0-amine-assisted cannizzaro reaction of glyoxal with new 2,6-diaminoanilines

Article abstract of DOI:10.1002/ejoc.200900079

The preparation of several new o-amine-substituted anilines was achieved according to a new bifunctional molecular design, and their reactions with glyoxal were conducted. Can- nizzaro reactions of glyoxal proceeded using specifically designed anilines, such as 2,6-dipyrrolidinyl-, 2,6-dipiperidinyl-, 2,6-dimorpholinyl-, and 2-pyrrolidinyl-aniline, which are new and can easily be synthesized by substitution of halogen-substituted nitrobenzene with amines and subsequent reduction with hydrogen, to form a-hydroxy acetamide and a-amino acetamide derivatives, as a result of the Cannizzaro reaction. In comparison with the reaction of glyoxal with p- pyrrolidinylaniline to form a common diimine product, the reaction with o-pyrrolidinylaniline leads only to a-hydroxy amides, strongly suggesting that the abnormal Cannizzaro reactions are attributed to the existence of basic nitrogen atoms at the o-positions, which suppress diimine formation and assist the generation of acetamides.

Full text of DOI:10.1002/ejoc.200900079

FULL PAPER
DOI: 10.1002/ejoc.200900079
o-Amine-Assisted Cannizzaro Reaction of Glyoxal with New
2,6-Diaminoanilines
Yuhki Irie,[a] Yuji Koga,[a] Taisuke Matsumoto,[b] and Kouki Matsubara*[a]
Keywords: Cannizzaro reaction / Amination / Neighbouring group assistance / Nucleophilic substitution
The preparation of several new o-amine-substituted anilines
reaction. In comparison with the reaction of glyoxal with p-
was achieved according to a new bifunctional molecular de-
pyrrolidinylaniline to form a common diimine product, the
sign, and their reactions with glyoxal were conducted. Can- reaction with o-pyrrolidinylaniline leads only to α-hydroxy
nizzaro reactions of glyoxal proceeded using specifically de- amides, strongly suggesting that the abnormal Cannizzaro
signed anilines, such as 2,6-dipyrrolidinyl-, 2,6-dipiperidinyl-, reactions are attributed to the existence of basic nitrogen
2,6-dimorpholinyl-, and 2-pyrrolidinyl-aniline, which are
new and can easily be synthesized by substitution of halo-
gen-substituted nitrobenzene with amines and subsequent
reduction with hydrogen, to form α-hydroxy acetamide and
α-amino acetamide derivatives, as a result of the Cannizzaro
atoms at the o-positions, which suppress diimine formation
and assist the generation of acetamides.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
Introduction
Reactions of aromatic amines are well known as the
nucleophilic addition and substitution of olefins, aromatic
compounds, and carbonyl compounds to form N–C bond-
forming products, such as alkyl and arylamines, imines, and
amides.[1] However, studies on versatile, substituted primary
anilines at the aromatic ring are not so numerous. For ex-
ample, the substituents introduced into anilines were iso-
propyl groups, aromatic groups,[2] and alkyne groups[3] at
the o-positions, which were used as building units of specific
metal ligands for sterically demanding metal catalysts.
We thought of a bifunctional molecule, which was in-
spired from the tagged molecules in organic synthesis[4] and
was designed in order to expand the variation of the aniline
reactions. That is, the molecule having a catalyst group at
the neighbouring position of the reactant group, –NH2, as
shown in Figure 1. Catalysts close to the reactant group
could perform specific reactions to yield other products
than those generally expected in the standard reactions of
common anilines. In this study, we applied this concept to
a series of anilines first having tertiary or secondary amines
at the o-positions, in which the o-amine could act as a base
in the nucleophilic reactions of anilines.
Figure 1. A designed model for a bifunctional molecule.
A series of o-monoamine-substituted anilines have al-
ready been reported.[5] However, although some primary
anilines having amines at the 2- and 6-positions have been
synthesized until now,[6] difficulties in the synthesis are un-
known to prepare various new 2,6-diaminoanilines. This
prompted us to prepare several new 2,6-substituted anilines
having secondary or primary amino substituents. After sev-
eral attempts, we found an efficient synthetic route to the
anilines via double amination of 2,6-dihalonitrobenzene[7]
and subsequent reduction to achieve the preparation of 2,6-
diaminoanilines [Scheme 1 (a)]. Furthermore, we found that
reactions of the anilines with glyoxal formed not diimine
but α-hydroxy-acetamides as a result of the Cannizzaro re-
action, which was induced by successful assistance the o-
amine substituents [Scheme 1 (b)].
The Cannizzaro reaction is a hydride transfer from alde-
hyde,[8] which is added by a nucleophile, to another car-
bonyl carbon under a basic or acidic condition to form a
carboxylic acid derivative and alcohol in an intermolecular
version[9] or α-hydroxy-carboxylic acid derivatives in an in-
tramolecular version.[10–12] Several studies on intramolecu-
lar Cannizzaro reactions have been reported, such as reac-
[a] Department of Chemistry, Faculty of Science, Fukuoka Univer-
sity,
Nanakuma, Jonan-ku, Fukuoka 814-0180, Japan
Fax: +81-92-871-6631
E-mail: kmatsuba@fukuoka-u.ac.jp
[b] Analysis Center, Advanced Institute for Materials Chemistry
and Engineering, Kyushu University,
Kasuga, Fukuoka, Japan
Supporting information for this article is available on the
WWW under http://www.eurjoc.org/ or from the author.
Eur. J. Org. Chem. 2009, 2243–2250
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2243

Y. Irie, Y. Koga, T. Matsumoto, K. Matsubara
FULL PAPER
base (Cs2CO3) in toluene, whereas these halogens were also
substituted by nucleophilic substitution of electron-poor
haloarene with amine as shown in Method B. In the synthe-
sis of pyrrolidine-substituted amine, substitution of 2,6-di-
bromo-1-nitrobenzene via Method A was completed in 36
h using the Pd compound (2 mol-%) and the ligand (4 mol-
%), whereas its chloride analog was substituted in 96 h
using the Pd catalyst (5 mol-%) and the ligand (7.5 mol-%),
to form 1 in high yields in both cases. When these reactions
were quenched after 12 h, the substitution conversions from
2,6-dichloro-1-nitrobenzene in Methods A and B were 30
and 32%, respectively, indicating that Method B is better
than Method A for obtaining 2,6-diamino-1-nitrobenzene
(1). So, we prepared some 2,6-amine-substituted nitroben-
zene derivatives with secondary amines almost quantita-
tively from 2,6-dichloro-1-nitrobenzene via Method B
(Table 1, entries 1–3), but primary anilines did not react
with 2,6-dichloro-1-nitrobenzene in Method B. Diamin-
ation of 2,6-dichloronitrobenzene with anilines was
achieved via Method A using palladium acetate and an N-
heterocyclic carbene ligand, bis(2,6-diisopropylphenyl)imid-
azol-2-ylidene (IPr). 1d, 1e, and 1f were formed after isola-
tion by silica gel column chromatography (entries 4–6) in
72, 74, and 56% yield, respectively, higher than those using
BINAP. The palladium-catalyzed substitution also formed
monoamine-substituted products 2 in 48 h in low yields, 6–
17%. The nitrobenzene derivatives 1a–1d were fully iden-
Scheme 1. Synthetic route for (a) 2,6-diaminoaniline and (b) α-hy-
droxy amide.
tions of glyoxal derivatives with alkoxy[10] and amide
anions[11] under basic conditions, and alcohol in the pres-
ence of Lewis acids.[12]
The products, α-hydroxy amide derivatives, are com-
monly known as the significant core structure for anticon-
vulsion drugs[13] and fundamental components included in
other pharmaceutical compounds[14] and agrochemicals,[15]
and also significant synthetic precursors of biologically
active substances.[16] Thus, several synthetic routes to α-hy-
droxy amides have been developed.[17] This is the first intra-
molecular Cannizzaro reaction of glyoxal with amine in the
absence of strong bases and Lewis acids to form α-hydroxy
amide derivatives, showing the specific reactivity of novel
o-amine-substituted anilines.
1
tified by means of H and 13C NMR and IR spectroscopy
and elemental analysis.
Table 1. Amination of 2,6-dichloro-1-nitrobenzene.
Results and Discussion
Synthesis of 2,6-Diaminoanilines
We explored two synthetic routes for the novel 2,6-
amine-substituted aniline: one is that 2,6-dibromoaniline is
acetylated by acetyl chloride, bromides at the 2,6-positions
are substituted with amines by a palladium catalyst, and
then the acetyl unit is deprotected to form the desired prod-
uct; the other is that 2,6-dihalonitrobenzene is substituted
with amines at the 2,6-positions and subsequently, the nitro
group is reduced. As a result, while (2,6-dibromophenyl)-
acetamide, which was obtained from 2,6-dibromoaniline
and acetyl chloride in 69% yield, could not react with
amines, 2,6-dihalonitrobenzene was successfully substituted
with cyclic amines to form 2,6-diaminonitrobenzene in high
yields and subsequent reduction formed 2,6-diaminoaniline
in excellent yields. The conditions of the amination pro-
cesses are shown below as Methods A and B:
Entry
Method
–NR1R2
Time [h]
% Yield[a]
1
2
1
2
3
4
5
6
B
B
B
a
b
c
d
e
f
120
216
72
48
48
97
99
85
72
74
57
0
0
0
14
6
17
[b]
AЈ
[b]
AЈ
[c]
AЈЈ
48
[a] The products were isolated by silica gel column chromatography.
[b] Method AЈ: Pd(OAc)2 (3.0 mol-%) and IPr (4.5 mol-%) were
used. [c] Method AЈЈ: Pd(OAc)2 (4.0 mol-%) and IPr (8.0 mol-%)
were used.
Method A: Amine (3.0 equiv.), Pd(OAc)2, BINAP, Cs2CO3
(2.8 equiv.), and toluene were added and the mixture was
heated at 100 °C.
Method B: Amine (30.0 equiv.) was added and the mixture
was refluxed without adding solvents.
Subsequently, the products, 2,6-diamino-1-nitrobenzene
In Method A, halogen atoms in 2,6-dihalonitrobenzene (1), were reduced by hydrogen in the presence of Pd/C (Pd
were substituted with amine in the presence of a palladium content: 5%) in ethanol to form 2,6-diaminoaniline (3). The
catalyst [Pd(OAc)2 (2 mol-%) and BINAP (4 mol-%)] and a reduction of 1a–1c having the cyclic amino substituents
2244
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2243–2250

Intramolecular Cannizzaro Reaction of Glyoxal with Anilines
proceeded quantitatively to yield a series of 2,6-diaminoani- Table 3. Reactions of glyoxal with 2,6-diamino- or 2-aminoanilines
3 (the products were isolated by silica gel column chromatography).
lines 3a–3c in excellent yields (Table 2, entries 1–3). In the
reduction of 1d having anilines at the 2,6-positions, an ani-
line derivative 3d was generated quantitatively (entry 4), but
anilines having bulky substituents, such as 2,6-bis(2,6-diiso-
propylanilino)nitrobenzene (3e) and 2,6-bis(2,4,6-trimeth-
ylanilino)nitrobenzene (3f), could hardly be reduced (en-
tries 5–6).
Table 2. Reduction of 2,6-diamino-1-nitrobenzene (1).
Entry
–NR1R2
mol-% Pd/C
Time [h]
% Yield[a]
1
2
3
4
5
6
a
b
c
d
e
f
5
2.5
5
10
50
50
12
14
5
5
12
12
88
98
94
89
4 (30)[b]
– (70)[b]
[a] The products were isolated by silica gel column chromatography.
[b] Isolated yields of starting material 1.
The aniline derivatives 3a–3e were also fully charac-
terized. The IR resonances due to the NH2 stretching vi-
bration of the primary anilines 3a–3e appear at 3300–
3450 cm–1. The existence of the primary amino groups was
1
revealed by the observation of the H resonances at δ =
4.02, 4.26, and 4.27 in the spectra for 3a, 3b, and 3c and at
δ = 3.90 and 3.73 ppm for 3d and 3e due to the amino
1
protons. In the H and 13C NMR spectra for 3a–3c similar
characteristic sets of high-field signals from δH = 6.68 to
6.92 and δC = 112–141 ppm appear due to the amine-substi-
tuted aromatic groups.
The influence of the number of amine substituents in the
aniline on the Cannizzaro reaction was also studied. Inter-
estingly, as shown in entry 5 of Table 3, o-pyrrolidinylani-
line 3g also reacts with glyoxal to form α-hydroxy amide 4g
in low yield (15%), but does not form imine derivatives.
Compound 4g was characterized by NMR and IR spec-
troscopy and elemental analysis. IR signals due to the car-
Reactions of 2,6-Diaminoanilines with Glyoxal
The new series of aniline derivatives 3a–3d was treated bonyl CO stretching band and NH and OH stretching
with glyoxal. We found that no imines were formed, instead broad bands were detected at 1670 cm–1 and around
the new α-hydroxy amides 4a and 4b were generated from 3300 cm–1, respectively.
3a and 3b (Table 3, entries 1–2). Interestingly, from 3c, the
Several studies on nucleophilic substitution of glyoxal
α-amino carboxamide 4c was obtained (entry 3). The yields with 2,6-substituted anilines to form only diimines have
depend on the amount of glyoxal: when the amount of gly- been reported previously.[2–3] So, we attribute these abnor-
oxal was increased, the yield of the products also increased mal reactions to some specific reactivity of the o-amine-
within 24 h. Generation of 4b was almost quantitatively, substituted anilines as mentioned in the Introduction of this
whereas the starting materials were also obtained in the re- article. We were interested to see which rôle tertiary alk-
action of 3a and 3c, probably because of the lower reactivity ylamines like 3a–c and 3g could play in the Cannizzaro re-
of 3a and 3c than that of 3b. From the crystal structures action. Triethylamine was added to the reaction mixture of
of the products and some experiments as shown below, we glyoxal with a 2,6-dialkylaniline, e.g. 2,6-diisopropylaniline.
conclude that a Cannizzaro reaction proceeds to form 4a– In the presence of triethylamine (2 equiv. to aniline), the
4c. Aniline-substituted aniline 3d also reacts with glyoxal to reaction of glyoxal with the aniline yields diimine com-
form the imidazole 5 as shown in Table 3, entry 4.
pound in high yield as shown in Scheme 2, no α-hydroxy
Eur. J. Org. Chem. 2009, 2243–2250
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2245

Y. Irie, Y. Koga, T. Matsumoto, K. Matsubara
FULL PAPER
amide is formed.[18] We also investigated the effect of the lengths were 1.233(2) and 1.229(2) Å] but a single C–O
substitution position of the amino substituents. When a 4- bond length; consequently, an alcohol was generated in
aminoaniline, for instance 4-pyrrolidinylaniline 3h, was these reactions. Also, the C2–N4 bond lengths in 4c are
added to glyoxal, a normal diimine 6 was obtained in 98% 1.436(3) Å, suggesting that a C–N single bond was formed
yield and no formation of α-hydroxy amide was detected as a result of hydride rearrangement. On the other hand, it
(Scheme 3), suggesting that the presence of basic amines in was found that the reaction of 2,6-dianilinoaniline 3d with
the neighbourhood of the nucleophilic nitrogen in 3a–c and glyoxal did not yield α-hydroxy amide or diimine but the
3g is essential for the Cannizzaro reaction. We tested reac imidazole derivative 5, as shown in part (d) of Figure 2.
tions of 1,2-diones, such as phenylglyoxal and 1,2-propane-
dione, with the anilines 3 but, in all cases, we obtained start-
ing materials mostly and, unfortunately, hydroxy amides
and imines were not obtained, probably due to steric factors
Mechanistic Considerations
The above results demonstrated that the presence of terti-
ary amines close to the NH2 moiety of aniline in 3 clearly
affects the reactivity of 3 with glyoxal so as not to yield
diimines, interestingly. As discussed in the literature on
Cannizzaro reactions,[9] we investigated whether hydride
transfer occurred or not in the formation of the amide com-
pounds 4. When [D4]methanol was used as a solvent in the
reaction of 3b with glyoxal, the H and H NMR spectro-
scopic analysis of the isolated α-hydroxy amide 4b revealed
that deuterium was not incorporated (Ͻ 10%), at least not
within 1 h at 60 °C (Scheme 4).[20] This shows that no pro-
ton exchange between hydrogen atoms derived from glyoxal
and the solvent occurs. So the process does not proceed
through an ene-diol intermediate, which is proposed in the
Lobry de Bruyn-Alberda van Ekenstein reaction.[21]
and/or low basicity of o-amines in the anilines.
1
2
Scheme 2. The reaction of glyoxal with 2,6-diisopropylaniline in the
presence of triethylamine.
Scheme 3. The reaction of glyoxal with 4-pyrrolidinylaniline 3h.
Although an intermolecular hydride-transfer mechanism
cannot be ruled out, this process might be disfavoured, be-
cause, at present, we cannot find factors inhibiting the in-
tramolecular hydride transfer. The proposed intramolecular
reaction mechanism is depicted in Scheme 5. In the tetrahe-
X-ray Crystal Structures of 4a, 4c, 4g, and 5
Finally, the crystal structures of products 4a, 4c, 4g, and dral intermediate having a hydroxy group, a basic lone pair
5 obtained from anilines 3 with glyoxal, α-hydroxy amide, of the nitrogen of o-substituted amines may easily interact
α-amino amide, and imidazole derivatives were determined with this hydroxy group and abstract a proton to induce
by single-crystal X-ray diffraction studies.[19] The structures hydride transfer from the tetrahederal carbon to the adja-
of these compounds are depicted in Figure 2. Table 4 lists cent carbonyl carbon to form a zwitter-ionic amide. In
the representative bond lengths and angles of these com- other words, the hydride transfer in the Cannizzaro reaction
pounds. The C2–O2 bond lengths of α-hydroxy amides 4a may be assisted by the neighbouring basic amine. Such as-
and 4g are 1.409(2) and 1.417(2) Å, respectively, and thus sistance of amines in the hydride rearrangement has never
do not correspond to a double bond [cf. C1–O1 bond been reported before to the best of our knowledge.
Table 4. Representative bond lengths [Å] and angles [°] of 4a, 4c, 4g, and 5.
4a
4c
4g
5
Bond lengths
N1–C1
N1–C3
C1–O1
C1–C2
C2–O2
1.336(2)
1.432(2)
1.233(2)
1.519(2)
1.409(2)
N1–C1
N1–C3
C1–O1
C1–C2
C2–N4
1.364(3)
1.433(3)
1.225(3)
1.517(3)
1.436(3)
N1–C1
N1–C3
C1–O1
C1–C2
C2–O2
1.339(2)
1.415(2)
1.229(2)
1.517(2)
1.417(2)
N1–C1
N1–C2
N2–C3
N2–C8
N3–C1
N3–C14
1.310(2)
1.394(2)
1.393(2)
1.393(2)
1.376(2)
1.422(2)
Bong angles
C1–N1–C3
N1–C1–O1
N1–C1–C2
O1–C1–C2
C1–C2–O2
125.10(14)
124.82(15)
115.85(14)
119.33(15)
114.16(14)
C1–N1–C3
N1–C1–O1
N1–C1–C2
O1–C1–C2
C1–C2–N4
C2–N4–C17
123.59(19)
121.4(2)
116.66(17)
121.63(19)
108.69(17)
125.93(18)
C1–N1–C3
N1–C1–O1
N1–C1–C2
O1–C1–C2
C1–C2–O2
127.93(14)
126.01(15)
114.78(14)
119.21(15)
112.38(14)
N1–C1–N3
N1–C2–C3
C1–N1–C2
C1–N3–C7
C3–N2–C8
114.22(16)
127.93(16)
103.93(15)
106.04(14)
127.93(16)
2246
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2243–2250

Intramolecular Cannizzaro Reaction of Glyoxal with Anilines
Figure 2. Crystal structures of (a) 4a, (b) 4c, (c) 4g, and (d) 5 (50% probability thermal ellipsoids).
Scheme 4. The reaction of glyoxal with 2,6-dipyrrolidinylaniline in
the presence of [D4]methanol.
Scheme 6. Proposed mechanism for the formation of 5.
The relationship between the products, α-hydroxy amide
and α-amino amide, with the basicity of o-substituted
amines like pyrrolidine, piperidine, and morpholine may
also support the above mechanism. The reactions of glyoxal
with the strongly basic amines 3a and 3b containing pyrrol-
idine (pKb = 2.9) and piperidine (pKb = 2.8) form the α-
hydroxy amides 4a and 4b, whereas less basic amines like
morpholine (pKb = 5.6) in 3c provid the α-amino amide 4c,
which may be generated via imine formation at one of two
carbonyl groups in glyoxal. A tetrahedral intermediate
Scheme 5. Proposed mechanism for o-amine-assisted intramolecu-
lar Cannizzaro reaction.
Although decarboxylation of α-keto carboxylic acid de- formed by the addition of a second aniline molecule poss-
rivatives with o-phenylenediamine to form benzimidazole ibly induces the formation of 4c by intramolecular Canniz-
compounds is known,[22] such benzimidazole derivatives zaro reaction, as a result of the increased acidity of the hy-
have not been formed by the condensation using α-keto car- droxy proton to form.
bonyl compounds like glyoxal. The proposed mechanism
The reaction true mechanism for the formation of com-
for the formation of 5 is shown in Scheme 6. After the elec- pound 5 may be more complex, and several plausible reac-
trophilic imine is attacked by the o-substituted amine, de- tion mechanisms can be proposed. We assume that the first
formylation might proceed to form 5.
imine formation and then intramolecular nucleophilic at-
Eur. J. Org. Chem. 2009, 2243–2250
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2247

Y. Irie, Y. Koga, T. Matsumoto, K. Matsubara
FULL PAPER
1b: Yellow solid, yield 5.75 g, 99.5%. C16H23N3O2 (289.4): calcd.
C 66.41, H 8.01, N 14.52; found C 66.39, H 7.98, N 14.25. 1H
NMR (CDCl3): δ = 7.30 (t, J = 8.1 Hz, 1 H), 6.92 (d, J = 8.1 Hz,
2 H), 2.91–2.88 (m, 8 H), 1.66–1.61 (m, 8 H), 1.53–1.47 (m, 4 H)
ppm. 13C NMR (CDCl3): δ = 146.74, 145.74, 130.44, 117.30, 54.38,
tack of the secondary o-amines to the imine carbon might
proceed to form an imidazole compound, accompanied by
elimination of the other carbonyl group.
26.35, 24.08 ppm. IR (neat): ν = 1537 (N=O) cm–1.
˜
Conclusions
1c: Pale-yellow solid, yield 4.97 g, 84.5%. C14H19N3O4 (293.32):
calcd. C 57.33, H 6.53, N 14.33; found C 57.28, H 6.39, N 14.28.
1H NMR (CDCl3): δ = 7.40 (t, J = 8.1 Hz, 1 H), 7.03 (d, J =
8.2 Hz, 2 H), 3.77 (t, J = 4.5 Hz, 8 H), 2.97 (m, J = 4.6 Hz, 8 H)
ppm. 13C NMR (CDCl3): δ = 146.11, 145.16, 131.01, 118.24, 67.15,
A series of new anilines containing cyclic amine or ani-
line moieties at the 2- and/or 6-positions was synthesized
by a convenient procedure. Nucleophilic substitution or
palladium-catalyzed amination of 2,6-dihalonitrobenzene
with a cyclic or aromatic amine results in the formation of
2,6-diaminonitrobenzene 1 in high yields. Subsequent re-
duction of 1 with Pd/C and hydrogen affords the anilines 3
in excellent yields. These anilines have a special reactivity
towards glyoxal to form α-hydroxy amide, α-amino amide,
and imidazole derivatives, the structures of which were de-
termined by spectroscopic methods and X-ray crystallogra-
phy. The reactions yielding amide compounds are the first
examples of intramolecular Cannizzaro reactions with ani-
lines where no addition of Lewis acids or strong bases is
needed. Obviously the reaction is significantly assisted by
the basic o-substituted amines. Other 1,2-dicarbonyl com-
pounds do not react with these anilines. We are now going
to expand the above bifunctional system to develop various
new organic reactions.
53.15 ppm. IR (neat): ν = 1531 (N=O), 1110 (C–O) cm–1.
˜
The synthetic method of 1g and 1h was similar to that for 1a–1c.
In these cases, the reaction mixtures were stirred refluxed (at
108 °C) for 2 h to yield 1g as an orange solid (yield 7.54 g, 98%)
and 1h as a yellow solid (yield 1.89 g, 98%).
1g: 1H NMR (CDCl3): δ = 7.74 (dd, J = 8.2, 1.6 Hz, 1 H), 7.36
(dt, J = 8.7, 1.7 Hz, 1 H), 6.90 (dd, J = 8.6, 1.0 Hz, 1 H), 6.71 (dt,
J = 8.2, 1.1 Hz, 1 H), 3.23–3.20 (m, 4 H), 2.00–1.97 (m, 4 H) ppm.
1h: 1H NMR (CDCl3): δ = 8.12 (dd, J = 9.3, 1.9 Hz, 2 H), 6.47
(dd, J = 9.2, 1.0 Hz, 2 H), 3.40 (t, J = 6.5 Hz, 4 H), 2.09–2.06 (m,
4 H) ppm.
Preparation of 2,6-Dianilinonitrobenzenes 1d–1f: In a typical exam-
ple, to a 100 mL Schlenk flask were added 2,6-dichloro nitroben-
zene (3.00 g, 15.6 mmol), IPr (273 mg, 0.702 mmol), Pd(OAc)2
(105 mg, 0.468 mmol), Cs2CO3 (14.2 g, 43.7 mmol), aniline
(4.3 mL, 46.8 mmol), and toluene (32 mL), and the mixture was
stirred at 100 °C for 48 h. After the resulting suspension was cooled
to room temperature, the precipitate in the reaction mixture was
removed by filtration and washing with CH2Cl2. After the solvent
was evaporated, the residual black solid was dissolved in CH2Cl2
and crystallized. The crystals were washed with cold methanol to
yield 1d as black crystals (yield 3.24 g, 68%). As the filtered solu-
tion contained 1d and aniline, the solvent was evapolated and, sub-
sequently, aniline was removed by heating the mixture at 100 °C to
give a black solid. The solid mixture was separated by silica gel
column chromatography to yield disubstituted compound 1d (black
solid, yield 235 mg, 4.9%) and monosubstituted compound 2d (red
oil, yield 554 mg, 14%).
Experimental Section
General: Catalytic reactions were conducted under inert gas atmo-
sphere using standard Schlenk techniques and
a glove box
(MBraun UniLab). Toluene as a solvent for catalytic reactions was
distilled from benzophenone ketyl and stored under nitrogen.
Other reagents were used as purchased. Column chromatography
of organic products was carried out using silica gel [Kanto Kagaku,
silica gel 60  (spherical, neutral)]. The 1H NMR spectra were
taken with a VARIAN Mercury Y plus 400 MHz spectrometer at
room temperature. Chemical shifts (δ) were recorded in ppm from
the signal assigned as [D]chloroform, which was passed through a
column with neutral alumina before use. IR spectra were recorded
with a Perkin–Elmer Spectrum One spectrometer equipped with a
universal diamond ATR (wavenumbers given in cm–1). Elemental
analyses were carried out with a YANACO CHN Corder MT-5,
AUTO-SAMPLER. Antipyrine was used as the standard sample.
Mass spectra (EHMS) were recorded with a JEOL JMS-GCmateII.
1d: C18H15N3O2 (305.34): calcd. C 70.81, H 4.95, N 13.76; found
1
C 70.64, H 5.16, N 13.79. H NMR (CDCl3): δ = 9.60 (br., 2 H),
7.39 (t, J = 6.8 Hz, 4 H), 7.27 (d, J = 6.45 Hz, 4 H), 7.19 (t, J =
7.4 Hz, 2 H), 7.03 (t, J = 8.4 Hz, 1 H), 6.49 (d, J = 8.4 Hz, 2 H)
ppm. 13C NMR (CDCl3): δ = 144.85, 139.56, 135.29, 129.55,
125.08, 124.68, 124.20, 103.82 ppm. IR (neat): ν = 3368 (N–H),
˜
1570 (N=O) cm–1.
The disubstituted compound 1e was crystallized from hexane to
yield purple crystals (total yield 5.45 g, 74%).
Preparation of 2,6-Diaminonitrobenzenes 1a–c and Monoaminoni-
trobenzenes 1g and 1h: In a typical example, 2,6-dipyrrolidinyl-1-
nitrobenzene (1a) was prepared as follows. To a 200 mL round-
bottom flask were added 2,6-dichloronitrobenzene (3.84 g,
20.0 mmol) and pyrrolidine (51.0 mL, 600 mmol). The mixture was
refluxed (at 118 °C) for 120 h. After the reaction, pyrrolidine was
evaporated under reduced pressure to yield a red solid, which was
then dissolved in CH2Cl2 (50 mL) and washed with water. Removal
of the solvent afforded an orange solid (yield 5.05 g, 97%).
1e: C30H39N3O2 (473.66): calcd. C 76.07, H 8.30, N 8.87; found C
1
76.04, H 8.17, N 8.76. H NMR (CDCl3): δ = 9.75 (br., 2 H), 7.35
(t, J = 7.7 Hz, 2 H), 7.25 (d, J = 8.4 Hz, 4 H), 6.79 (t, J = 8.3 Hz,
1 H), 5.48 (d, J = 8.4 Hz, 2 H), 3.11 (sept, J = 6.9 Hz, 4 H), 1.20
(d, J = 6.8 Hz, 12 H), 1.15 (d, J = 6.9 Hz, 12 H) ppm. 13C NMR
(CDCl3): δ = 148.35, 147.08, 136.32, 133.42, 128.27, 124.03, 121.47,
101.04, 28.39, 24.66, 23.04 ppm. IR (neat): ν = 3372 (N–H), 1573
˜
(N=O) cm–1.
1a: C14H19N3O2 (261.32): calcd. C 64.35, H 7.33, N 16.08; found
C 64.20, H 7.26, N 15.81. 1H NMR (CDCl3): δ = 7.09 (t, J =
8.3 Hz, 1 H), 6.34 (d, J = 8.3 Hz, 2 H), 3.23–3.20 (m, 8 H), 1.92–
1.89 (m, 8 H) ppm. 13C NMR (CDCl3): δ = 143.49, 131.38, 130.20,
1f: Purple solid, yield 3.43 g, 57%. C24H27N3O2 (389.50): calcd. C
1
74.01, H 6.99, N 10.79; found C 74.00, H 6.96, N 10.82. H NMR
(CDCl3): δ = 9.62 (br., 2 H), 6.96 (s, 4 H), 6.82 (t, J = 8.3 Hz, 1
106.18, 49.54, 25.40 ppm. IR (neat): ν = 1510 (N=O) cm–1.
H), 5.50 (d, J = 8.3 Hz, 2 H), 2.32 (s, 6 H), 2.18 (s, 12 H) ppm. 13
C
˜
2248
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2243–2250

Intramolecular Cannizzaro Reaction of Glyoxal with Anilines
NMR (CDCl3): δ = 147.13, 136.85, 136.56, 136.17, 133.56, 129.27, 4a: C16H23N3O2 (289.4): calcd. C 66.41, H 8.01, N 14.52; found C
1
121.91, 100.37, 20.96, 18.15 ppm. IR (neat): ν = 3347 (N–H), 1571
66.57, H 7.90, N 14.12. H NMR (CDCl3): δ = 7.08 (t, J = 8.2 Hz,
1 H), 6.87 (br., 1 H) 6.36 (br., 2 H) 3.71 (br., 2 H), 3.24 (br., 8 H),
3.07 (br., 1 H), 1.90 (br., 8 H) ppm. 13C NMR (CDCl3): δ = 174.92,
149.25, 129.21, 110.97, 106.45, 59.60, 50.88, 25.52 ppm. IR (neat):
˜
(N=O) cm–1.
Preparation of Diaminoanilines 3a–e and Monoaminoanilines 3g and
3h: In a typical example, 2,6-dipyrrolidinylaniline 3a was obtained
as follows. To a 1 L round-bottom flask were added 2,6-dipyrrolid-
inyl-1-nitrobenzene 1a (4.54 g, 17.4 mmol), EtOH (350 mL), 5%
Pd/C (1.85 g, 0.87 mmol). The flask was evacuated and filled with
hydrogen gas (1 atm). The mixture was stirred at 5 °C for 5–14 h.
Filtration of the resulting suspension and removal of the solvent
gave a pale-yellow solid (3.67 g), which was dissolved in ethanol
(20 mL) in a 500 mL Erlenmeyer flask by heating. After the solu-
tion was cooled to –30 °C, ice-cold water (80 mL) was slowly
added. The precipitate was filtered and dried to yield 3a as a white
solid (yield 3.55 g, 88%).
ν = 3218 (N–H, O–H), 1673 (C=O) cm–1.
˜
The product 4b was obtained only by washing the residual solid
after evaporation with diethyl ether (pale-yellow solid, yield
210.2 mg, 66%): C18H27N3O2 (317.43): calcd. C 68.11, H 8.57, N
13.24; found C 67.83, H 8.58, N 13.35. 1H NMR (CDCl3): δ = 7.17
(t, J = 8.1 Hz, 1 H), 6.80 (d, J = 8.1 Hz, 2 H), 4.25 (d, J = 4.9 Hz,
2 H), 3.03 (br., 1 H), 2.81 (t, J = 5.2 Hz, 8 H), 1.69–1.63 (m, 8 H),
1.57–1.51 (m, 4 H) ppm. 13C NMR (CDCl3): δ = 172.37, 150.34,
127.77, 124.29, 115.06, 61.43, 52.94, 26.51, 24.23 ppm. IR (neat):
ν = 3320, 3264 (N–H, O–H), 1660 (C=O) cm–1.
˜
4c: Pale-yellow solid, yield 26.4 mg, 16%. C30H42N6O5 (566.3):
calcd. C 63.58, H 7.47, N 14.83; found C 63.13, H 7.21, N 14.49.
1H NMR (CDCl3): δ = 7.24 (t, J = 7.9 Hz, 2 H), 6.86 (d, J =
8.0 Hz, 4 H), 6.83 (br., 2 H), 4.38 (br., 2 H), 3.74 (br., 8 H), 3.61
(br., 8 H), 2.90 (br., 8 H), 2.79 (br., 8 H) ppm. 13C NMR (CDCl3):
δ = 172.71, 149.49, 142.77, 136.57, 128.33, 125.94, 121.06, 115.97,
3a: C14H21N3 (231.34): calcd. C 72.69, H 9.15, N 18.16; found C
72.56, H 9.03, N 10.82. 1H NMR (CDCl3): δ = 6.75–6.68 (m, 3 H),
4.02 (br., 2 H), 3.09–3.06 (m, 8 H), 1.93–1.90 (m, 8 H) ppm. 13C
NMR (CDCl3): δ = 137.94, 135.85, 117.51, 112.49, 50.76, 24.09
ppm. IR (neat): ν = 3406, 3319 (N–H) cm–1.
˜
3b: White solid, yield 3.04 g, 98%. C16H25N3 (259.39): calcd. C
115.83, 67.21, 67.14, 51.90, 51.48, 47.39 ppm. IR (neat): ν = 3258
˜
1
(N–H, O–H), 1681 (C=O), 1108, 1100 (C–O) cm–1.
74.09, H 9.71, N 16.20; found C 74.35, H 9.65, N 16.00. H NMR
(CDCl3): δ = 6.77–6.69 (m, 3 H), 4.26 (br., 2 H), 2.86 (br., 8 H),
1.73–1.68 (m, 8 H), 1.57 (br., 4 H) ppm. 13C NMR (CDCl3): δ =
140.74, 136.39, 117.38, 114.33, 52.57, 26.92, 24.43 ppm. IR (neat):
4g: Brown solid, yield 64.8 mg, 15%. This material was pure
enough for further use; purification by recrystallization from meth-
anol yielded pale-yellow crystals. C12H16N2O2 (220.3): calcd. C
ν = 3419, 3406, 3324 (N–H) cm–1.
˜
1
65.43, H 7.32, N 12.72; found C 64.72, H 7.03, N 12.02. H NMR
3c: White solid, yield 3.76 g, 94%. C14H21N3O2 (263.3): calcd. C
63.85, H 8.04, N 15.96; found C 63.69, H 8.00, N 16.03. H NMR
(CDCl3): δ = 9.13 (br., 1 H), 8.15 (dd, J = 7.4, 1.3 Hz, 1 H), 7.11–
7.00 (m, 3 H), 4.22 (br., 2 H), 3.04–3.03 (m, 4 H), 1.94–1.91 (m, 4
H) ppm. 13C NMR (CDCl3): δ = 169.67, 140.91, 131.03, 124.70,
1
(CDCl3): δ = 6.82–6.74 (m, 3 H), 4.27 (br., 2 H), 3.86 (t, J = 4.5 Hz,
8 H), 2.94 (t, J = 4.6 Hz, 8 H) ppm. 13C NMR (CDCl3): δ = 139.11,
123.29, 120.92, 118.80, 62.71, 51.90, 24.45 ppm. IR (neat): ν =
˜
˜
3391, 3348 (N–H, O–H), 1662 (C=O) cm–1.
136.19, 117.68, 114.77, 67.48, 51.27 ppm. IR (neat): ν = 3423, 3319
(N–H), 1111 (C–O) cm–1.[23]
5: Pale-yellow solid, yield 156.8 mg, 55%. C19H15N3 (285.3): calcd.
C 79.98, H 5.30, N 14.73; found C 79.74, H 5.48, N 14.73. 1H
NMR (CDCl3): δ = 8.03 (br., 1 H), 7.60–7.53 (m, 4 H), 7.49–7.45
(m, 1 H), 7.35–7.34 (m, 4 H), 7.21–7.20 (m, 2 H), 7.00 (br., 1 H)
ppm. 13C NMR (CDCl3): δ = 141.98, 140.11, 136.53, 135.88,
134.31, 133.87, 129.98, 129.29, 127.96, 124.65, 123.99, 121.68,
119.16, 104.94, 101.52 ppm.
3d: Pale-purple solid, yield 2.56 g, 89%. C18H17N3 (275.4): calcd.
C 78.52, H 6.22, N 15.26; found C 78.55, H 6.40, N 15.16. 1H
NMR (CDCl3): δ = 7.23 (t, J = 7.8 Hz, 4 H), 7.00 (d, J = 8.0 Hz,
2 H), 6.84 (t, J = 7.4 Hz, 2 H), 6.78 (d, J = 8.0 Hz, 4 H), 6.73 (t,
J = 8.0 Hz, 1 H), 5.25 (br., 2 H), 3.90 (br., 2 H) ppm. 13C NMR
(CDCl3): δ = 145.18, 137.66, 129.51, 129.33, 120.85, 119.49, 118.52,
115.38 ppm. IR (neat): ν = 3428, 3397, 3334 (N–H) cm–1.
˜
The Reaction of Glyoxal with 4-Pyrrolidinylaniline 3h: To a 20 mL
Schlenk tube were added 4-pyrrolidinylaniline (0.1622 g,
1.0 mmol), 1-propanol (0.9 mL), and a 40% glyoxal solution
(57.3 µL, 0.5 mmol). The mixture was stirred at 70 °C for 24 h. The
reaction mixture was filtered and washed with cold methanol to
yield 6 as a brown solid (yield 158.1 mg, 91%). C22H26N4 (346.5):
calcd. C 76.27, H 7.56, N 16.17; found C 76.02, H 7.55, N 16.07.
1H NMR (CDCl3): δ = 8.47 (s, 2 H), 7.36 (d, J = 8.9 Hz, 4 H),
6.58 (d, J = 8.9 Hz, 4 H), 3.36–3.32 (m, 8 H), 2.04–2.01 (m, 8 H)
ppm. 13C NMR (CDCl3): δ = 154.24, 147.91, 138.24, 123.46,
3e: Purple solid, which was separated by column chromatography,
yield 16.7 mg, 3.5%. EHMS Calcd for C30H41N3: (M+) 443.3301.
1
Found: 443.3333. H NMR (CDCl3): δ = 7.22–7.20 (m, 6 H), 6.43
(t, J = 8.0 Hz, 1 H), 5.87 (d, J = 8.0 Hz, 2 H), 4.88 (br., 2 H), 3.73
(br., 2 H), 3.14 (sept, J = 6.8 Hz, 4 H), 1.21 (br., 12 H), 1.11 (br.,
12 H) ppm. 13C NMR (CDCl3): δ = 144.88, 137.53, 137.37, 125.57,
124.78, 123.66, 120.14, 109.38, 27.98, 24.64, 23.15 ppm. IR (neat):
ν = 3372, 3302 (N–H) cm–1.
˜
The synthetic method for 3g[5] and 3h[24] was similar to that for 3a–
3d.
111.88, 47.68, 25.52 ppm. IR (neat): ν = 1598 (C=N) cm–1.
˜
X-ray Crystallography: Single crystals of α-hydroxy amide, α-amino
amide, and imidazole derivatives 4a, 4c, 4g, and 5 suitable for X-
ray diffraction studies were grown in the methanol/diethyl ether,
dichloromethane/hexane, methanol, and toluene/hexane solutions,
respectively. All data were collected on a Rigaku Saturn CCD dif-
fractometer at 123 K using mirror-focused graphite-monochro-
mated Mo-Kα radiation (λ = 0.71070 Å). The structures were
solved by direct methods and refined by full-matrix least-squares
fitting based on F2 using the PC version of the program SHELXL
97-2.[25] All H atoms were located at ideal positions and were in-
cluded in the refinement, but were restricted to ride on the atom
to which they were bonded.
The Reactions of Glyoxal with 2,6-Diaminoanilines 3a–c and 2-Ami-
noaniline 3g: In a typical example, the reaction of glyoxal with 2,6-
dipyrrolidinylaniline 3a was as follows. To a 20 mL Schlenk tube
were added 2,6-dipyrrolidinylaniline (0.2 g, 0.865 mmol), 1-propa-
nol (0.78 mL), and a 40% glyoxal solution (99.2 µL, 0.865 mmol)
under an argon atomsphere. After the tube was sealed, the mixture
was stirred at 70 °C for 24 h. The volatile compounds were removed
under reduced pressure to obtain a brown solid. The residual solid
was separated by silica gel column chromatography using ethyl ace-
tate as an eluent to yield 4a as a pale yellow solid (yield 133.5 mg,
53%).
Eur. J. Org. Chem. 2009, 2243–2250
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
2249

Y. Irie, Y. Koga, T. Matsumoto, K. Matsubara
FULL PAPER
J. Chem. 1981, 59, 1061–1067; c) Y. Vida, E. Perez-Inestrosa,
R. Suau, Tetrahedron Lett. 2005, 46, 1575–1577.
[11] K. Ishihara, T. Yano, Org. Lett. 2004, 6, 1983–1986.
[12] A. E. Russell, S. P. Miller, J. P. Morken, J. Org. Chem. 2000, 65,
8381–8383.
Supporting Information (see also the footnote on the first page of
this article): 1H and 13C NMR spectra for the new organic com-
pounds and additional X-ray measurement data.
[13] S. L. Shapiro, I. M. Rose, L. Freedman, J. Am. Chem. Soc.
1959, 81, 6322–6329.
Acknowledgments
[14] a) T. Punniyamurthy, J. Iqbal, Tetrahedron Lett. 1997, 38,
4463–4466; b) K. X. Chen, F. G. Njoroge, J. Pichardo, A.
Prongay, N. Butkiewicz, N. Yao, V. Madison, V. Girijavallab-
han, J. Med. Chem. 2006, 49, 567–574.
This work was financially supported by the Japanese Society for
the Promotion of Science (Grants-in-Aid for Scientific Research
19750084).
[15] M. Hladik, J. J. Hsiao, A. Lynnroberts, Environ. Sci. Technol.
2005, 39, 6561–6574.
[1] a) N. V. Sidgwick, I. T. Millar, H. D. Springall, The Organic
Chemistry of Nitrogen, 3rd ed, Clarendon Press, Oxford, 1966;
b) G. Sauvé, V. S. Rao, in Comprehensive Organic Functional
Group Transformations (Eds.: A. R. Katritzky, O. Meth-Cohn,
C. W. Rees), Pergamon, Oxford, 1995, vol. 2, pp. 752–817; c)
J. L. Chiara, in Comprehensive Organic Functional Group Trans-
formations II (Eds.: A. R. Katritzky, R. J. K. Taylor), Perga-
mon, Oxford, 2005, vol. 2, p. 731–852.
[2] a) D. Meinhard, M. Wegner, G. Kipiani, A. Hearley, P. Reuter,
S. Fischer, O. Marti, B. Rieger, J. Am. Chem. Soc. 2007, 129,
9182–9191; b) J. Gavenonis, T. D. Tilley, Organometallics 2004,
23, 31–43; c) I. Göttker-Schnetmann, P. Wehrmann, C. Röhr,
S. Mecking, Organometallics 2007, 26, 2348–2362.
[3] Y.-K. Lim, X. Jiang, J. C. Bollinger, D. Lee, J. Mater. Chem.
2007, 17, 1969–1980.
[4] a) J. Yoshida, K. Itami, Chem. Rev. 2002, 102, 3693–3716; b)
K. Itami, T. Nokami, Y. Ishimura, K. Mitsudo, T. Kamei, J.
Yoshida, J. Am. Chem. Soc. 2001, 123, 11577–11585; c) K. It-
ami, K. Mitsudo, J. Yoshida, Angew. Chem. Int. Ed. 2001, 40,
2337–2339.
[5] M. D. Nair, R. Adams, J. Am. Chem. Soc. 1961, 83, 3518–3521.
[6] B. K. Trivedi, A. Holmes, T. L. Stoeber, C. J. Blankley, W. H.
Roark, J. A. Picard, M. K. Shaw, A. D. Essenburg, R. L. Stan-
field, B. R. Krause, J. Med. Chem. 1993, 36, 3300–3307, and
some patents described about such anilines.
[7] J. P. Wolfe, S. L. Buchwald, J. Org. Chem. 2000, 65, 1144–1157.
[8] a) J. March, Advanced Organic Chemistry, 4th ed., John
Wiley & Sons, New York, 1992, p. 1233–1235; b) S. Canniz-
zaro, Justus Liebigs Ann. Chem. 1853, 88, 129–130; c) T. A.
Geissman, Org. React. 1944, 11, 94–113; d) C. G. Swain, A. L.
Powell, W. A. Sheppard, C. R. Morgan, J. Am. Chem. Soc.
1979, 101, 3576–3583; e) S. K. Chung, J. Chem. Soc., Chem.
Commun. 1982, 480–481; f) E. C. Ashby, D. T. Coleman III,
M. P. Gamasa, Tetrahedron Lett. 1983, 24, 851–854.
[9] Recent examples a) D. Basavaiah, D. S. Sharada, A. Veerend-
har, Tetrahedron Lett. 2006, 47, 5771–5774; b) K. Yoshizawa,
S. Toyota, F. Toda, Tetrahedron Lett. 2001, 42, 7983–7985; c)
A. Sharifi, M. M. Mojtahedi, M. R. Saidi, Tetrahedron Lett.
1999, 40, 1179–1180.
[10] a) K. Bowden, A. M. Butt, M. Streater, J. Chem. Soc. Perkin
Trans. 2 1992, 567–571; b) R. S. McDonald, C. E. Sibley, Can.
[16] a) J. W. Jaroszewski, E. S. Olafsdottir, P. Wellendorph, J. Chris-
tensen, H. Franzyk, B. Somanadhan, B. A. Budnik, L. B.
Jørgensen, V. Clausen, Phytochemistry 2002, 59, 501–511; b)
N. R. Candeias, P. M. P. Gois, C. A. M. Afonso, J. Org. Chem.
2006, 71, 5489–5497; c) N. Kambe, T. Inoue, N. Sonoda, Org.
Synth. 1995, 72, 154–162; d) N. R. Candeias, P. M. P. Gois,
C. A. M. Afonso, J. Org. Chem. 2006, 71, 5489–5497.
[17]
a) J. M. Concellón, E. Bardales, Org. Lett. 2003, 5, 4783–4785;
b) T. Kurz, K. Widyan, Tetrahedron 2005, 61, 7247–7251; c)
S. E. Denmark, Y. Fan, J. Am. Chem. Soc. 2003, 125, 7825–
7827; d) Q. Xia, B. Ganem, Org. Lett. 2002, 4, 1631–1634; e)
M. Adamczyk, J. Grote, S. Rege, Tetrahedron: Asymmetry
1997, 8, 2509–2512.
[18]
a) A. J. Arduengo III, R. Krafczyk, R. Schmutzler, Tetrahedron
1999, 55, 14523-l4534; b) J. Huang, S. P. Nolan, J. Am. Chem.
Soc. 1999, 121, 9889–9890; c) M. B. Abrams, B. L. Scott, R. T.
Baker, Organometallics 2000, 19, 4944–4956.
[19] CCDC-717675 (for 4a), -717676 (for 4c), -717677 (for 4g),
-717678 (for 5) contain the supplementary crystallographic
data (excluding structure factors) for this paper. These data can
be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[20] Finally, the deuterium content of the α-hydrogen in 4b was
30% after 16 h at 70 °C, probably due to enolization by the
basic amines.
[21] Lobry de Bruyn-Alberda van Ekenstein reaction; see review ar-
ticle: J. C. Speck, Adv. Carbohydr. Chem. 1958, 13, 63–103.
[22] a) F. Szydlo, B. Andrioletti, E. Rose, C. Duhayon, Tetrahedron
Lett. 2008, 49, 3749–3751; b) R. Duval, G. Lewin, R. Hocque-
miller, Bioorg. Med. Chem. 2003, 11, 3439–3446.
[23] R. Varala, E. Ramu, M. M. Alam, S. R. Adapa, Synlett 2004,
10, 47–1750.
[24] Syntheses of compounds 1c and 3c are described in a patent,
no detailed characterizations are given: J. R. Brian, G. R. Da-
vid, F. Gillian (Knoll Aktiengesellschaft, Germany), WO96/
17612, 1996. CAS numbers: 179900-23-3 (1c), 179900-24-4 (3c).
[25] G. M. Sheldrick, SHELXL-97, A program for crystal structure
refinement, release 97-2, University of Göttingen, 1997.
Received: January 25, 2009
Published Online: March 31, 2009
2250
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 2243–2250