-soluble property.
O
H2N−NH2·H2O
+ 2
O
Cl
(72%)
O
O
NBS
(88%)
O
O
O
N
N
O
H
H
The Mitsunobu reaction is popular in organic synthesis to
introduce an acidic nucleophile (carboxylic acid, phenol, imide,
etc.) to an alcoholic function under mild conditions.2 The
notable property of this reaction is complete inversion of
stereochemistry at the alcoholic function. The original procedure
consists of mixing an alcohol and a nucleophile with diethyl
azodicarboxylate (DEAD) and triphenylphosphane at ambient
temperature.3 This oxidant–reductant combination to cause de-
hydration in the Mitsunobu reaction is still widely used although
diisopropyl azodicarboxylate (DIAD) is also conveniently em-
ployed as a more stable analogue of DEAD. A major drawback
of this process is formation of two co-products, hydrazinedicar-
boxylate and phosphane oxide, which must be removed from the
reaction mixture to isolate a target compound. Many efforts have
been undertaken to solve this issue: e.g., one of the reagents is
supported by polymer or solid to perform the separation by filtra-
tion, or is attached to an acidic, basic, or fluorous function to al-
low the separation by extraction into a basic or acidic aqueous
layer or into fluorous solvent. These modifications of the original
oxidant or reductant are considered to intend a separation-free or
separation-friendly process.4,5
Formation of triphenylphosphane oxide is practically insig-
nificant matter in the separation process due to its facile crystal-
lization in non-polar solvents. In contrast, separation of the other
co-product, diethyl or diisopropyl hydrazinedicarboxylate re-
quires highly capable column chromatography to isolate a target
compound. One also often faces a problem that several side
products of the hydrazinedicarboxylate cannot be removed com-
pletely from the product. In situ recycling of DEAD that makes it
a catalyst is one of the solutions.6 Total removal of the hydra-
zinedicarboxylate and the side-products can be performed by
polymerization of norbornene units incorporated in the reagent
as to be called impurity annihilation.7 Very recently, di-4-
chlorobenzyl analogue was reported,8 where the hydrazinedicar-
boxylate can be removed by the crystallization. We have made
another approach to this issue by molecular design to enable
the separation of a hydrazinedicarboxylate by extraction with
neutral water. The preparation cost and handling were also
considered to compete with DEAD and DIAD in the commercial
source. Di-2-methoxyethyl azodicarboxylate (DMEAD) is a new
reagent, which will be detailed in this report.
1
O
O
O
O
O
N
N
O
Di-2-methoxyethyl azodicarboxylate
(DMEAD)
Scheme 1. Synthesis of DMEAD.
Di-2-dimethoxyethyl hydrazinedicarboxylate (1) was pre-
pared by mixing hydrazine hydrate and 2-methoxyethyl chloro-
formate in ethanol in the presence of sodium carbonate (72%
yield after recrystallization, Scheme 1).9,10 Solubility of 1 is
good in chloroform (0.25 g/mL) and ethyl acetate (0.1), but very
poor (<0:01) in less polar ether and toluene. In contrast, 1 is well
soluble in water (0.55 g/mL). The solubility is better than
those of the methyl analogue (0.15) and the ethyl or isopropyl
analogue (<0:01). Since 1 is not only a precursor of DMEAD
but also a co-product in the Mitsunobu reaction, the observed
solubility supports our molecular design intended to separate 1
after the Mitsunobu reaction by extraction with neutral water.
After several trials of oxidation of 1 under different
conditions,9 NBS in toluene11 was found to result in the best re-
sults to give DMEAD. Extraction and single recrystallization
(toluene/hexane) of the reaction mixture gave essentially pure
DMEAD in 88% yield.12,13 It should be noted that distillation
required for preparation of DEAD and DIAD can be omitted
in the case of DMEAD.
OBz
OH
a
b
c
(83%, > 99% pure)
+ PhCOOH (1.1 equiv.)
+ PPh3 (1.1 equiv.)
+ DMEAD (1.1 equiv.)
in THF
POPh3 (1.1 equiv., 90% pure)
1 (0.95 equiv., > 97% pure)
Scheme 2. A procedure for the Mitsunobu reaction: (a) Extrac-
tion with toluene and water, and aqueous layer was concentrated;
(b) Toluene layer was dried, concentrated, suspended in hexane,
and filtered; (c) Filtrate was purified by a silica-gel column.
Copyright Ó 2007 The Chemical Society of Japan

Chemistry Letters Vol.36, No.4 (2007)
567
Table 1. Product yieldsa and purities in the reaction of (R,R)-
2,4-pentanediol with varied nucleophiles
References and Notes
1
2
Graduate student from Toyo Kasei Kogyo, Co. Ltd.
O. Mitsunobu, Synthesis 1981, 1; D. L. Hughes, Org. React.
1992, 42, 335.
DMEAD or DIAD
+ NuH
PPh3
OH Nu
OH OH
3
4
O. Mitsunobu, M. Yamada, Bull. Chm. Soc. Jpn. 1967, 40, 2380.
For reviews, see: R. Dembinski, Eur. J. Org. Chem. 2004, 2763;
S. Dandapani, D. P. Curran, Chem. Eur. J. 2004, 10, 3130.
J. C. Poupon, A. A. Boezio, A. B. Charette, Angew. Chem., Int.
Ed. 2006, 45, 1415.
DMEAD
DIAD
Nucleophile (pKa)
Yield Purity Yield Purity
/%b
5
/%
/%
/%
Benzoic acid (4.20)
74
79
84
82
83
100
100
100
100
100
74
83
80
83
91
72
64
6
7
T. Y. S. But, P. H. Toy, J. Am. Chem. Soc. 2006, 128, 9636.
A. G. M. Barrett, R. S. Roberts, J. Schroder, Org. Lett. 2000, 2,
2999.
B. H. Lipshutz, D. W. Chung, B. Rich, R. Corral, Org. Lett.
2006, 8, 5069.
4-Nitrobenzoic acid (3.44)
Phenol (9.95)
100
77
8
9
4-Methoxyphenol (10.20)
4-Methoxycarbonylphenol (8.74)
66
N. Rabjohn, Org. Synth. 1955, Coll. Vol. 3, 375; J. C. Kauer,
Org. Synth. 1963, Coll. Vol. 4, 411.
aThe yields are based on the amounts of the nucleophiles. Diaster-
eomeric purities were confirmed to be >98% by NMR. bThe most
impurity was diisopropyl hydrazinedicarboxylate.
10 A solution of hydrazine hydrate (15 g, 300 mmol) in 99.5%
ethanol (75 mL) was cooled to 5 ꢀC with an ice–water bath. 2-
Methoxyethyl chloroformate (41.52 g, 300 mmol) was added
dropwise in keeping the temperature below 20 ꢀC. After 5 min,
2-methoxyethyl chloroformate (41.52 g, 300 mmol) was added
simultaneously with a solution of sodium carbonate (31.76 g,
300 mmol) in water (120 mL) below 20 ꢀC. After 1 h, the
mixture was concentrated under vacuum, and then treated with
acetone (100 mL) to precipitate inorganic salts. Filtration, con-
centration, and purification by recrystallization with a mixture
of acetone (75 mL) and toluene (120 mL) gave 50.91 g of a
colorless solid (1, 71.9% yield). mp 71.3–76.5 ꢀC; Anal. Calcd
for C8H16N2O6: C, 40.68; H, 6.83; N, 11.96%. Found: C,
Easy isolation of the Mitsunobu product was demonstrated
with recovery of both of the co-products by using DMEAD.
(S)-2-Octanol (2 g) was converted to its benzoate ester with
1.1 equiv. of reagents, benzoic acid, triphenylphosphane, and
DMEAD, in THF at room temperature (Scheme 2). After con-
centration, the mixture was dissolved in toluene and washed with
water. The water layer was re-extracted with toluene once more.
The aqueous layer contained only 1, and the purity determined
1
by the H NMR after concentration was >98%. The yield of 1
based on the employed amount (1.1 equiv.) of DMEAD was
86%, which is 95% of the theoretical value. The organic layer
was dried, concentrated, and suspended in hexane. Triphenyl-
phosphane remained in part was consumed during the process
and almost all the phosphane oxide was recovered as crystalline
by the filtration. Some side products produced from DMEAD
were included in the crystalline part. The filtrate was concentrat-
ed and purified by a silica-gel column to give the (R)-benzoate
product in 83% yield. Required amount of silica gel for the
separation was less than a half of that in the usual process
with DEAD or DIAD. Inversion of the stereochemistry was
confirmed by the optical rotation.
The reactivity of DMEAD in the Mitsunobu reaction was
further demonstrated with (R,R)-2,4-pentanediol, where excess
use of the reagents is not allowed to obtain a singly modified
product. Table 1 summarizes isolated yields of the product with
different nucleophiles by using DMEAD in addition to the same
reactions with DIAD as reference.14 The isolated yields were
very similar between DMEAD and DIAD to conclude that they
have no difference in the reactivity in the Mitsunobu reaction.
The product purity was not always high owing to the difficulty
in the chromatographic separation of the hydrazine analogue
when DIAD was employed. In contrast, complete separation to
give the pure products was achieved in the reactions with
DMEAD owing to the high polarity of 1 and its analogues as side
products.15
40.88; H, 7.49; N, 12.07%. IR (KBr) 1755 cmÀ1 1H NMR
;
(400 MHz, CDCl3) ꢀ 7.02 (brs, 2H), 4.27–4.25 (m, 4H), 3.58–
3.56 (m, 4H), 3.49 (s, 6H); 13C NMR (100 MHz, CDCl3) ꢀ
156.59, 70.40, 64.68, 58.69.
11 L. A. Carpino, P. J. Crowley, Org. Synth. 1973, Coll. Vol. 5,
160.
12 To a solution of 1 (45 g, 190.5 mmol) and pyridine (15.09 g, 1.0
equiv.) in toluene (450 mL) was added N-bromosuccinimide
(37.31 g, 1.1 equiv.) slowly at room temperature. After vigorous
stirring for 2 h, the reaction mixture was washed with water
(180 mL Â 2), dried over magnesium sulfate, concentrated
under vacuum, and then purified by recrystallization with a
mixture of toluene (67.5 mL) and hexane (337.5 mL) to give
39.26 g of DMEAD as yellow prisms (88.0% yield). mp 39.9–
40.4 ꢀC; Anal. Calcd for C8H14N2O6: C, 41.03; H, 6.03; N,
11.96%. Found: C, 41.09; H, 6.65; N, 11.98%. IR (KBr)
1
1782 cmÀ1; H NMR (400 MHz, CDCl3) ꢀ 4.51–4.49 (m, 4H),
3.66–3.63 (m, 4H), 3.32 (s, 6H); 13C NMR (100 MHz, CDCl3)
ꢀ 160.04, 69.45, 67.85, 58.84.
13 Decomposition temperature observed by DSC for DMEAD was
210 ꢀC, which is somewhat lower than that of DEAD (227 ꢀC by
our measurement).
14 A ratio of (R,R)-2,4-pentanediol, nucleophile, triphenylphos-
phane, and DMEAD (DIAD) is fixed to be 1/0.85/1/1. All
the reactions were carried out at room temperature by addition
of one of the azodicarboxylates to a mixture of the other three,
except for the case with 4-nitorobenzoic acid that was added
at last. The product was isolated by extraction and silica-gel
column chromatography.
15 The efficiency of the product isolation depends on the polarity of
the hydrazine analogues. The R f value on TLC (SiO2, elution
with a mixture of hexane and ethyl acetate = 1/1) was 0.65
for diisopropyl analogue (0.44 for diethyl), while very small
0.08 for 1.
In the present study, we have shown that 1 formed as a
co-product in the Mitsunobu reaction with DMEAD could easily
be removed by simple extraction with neutral water, and that
DMEAD can be a separation-friendly alternative of DEAD in
the Mitsunobu reaction. The production cost of DMEAD is
clearly less than the other alternatives so far reported to solve
the separation problem, and the preparation is even easier than
that of DEAD or DIAD.
Published on the web (Advance View) March 27, 2007; doi:10.1246/cl.2007.566